Machine learning for the prediction of acute kidney injury in patients with sepsis

Risk prediction for acute kidney disease and adverse outcomes in patients with chronic obstructive pulmonary disease: an interpretable machine learning approach

BACKGROUND

Little is known about acute kidney injury (AKI) and acute kidney disease (AKD) in patients with chronic obstructive pulmonary disease (COPD) and COPD mortality based on the acute/subacute renal injury. This study develops machine learning models to predict AKI, AKD, and mortality in COPD patients, utilizing web applications for clinical decisions.

METHODS

We included 2,829 inpatients from January 2016 to December 2018. Data were split into 80% for training and 20% for testing. Eight machine learning algorithms were used, and model performance was evaluated using various metrics. SHAP was used to visualize the decision process. The best models, assessed using AUROC were used to develop web applications for identifying high-risk patients.

RESULTS

The incidence rates were 13.71% for AKI and 15.11% for AKD. The overall mortality rate was 4.84%. LightGBM performed best with AUROC of 0.815, 0.827, and 0.934 in AKI, AKD, and mortality, respectively. Key predictors for AKI were Scr, neutrophil percentage, cystatin c, BUN, and LDH. For AKD, the key predictors were age, AKI grade, HDL-C, Scr, and BUN. The key predictors for mortality included the use of dopamine and epinephrine drugs, cystatin c, renal function trajectory, albumin, and neutrophil percentage. Force plots visualized the prediction process for individual patients.

CONCLUSIONS

The incidence of AKI and AKD is significant in patients with COPD. Renal function trajectory is crucial for predicting mortality in these patients. Web applications were developed to predict AKI, AKD, and mortality, improving prognosis by identifying high-risk patients and reducing adverse events and disease progression.

Identification of pyroptosis-related gene S100A12 as a potential diagnostic biomarker for sepsis through bioinformatics analysis and machine learning

Sepsis is a non-discriminatory inflammatory reaction that can result in a diverse array of organ dysfunctions, which can be fatal. Pyroptosis is a programmed mechanism of cell death that is distinguishable from apoptosis and other forms of cellular demise. However, the role of pyroptosis in sepsis remains to be further explored. In this study, by employing a combination of the difference analysis, WGCNA, Friends' analysis, and machine learning, the central gene S100A12 was successfully identified. S100A12 demonstrated superb diagnostic capabilities in both the integrated and external validation datasets. Furthermore, significant disparities were observed in the levels of monocytes, eosinophils, and neutrophils between sepsis patients and the control group, as per the findings of immune infiltration analysis. The aforementioned immune infiltrating cells exhibited an increase in expression levels among patients diagnosed with sepsis and were found to be significantly and positively associated with S100A12 expression. The results of the single-cell analysis indicated a significant expression of S100A12 in both neutrophils and monocytes, which was in complete alignment with the outcomes of immune infiltration. In summary, the pyroptosis-related gene S100A12 represents a potential biomarker for the diagnosis and treatment of sepsis.

Predicting rapid kidney function decline in middle-aged and elderly Chinese adults using machine learning techniques

The rapid decline of kidney function in middle-aged and elderly people has become an increasingly serious public health problem. Machine learning (ML) technology has substantial potential to disease prediction. The present study use dataset from the Chinese Health and Retirement Longitudinal Study (CHARLS) and utilizes advanced Gradient Boosting algorithms to develop predictive models. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to identify the key predictors, and multivariate logistic regression was utilized to validate the independent predictive power of the variables. Furthermore, the study integrated SHapley Additive exPlanations (SHAP) to boost the interpretability of the model. The findings show that the Gradient Boosting Model demonstrated robust performance across both the training and test datasets. Specifically, it attained AUC values of 0.8 and 0.765 in the training and test sets, respectively, while achieving accuracy scores of 0.736 and 0.728 in these two datasets. LASSO regression identified key influencing factors, including estimated glomerular filtration rate (eGFR), age, hemoglobin (Hb), glucose, and systolic blood pressure (SBP). Multivariate linear regression further confirmed the independent associations between these variables and rapid kidney function deterioration (P < 0.05). This study developed a risk assessment model for rapid kidney function deterioration that is applicable to middle-aged and elderly populations in China.

Phellodendrine Exerts Protective Effects on Intra-abdominal Sepsis by Inactivating AKT/NF-kB Signaling

Acute kidney injury (AKI) and acute lung injury (ALI) are major complications of intra-abdominal sepsis, leading to increased mortality. Phellodendrine (PHE) is a characteristic and important active ingredient of Phellodendri Cortex, possessing multiple pharmacological properties. This study intends to explore the effect of PHE on intra-abdominal sepsis-induced AKI and ALI. An intra-abdominal infection-induced rat model of sepsis was established by fecal intraperitoneal injection, followed by the administration of PHE. ELISA was used to determine plasma levels of inflammatory cytokines. Hematoxylin-eosin, Periodic acid Schiff, and Masson trichrome staining were employed for histopathological analysis of rat kidney and lung tissues. Western blotting was used to estimate the AKT/NF-κB signaling-related protein levels. The results showed that PHE improved the survival rate of septic rats and reduced plasma levels of proinflammatory cytokines. PHE administration attenuated pathological lesions in the kidneys and lungs of septic rats. Mechanistically, PHE treatment blocked AKT/NF-κB signaling in septic rats' kidneys and lungs. In conclusion, PHE ameliorates intra-abdominal sepsis-induced kidney and lung injury possibly by inactivating AKT/NF-kB signaling.

Attention to early stages: predicting acute kidney injury in a post cardiosurgical ICU setting using an inclusive time-to-event model

BACKGROUND

Acute kidney injury (AKI) is a critical complication in intensive care units (ICUs) that is known to have multifaceted impacts. However, as AKI is often detected too late, early prediction is crucial for timely intervention.

METHODS

We used an attention-based time-to-event model to estimate the risk of a patient's first AKI incidence in a post-cardiosurgical ICU setting, irrespective of commonly employed limitations such as focusing on severe stages (2 & 3). Pre-, intra-, and postoperative data from 8564 adult patients were included, and AKI was defined by adhering to the full Kidney Disease: Improving Global Outcomes (KDIGO) definition. Models were primarily evaluated using the concordance index (CI).

RESULTS

70.4 % of patients developed AKI, with stage 1 being the most frequent initial stage (88.1 %). The attention-based network outperformed our baseline model, achieving CIs of 0.80, 0.72, and 0.69 for ranking event risks up to 6, 12, and 24 h prior to the onset. In terms of converting the task to a classification problem for literature comparison, we obtained areas under the receiver operator characteristic curve (auROCs) of 0.82-0.73. Performance improved for severe AKIs only, yielding CIs of 0.92, 0.85, and 0.75, and auROCs ranging between 0.94 and 0.78.

CONCLUSION

We demonstrated the importance of early-stage AKI predictions and presented a novel approach to achieve this. Under similar assumptions, our results showed improvement and approached outcomes comparable to the literature. While practical validation is pending, we are confident that our approach proves useful in assisting physicians to prevent AKI development.

Development and validation of a prediction model for the depressive symptom risk in commercial airline pilots

Background/aims

Shift workers, such as medical personnel, and pilots, are facing an increased risk of depressive symptoms. Depressive symptoms significantly impact an individual's quality of life and affect work performance, decision-making abilities, and overall public safety. This study aims to establish a multidimensional depressive symptom prediction model based on a large sample of commercial airline pilots to facilitate early identification, prevention, and personalized intervention strategies.

Methods

This population-based study included 11,111 participants, with 7918 pilots in the training set and 3193 pilots in the external validation set. Depressive symptom severity was assessed using the Patient Health Questionnaire-9 (PHQ-9). Physiological, psychological, and lifestyle factors potentially associated with depressive symptom risk were collected. The optimal predictors for model development were selected using the Boruta algorithm combined with the LASSO method, and a nomogram was developed using multivariate logistic regression to predict depressive symptoms in pilots. The model performance was evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves, and accuracy measures, such as the Brier score and Spiegelhalter z-test. Additionally, decision curve analysis (DCA) was performed to assess the model's clinical utility.

Results

A total of 7918 pilots were included in the training set and 3193 were included in the external validation set. Five characteristic indicators were selected based on their significance in the prediction of depressive symptom risk: living status, alcohol drinking, family history of mental health disorder, subjective health, and subjective sleep quality. The model showed acceptable overall discrimination (AUCtrain = 0.836, 95%CI 0.818 to 0.854; AUCvalidation = 0.840, 95%CI 0.811 to 0.868) and calibration (Brier scoretrain = 0.048; Brier scorevalidation = 0.051). The decision curve analysis showed that the net benefit was superior to intervening on all participants or not intervening on all participants.

Conclusions

This study provides a reliable tool for early prediction and customized management of depressive symptoms among commercial airline pilots. This approach promotes the development of the field by transitioning from passive mental health care to active mental health prevention, emphasizing personalized prevention strategies.

Development and validation of a quick screening tool for predicting neck pain patients benefiting from spinal manipulation: a machine learning study

BACKGROUND

Neck pain (NP) ranks among the leading causes of years lived with disability worldwide. While spinal manipulation is a common physical therapy intervention for NP, its variable patient responses and inherent risks necessitate careful patient selection. This study aims to develop and validate a machine learning-based prediction model to identify NP patients most likely to benefit from spinal manipulation.

METHODS

This multicenter study analyzed 623 NP patients in a retrospective cohort and 319 patients from a separate hospital for external validation, with data collected between May 2020 and November 2024. Treatment success was defined as achieving ≥ 50% reduction in Numerical Rating Scale (NRS) and ≥ 30% reduction in Neck Disability Index (NDI) after two weeks of spinal manipulation. We compared data imputation methods through density plots, and conducted δ-adjusted sensitivity analysis. Then employed both Boruta algorithm and LASSO regression to select relevant predictors from 40 initial features, and four feature subsets (Boruta-selected, LASSO-selected, intersection, and union) were evaluated to determine the optimal combination. Nine machine learning algorithms were tested using internal validation (70% training, 30% testing) and external validation. Performance metrics included Area Under the Receiver Operating Characteristic Curve (AUC), accuracy, F1-score, sensitivity, specificity, and predictive values. The SHAP framework enhanced model interpretability. Youden's Index was applied to determine the optimal predictive probability threshold for clinical decision support, and a web-based application was developed for clinical implementation.

RESULTS

The combined LASSO and Boruta algorithms identified nine optimal predictors, with the union feature set achieving superior performance. Among the algorithms tested, the Multilayer Perceptron (MLP) model demonstrated optimal performance with an AUC of 0.823 (95% CI 0.750, 0.874) in the test set, showing consistency between training (AUC = 0.829) and test performance. External validation confirmed robust performance (AUC: 0.824, accuracy: 0.765, F1 score: 0.76) with satisfactory calibration (Brier score = 0.170). SHAP analysis highlighted the significant predictive value of clinical measurements and patient characteristics. Based on Youden's Index, the optimal predictive probability threshold was 0.603, yielding a sensitivity of 0.762 and specificity of 0.802. The model was implemented as a web-based application providing real-time probability calculations and interactive SHAP force plots.

CONCLUSION

Our machine learning model demonstrates robust performance in identifying suitable candidates for spinal manipulation among neck pain patients, offering clinicians an evidence-based practical tool to optimize patient selection and potentially improve treatment outcomes.

Machine Learning for the Prediction of Acute Kidney Injury in Critically Ill Patients with Coronary Heart Disease: Algorithm Development and Validation

BACKGROUND

Acute kidney injury (AKI) frequently occurs in critically ill patients with coronary heart disease (CHD), and its development markedly elevates mortality rates and prolongs hospitalization duration. Early AKI prediction is crucial for timely intervention and amelioration of patient outcomes.

OBJECTIVE

This study aims to develop and verify a clinical prediction model for the occurrence of AKI upon admission in the critically ill CHD population through machine learning (ML).

METHODS

Data from the MIMIC-IV (version 2.2) database were gathered and included information on critically ill CHD individuals in the intensive care unit (ICU). The dataset was randomized into a training set (70%) and a test set (30%). LASSO regression was employed for feature variable selection. ML models, including logistic regression (LR), decision tree (DT), naive bayes (NB), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM), were constructed using the training set. The six models were compared in the test set to identify the best-performing model. Subsequently, the model was assessed by calibration curve and decision curve analysis(DCA). External validation was conducted using data from the Second Affiliated Hospital of Zhengzhou University. Ultimately, the predictive model was interpreted via SHapley Additive Explanations (SHAP) values.

RESULTS

2,711 ICU-admitted CHD patients were selected, with 1,809 (66.7%) having AKI. Thirteen variables were selected to construct the six ML models. XGBoost exhibited the best performance regarding discrimination (AUC =0.765, 95% CI 0.731-0.800), accuracy (0.725), and sensitivity (0.759). External validation using a cohort of 226 patients confirmed the strong generalizability of the XGBoost model (AUC = 0.835, 95% CI 0.782-0.887). Feature importance analyses derived from SHAP values, DT, RF, and XGBoost consistently identified five key predictors associated with the development of AKI: mechanical ventilation, use of antiplatelet agents, age, N-terminal pro B-type natriuretic peptide (NT-proBNP) levels, and acute physiology score III (APSIII).

CONCLUSIONS

ML models can serve as reliable tools for forecasting AKI in the critically ill CHD cohort. The XGBoost model is highly accurate and may aid doctors in identifying high-risk individuals for early intervention to lower mortality.

CLINICALTRIAL

Identification of Key Genes and Potential Therapeutic Targets in Sepsis-Associated Acute Kidney Injury Using Transformer and Machine Learning Approaches

Sepsis-associated acute kidney injury (SA-AKI) is a life-threatening complication of sepsis, characterized by high mortality and prolonged hospitalization. Early diagnosis and effective therapy remain difficult despite extensive investigation. To address this, we developed an AI-driven integrative framework that combines a Transformer-based deep learning model with established machine learning techniques (LASSO, SVM-RFE, Random Forest and neural networks) to uncover complex, nonlinear interactions among gene-expression biomarkers. Analysis of normalized microarray data from GEO (GSE95233 and GSE69063) identified differentially expressed genes (DEGs), and KEGG/GO enrichment via clusterProfiler revealed key pathways in immune response, protein synthesis, and antigen presentation. By integrating multiple transcriptomic cohorts, we pinpointed 617 SA-AKI-associated DEGs-21 of which overlapped between sepsis and AKI datasets. Our Transformer-based classifier ranked five genes (MYL12B, RPL10, PTBP1, PPIA, and TOMM7) as top diagnostic markers, with AUC values ranging from 0.9395 to 0.9996 (MYL12B yielding 0.9996). Drug-gene interaction mining using DGIdb (FDR < 0.05) nominated 19 candidate therapeutics for SA-AKI. Together, these findings demonstrate that melding deep learning with classical machine learning not only sharpens early SA-AKI detection but also systematically uncovers actionable drug targets, laying groundwork for precision intervention in critical care settings.

A machine-learning-derived online prediction model for depression risk in COPD patients: A retrospective cohort study from CHARLS

BACKGROUND

Depression associated with Chronic Obstructive Pulmonary Disease (COPD) is a detrimental complication that significantly impairs patients' quality of life. This study aims to develop an online predictive model to estimate the risk of depression in COPD patients.

METHODS

This study included 2921 COPD patients from the 2018 China Health and Retirement Longitudinal Study (CHARLS), analyzing 36 behavioral, health, psychological, and socio-demographic indicators. LASSO regression filtered predictive factors, and six machine learning models-Logistic Regression, Support Vector Machine, Multilayer Perceptron, LightGBM, XGBoost, and Random Forest-were applied to identify the best model for predicting depression risk in COPD patients. Temporal validation used 2013 CHARLS data. We developed a personalized, interpretable risk prediction platform using SHAP.

RESULTS

A total of 2921 patients with COPD were included in the analysis, of whom 1451 (49.7 %) presented with depressive symptoms. 11 variables were selected to develop 6 machine learning models. Among these, the XGBoost model exhibited exceptional predictive performance in terms of discrimination, calibration, and clinical applicability, with an AUROC range of 0.747-0.811. In validation sets encompassing diverse population characteristics, XGBoost achieved the highest accuracy (70.63 %), sensitivity (59.05 %), and F1 score (63.17 %).

LIMITATIONS

The target population for the model is COPD patients. And the clinical benefits of interventions based on the prediction results remain uncertain.

CONCLUSION

We developed an online prediction platform for clinical application, allowing healthcare professionals to swiftly and efficiently evaluate the risk of depression in COPD patients, facilitating timely interventions and treatments.

Relationships between vitamin C intake and COPD assessed by machine learning approaches from the NHANES (2017-2023)

Background

This research aims to explore the possible link between Vitamin C Intake (VCI) and the incidence of Chronic Obstructive Pulmonary Disease (COPD) in Americans aged over 20.

Methods

This study analyzed data from 10,757 participants with or without COPD from NHANES (2017-2023). The primary exposure variable, VCI, was grouped by quartiles. Missing data were handled via multiple imputations. A Directed Acyclic Graph (DAG) was used to pre-identify VCI -and COPD-related covariates. Variance Inflation Factor (VIF) eliminated highly collinear variables. Machine learning methods (LASSO, Random Forest, and XGBoost) screened variables. A weighted multivariate logistic regression model explored the VCI-COPD relationship. Restricted Cubic Spline (RCS) and threshold analysis examined non-linear relationships. Subgroup analysis and interaction tests ensured reliability. A nomogram showed the predictive factors' importance for COPD. Model performance was reported using the Area Under the Receiver Operating Characteristic Curve (AUC).

Results

In all models, we found that there was a negative correlation between VCI (≥50.1 mg/day) and the prevalence of COPD. The RCS and threshold analysis results show a negative correlation between COPD and VCI (≤135.6 mg/day). Subgroup analysis shows a negative association between VCI and the prevalence of COPD, specifically among females and individuals with dietary fiber intake in the second quartile (Q2). The AUC results show that our model has good diagnostic performance.

Limitations

The cross-sectional design limits causal inference and lacks external validation.

Conclusion

An elevated VCI within 50.1-135.6 is linked to a decreased risk for COPD.

Characteristics and prognosis of patients with pathogenic microorganism-positive sepsis AKI from ICU: a retrospective cohort study

Background

Sepsis-associated acute kidney injury (SA-AKI) carries a disproportionately high morbidity and mortality rate. While the synergism between dysregulated host response and renal vulnerability is increasingly recognized, the multifactorial drivers of poor prognosis remain poorly defined. The purpose of this study was to investigate the prognosis and clinical characteristics of patients with pathogenic microorganism-positive SA-AKI.

Method

Using a retrospective analysis approach, we extracted populations from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database that fulfilled the diagnostic criteria for confirmed sepsis with microbiological evidence of pathogenic organisms, and patients were divided into two cohorts according to with or without AKI. The severity of the disease in the two groups was collected for evaluation, and the clinical indicators and prognostic results of the patients were evaluated. The objective of this study was to explore the risk factors affecting the prognosis of patients with pathogenic microorganism-positive SA-AKI.

Outcome

The hospital mortality rate of AKI in patients with pathogenic microbial-positive sepsis was 18.96%. Further analysis showed that the use of vasoactive drug therapy, high lactate level, SAPS II score, SAPS III score, LODS score, and clinical indicators of prolonged hospital stay were independent risk factors for in-hospital mortality in patients with pathogenic microorganism-positive SA-AKI. Among them, SAPS III score plays an important role in predicting the prognosis of sepsis patients with AKI. Further studies found that lactate level was positively correlated with SAPS II score, SAPS III score, and LODS score.

Conclusion

The use of vasoactive drug therapy, high lactate level, SAPS II score, SAPS III score, and LODS score plays an important role in assessing the prognosis of patients with pathogenic microorganism-positive SA-AKI, and multivariate comprehensive assessment is significant in predicting the prognosis of sepsis AKI patients.

Association between Prognostic Nutritional Index and Mortality Risk in Critically Ill Patients with Chronic Obstructive Pulmonary Disease: A Retrospective Study

Background

The Prognostic Nutritional Index (PNI), an integrative measure of body's immune and nutritional status, has demonstrated its prognostic value across a range of diseases. However, its role in critically ill patients with Chronic Obstructive Pulmonary Disease (COPD) remains unclear. This study investigates the association between PNI levels and clinical outcomes in critically ill COPD patients, with a focus on identifying its role as a potential predictor of mortality.

Methods

A retrospective analysis of 1,250 critically ill COPD patients from the MIMIC-IV (v2.2) database was conducted. Patients were grouped by PNI tertiles. Primary and secondary outcomes were 28-day and 90-day mortality, respectively. Associations were evaluated using restricted cubic splines, Cox proportional hazards regression analysis, and Kaplan‒Meier survival curves. The predictive performance of PNI was assessed via receiver operating characteristic (ROC) curves analysis, and a nomogram integrating Boruta-selected features was developed to enhance clinical utility.

Results

The final cohort comprised 1,250 critically ill COPD patients, with observed mortality rates of 25.3% and 33.2% at 28 and 90 days, respectively. Higher PNI levels were associated with reduced risk of both 28-day and 90-day mortality [28-day HR: 0.95 (95% CI: 0.93-0.97), P < 0.001; 90-day HR: 0.94 (95% CI: 0.93-0.96), P < 0.001]. Restricted cubic spline analysis confirmed this trend. Furthermore, ROC analysis demonstrated the utility of PNI as a predictor for 28-day mortality (AUC: 0.61). Boruta-selected features reinforced the importance of PNI, and the constructed nomogram exhibited excellent predictive accuracy (AUC: 0.712).

Conclusion

Higher PNI is linked to reduced mortality risk in critically ill COPD patients, indicating its potential as a prognostic marker.

Development and application of an early prediction model for risk of bloodstream infection based on real-world study

BACKGROUND

Bloodstream Infection (BSI) is a severe systemic infectious disease that can lead to sepsis and Multiple Organ Dysfunction Syndrome (MODS), resulting in high mortality rates and posing a major public health burden globally. Early identification of BSI is crucial for effective intervention, reducing mortality, and improving patient outcomes. However, existing diagnostic methods are flawed by low specificity, long detection times and high demands on testing platforms. The development of artificial intelligence provides a new approach for early disease identification. This study aims to explore the optimal combination of routine laboratory data and clinical monitoring indicators, and to utilize machine learning algorithms to construct an early, rapid, and universally applicable BSI risk prediction model, to assist in the early diagnosis of BSI in clinical practice.

METHODS

Clinical data of 2582 suspected BSI patients admitted to the Chongqing University Central Hospital, from January 1, 2021 to December 31, 2023 were collected for this study. The data were divided into a modeling dataset and an external validation dataset based on chronological order, while the modeling dataset was further divided into a training set and an internal validation set. The occurrence rate of BSI, distribution of pathogens, and microbial primary reporting time were analyzed within the training set. During the feature selection stage, univariate regression and ML algorithms were applied. First, Univariate logistic regression was used to screen for predictive factors of BSI. Then, the Boruta algorithm, Lasso regression, and Recursive Feature Elimination with Cross-validation (RFE-CV) were employed to determine the optimal combination of predictors for predicting BSI. Based on the optimal combination, six machine learning algorithms were used to construct an early BSI risk prediction model. The best model was selected by models' performance, and the Shapley Additive Explanations (SHAP) method was used to explain the model. The external validation set was used to evaluate the predictive performance and generalizability of the selected model, and the research findings were ultimately applied in clinical practice.

RESULTS

The incidence of BSI among inpatients at the Chongqing University Central Hospital was 12.91%. Following further feature selection, a set of 5 variables was determined, including white blood cell count, standard bicarbonate, base excess of extracellular fluid, interleukin-6, and body temperature. BSI early risk prediction models were constructed using six machine learning algorithms, with the XGBoost model demonstrating the best performance, achieving an AUC value of 0.782 in the internal validation set and an AUC value of 0.776 in the external validation set. This model is made publicly available as an online webpage tool for clinical use.

CONCLUSIONS

This study successfully identified a set of 5 features by analyzing routine laboratory data clinical monitoring indicators among hospitalized patients. Based on this set, a machine learning-based early risk prediction model for BSI was constructed. The model is capable of early and rapid differentiation between BSI and non-BSI patients. The inclusion of minimal risk prediction factors enhances its applicability in clinical settings, particularly at the primary care level. To further improve the model's real-world applicability and more convenient for clinical use, the online application of the model could greatly improve the efficiency of BSI diagnosis and reducing patients' mortality.

A prognostic model for highly aggressive prostate cancer using interpretable machine learning techniques

Background

Extremely aggressive prostate cancer, including subtypes like small cell carcinoma and neuroendocrine carcinoma, is associated with poor prognosis and limited treatment options. This study sought to create a robust, interpretable machine learning-based model that predicts 1-, 3-, and 5-year survival in patients with extremely aggressive prostate cancer. Additionally, we sought to pinpoint key prognostic factors and their clinical implications through an innovative method.

Materials and methods

This study retrospectively analyzed data from 1,620 patients with extremely aggressive prostate cancer in the SEER database (2000-2020). Feature selection was performed using the Boruta algorithm, and survival predictions were made using nine machine learning algorithms, including XGBoost, logistic regression (LR), support vector machine (SVM), random forest (RF), k-nearest neighbor (KNN), decision tree (DT), elastic network (Enet), multilayer perceptron (MLP) and lightGBM. Model performance was evaluated using metrics such as AUC, accuracy (F1 score), confusion matrix, and decision curve analysis. Additionally, Shapley Additive Explanations (SHAP) were applied to interpret feature importance within the model, revealing the clinical factors that influence survival predictions.

Results

Among the nine models, the lightGBM model exhibited the best performance, with an AUC and F1 score of (0.8, 0.809) for 1-year survival prediction, (0.809, 0.751) for 3-year survival prediction, and (0.773, 0.611) for 5-year survival prediction. SHAP analysis revealed that M stage was the most important feature for predicting 1- and 3-year survival, while PSA level had the greatest impact on 5-year survival predictions. The model demonstrated good clinical utility and predictive accuracy through decision curve analysis and confusion matrix.

Conclusion

The lightGBM model has good predictive power for survival in patients with extremely aggressive prostate cancer. By identifying key clinical factors and providing actionable predictions, the model has the potential to enhance prognostic accuracy and improve patient outcomes.

Influence of age-adjusted shock index trajectories on 30-day mortality for critical patients with septic shock

Background

Septic shock poses a high mortality risk in critically ill patients, necessitating precise hemodynamic monitoring. While the age-adjusted shock index (ASI) reflects hemodynamic stability, the prognostic value of its dynamic trajectory remains unexplored. This study evaluates whether dynamic 24-h ASI trajectories predict 30-day mortality in septic shock patients.

Methods

This retrospective cohort study extracted data from the MIMIC-IV (derivation cohort, n = 2,559) and eICU-CRD (validation cohort, n = 2,177) databases. The latent category trajectory model (LCTM) classified ASI changes within 24 h of intensive care unit (ICU) admission. The association between ASI trajectory categories and 30-day mortality was evaluated using Kaplan-Meier (KM) method and Cox proportional-hazard models, reported as hazard ratios (HRs) and 95% confidence intervals (CIs).

Result

Three distinct ASI trajectories were explored: persistently low (Classes 1), initial high ASI sharply decreasing followed by instability (Classes 2), and steady ASI increase (Classes 3). KM curve revealed significantly higher 30-day mortality in Class 2 (32.1%) and Class 3 (38.7%) than Class 1 (12.3%) (P < 0.001). After fully adjusting for covariates, Class 2 (HR = 1.68, 95% CI: 1.25-2.25, P = 0.001) and Class 3 (HR = 1.87, 95% CI: 1.26-2.77, P = 0.002) showed elevated mortality risks in the derivation cohort. Validation cohort results were consistent (Class 2: HR = 1.92, 95% CI: 1.38-2.68, P = 0.001) and (Class 3: HR = 1.66, 95% CI: 1.09-2.54, P = 0.019). Triple-robust analyses and subgroup analyses confirmed the reliability of the results.

Conclusion

Dynamic 24-h ASI trajectories independently predict 30-day mortality in patients with septic shock, with unstable or rising patterns signaling high-risk subgroups. This underscores the clinical utility of real-time ASI monitoring for early risk stratification and tailored intervention.

Explainable machine learning model for prediction of 28-day all-cause mortality in immunocompromised patients in the intensive care unit: a retrospective cohort study based on MIMIC-IV database

OBJECTIVES

This study aimed to develop and validate an explainable machine learning (ML) model to predict 28-day all-cause mortality in immunocompromised patients admitted to the intensive care unit (ICU). Accurate and interpretable mortality prediction is crucial for clinical decision-making and optimal allocation of critical care resources for this vulnerable patient population.

METHODS

We utilized retrospective clinical data from the MIMIC-IV (version 2.2) database, encompassing ICU admissions at Beth Israel Deaconess Medical Center from 2008 to 2019. Eligible immunocompromised patients, including those with primary immunodeficiencies and chronic acquired conditions, such as hematological malignancies, solid tumors, and organ transplantation, were selected. Data were randomly split into training (80%) and testing (20%) cohorts. Ten ML models (logistic regression, XGBoost, LightGBM, AdaBoost, Random Forest, Gradient Boosting, Gaussian Naive Bayes, Complement Naive Bayes, Multilayer Perceptron, and Support Vector Machine) were developed and evaluated using area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), sensitivity, specificity, accuracy, and F1 score. Model explainability was achieved through SHapley Additive exPlanations (SHAP), and decision curve analysis (DCA) assessed clinical utility. In addition, Cox proportional hazards regression was conducted to evaluate the impact of predictive factors on time-to-event outcomes.

RESULTS

Among the evaluated models, the Support Vector Machine (SVM) demonstrated the highest AUROC of 0.863 (95% CI 0.834-0.890) and a notable AUPRC of 0.678 (95% CI 0.624-0.736). Key predictive factors consistently identified across multiple ML models included 24-h urine output, blood urea nitrogen (BUN) levels, presence of metastatic solid tumors, Charlson Comorbidity Index (CCI), and international normalized ratio (INR). SHAP analyses provided detailed insights into how these features influenced model predictions.

CONCLUSIONS

The explainable ML models based on various artificial intelligence methods demonstrated promising clinical applicability in predicting 28-day mortality risk among immunocompromised ICU patients. Factors such as urine output, BUN, metastatic solid tumors, CCI, and INR significantly contributed to prediction outcomes and may serve as important predictors in clinical practice.

Integration of Bioinformatics and Machine Learning Strategies Identifies Ferroptosis and Immune Infiltration Signatures in Peri-Implantitis

Peri-implantitis (PI) is a chronic inflammatory disease that ultimately leads to the dysfunction and loss of implants with established osseointegration. Ferroptosis has been implicated in the progression of PI, but its precise mechanisms remain unclear. This study explores the molecular mechanisms of ferroptosis in the pathology of PI through bioinformatics, offering new insights into its diagnosis and treatment. The microarray datasets for PI (GSE33774 and GSE106090) were retrieved from the GEO database. The differentially expressed genes (DEGs) and ferroptosis-related genes (FRGs) were intersected to obtain PI-Ferr-DEGs. Using three machine learning algorithms, the Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and Boruta, we successfully identified the most crucial biomarkers. Additionally, these key biomarkers were validated using a verification dataset (GSE223924). Gene set enrichment analysis (GSEA) was also utilized to analyze the associated gene enrichment pathways. Moreover, immune cell infiltration analysis compared the differential immune cell profiles between PI and control samples. Also, we targeted biomarkers for drug prediction and conducted molecular docking analysis on drugs with potential development value. A total of 13 PI-Ferr-DEGs were recognized. Machine learning and validation confirmed toll-like receptor-4 (TLR4) and FMS-like tyrosine kinase 3 (FLT3) as ferroptosis biomarkers in PI. In addition, GSEA was significantly enriched by the biomarkers in the cytokine-cytokine receptor interaction and chemokine signaling pathway. Immune infiltration analysis revealed that the levels of B cells, M1 macrophages, and natural killer cells differed significantly in PI. Ibudilast and fedratinib were predicted as potential drugs for PI that target TLR4 and FLT3, respectively. Finally, the occurrence of ferroptosis and the expression of the identified key markers in gingival fibroblasts under inflammatory conditions were validated by RT-qPCR and immunofluorescence analysis. This study identified TLR4 and FLT3 as ferroptosis and immune cell infiltration signatures in PI, unraveling potential novel targets to treat PI.

Predictive analytics for early detection of hospital-acquired complications: An artificial intelligence approach

BACKGROUND

Hospital-acquired complications (HACs) have an adverse impact on patient recovery by impeding their path to full recovery and increasing healthcare costs.

OBJECTIVE

The aim of this study was to create a HAC risk prediction machine learning (ML) framework using hospital administrative data collections within North Metropolitan Health Service (NMHS), Western Australia.

METHOD

A retrospective cohort study was performed among 64,315 patients between July 2020 to June 2022 to develop an automated ML framework by inputting HAC and the healthcare site to obtain site-specific predictive algorithms for patients admitted to the hospital in NMHS. Univariate analysis was used for initial feature screening for 270 variables. Of these, 77 variables had significant relationship with any HAC. After excluding non-contemporaneous data, 37 variables were included in developing the ML framework based on logistic regression (LR), decision tree (DT) and random forest (RF) models to predict occurrence of four specific HACs: delirium, aspiration pneumonia, pneumonia and urinary tract infection.

RESULTS

All models exhibited similar performance with area under the curve scores around 0.90 for both training and testing datasets. For sensitivity, DT and RF exceeded LR performance while on average, false positives were lowest for LR-based models. Patient's length of stay, Charlson Index, operation length and intensive care unit stay were common predictors.

CONCLUSION

Integrating ML-based risk detection systems into clinical workflows can potentially enhance patient safety and optimise resource allocation. LR-based models exhibited best performance.

IMPLICATIONS

We have successfully developed a "real-time" risk prediction model, where patient risk scores are calculated and reviewed daily.

Machine learning prediction models for multidrug-resistant organism infections in ICU ventilator-associated pneumonia patients: Analysis using the MIMIC-IV database

OBJECTIVE

This study aims to construct and compare four machine learning models using the MIMIC-IV database to identify high-risk factors for multidrug-resistant organism (MDRO) infection in Ventilator-associated pneumonia (VAP) patients.

METHODS

The study included 972 VAP patients from the MIMIC-IV database. Data encompassing demographic information, vital signs, laboratory results, and other relevant variables were collected. The class imbalance issue was addressed using the Synthetic Minority Over-sampling Technique (SMOTE). The dataset was randomly split into training and testing sets (8:2). LASSO regression and feature importance scores were used for feature selection. Clinical prediction models were built using logistic regression, XGBoost, random forest and gradient boosting machine. The performance of the models was evaluated through receiver operating characteristic(ROC) curve analysis.Model calibration was assessed using calibration curves and Brier scores. The effectiveness was evaluated through Decision Curve Analysis (DCA). SHAP was utilized for model interpretation.

RESULTS

Among 972 patients, 824 were non-MDROs-VAP and 128 were MDROs-VAP. Comparative analysis revealed statistically significant differences in various clinical parameters. XGBoost exhibited the best predictive performance, incorporating 20 features with an AUC of 0.831 (95 % CI: 0.785-0.877) on the test set. Calibration curves demonstrated robust consistency, corroborated by Decision Curve Analysis (DCA) affirming the clinical utility. SHAP analysis identified the most important features: red cell distribution width, duration of mechanical ventilation, anion gap, basophil percentage, and neutrophil percentage.

CONCLUSION

This study established and compared four machine learning models for MDROs infections in VAP patients. XGBoost was identified as the optimal predictor, and SHAP values provided insights into 20 independent risk factors, confirming its excellent predictive value.

IMPLICATIONS FOR CLINICAL PRACTICE

VAP is a common infection in ICU patients with a heightened risk of MDRO and increased mortality. The recognition of high bias in existing models calls for future research to employ rigorous methodologies and robust data sources, aiming to develop and validate more accurate and clinically applicable predictive models for MDROs infections in VAP patients.

Identification and experimental validation of biomarkers associated with mitochondrial and programmed cell death in major depressive disorder

Background

Major depressive disorder (MDD) is associated with mitochondrial dysfunction and programmed cell death (PCD), though the underlying mechanisms remain unclear. This study aimed to investigate the molecular pathways involved in MDD using a transcriptomic analysis approach.

Methods

Transcriptomic data related to MDD were obtained from public databases. Differentially expressed genes (DEGs), PCD-related genes (PCDs), and mitochondrial-related genes (MitoGs) were analyzed to identify key gene sets: PCD-DEGs and MitoG-DEGs. Correlation analysis (|correlation coefficient| > 0.9, p < 0.05) was performed to select candidate genes. Protein-protein interaction (PPI) network analysis and intersection of four algorithms were used to identify key candidate genes. Machine learning and gene expression validation were employed, followed by reverse transcription-quantitative polymerase chain reaction (RT-qPCR) for further validation. A nomogram was developed to predict MDD probability based on biomarkers. Additional analyses included immune infiltration, regulatory networks, and drug predictions.

Results

CD63, IL17RA, and IL1R1 were identified as potential biomarkers, with significantly higher expression levels in the MDD cohort. These findings were validated by RT-qPCR. A nomogram based on these biomarkers demonstrated predictive capacity for MDD. Differential immune cell infiltration was observed, with significant differences in nine immune cell types, including activated T cells and eosinophils, between the MDD and control groups. ATF1 was identified as a common transcription factor for CD63, IL17RA, and IL1R1. Shared miRNAs for CD63 and IL1R1 included hsa-miR-490-3p and hsa-miR-125a-3p. Drug prediction analysis identified 50 potential drugs, including verteporfin, etynodiol, and histamine, targeting these biomarkers.

Conclusion

CD63, IL17RA, and IL1R1 are key biomarkers for MDD, providing insights for diagnostic development and targeted therapies. The predictive nomogram and drug predictions offer valuable tools for MDD management.

Association between the (neutrophil + monocyte)/albumin ratio and all-cause mortality in sepsis patients: a retrospective cohort study and predictive model establishment according to machine learning

INTRODUCTION

Sepsis is a life-threatening condition characterized by widespread inflammatory response syndrome in the body resulting from infection. Previous studies have demonstrated that some inflammatory factors or nutritional elements contributed to deaths in patients diagnosed with sepsis. Nevertheless, the correlation between the (neutrophil + monocyte)/albumin (NMa) ratio and all-cause mortality of patients diagnosed with sepsis remains unclear. This study aims to investigate the association between the NMa ratio and all-cause mortality in sepsis patients and to develop a predictive model using machine learning techniques.

METHODS

The clinical data were harvested from 13,851 patients with sepsis from the MIMIC-IV (3.1) database. We divided the subjects into four groups based on quartiles of the NMa ratio. The main endpoint was 30-day all-cause mortality, and the secondary endpoint was 90-day all-cause mortality. The relationship between the NMa ratio and adverse prognosis was investigated employing Cox proportional hazard regression, restricted cubic splines, and Kaplan‒Meier curves. Moreover, we employed Boruta algorithm to evaluate the predictive potential of the NMa ratio and established the prediction models utilizing machine learning algorithms.

RESULTS

After adjusting for confounders, each unit increase in the NMa ratio was associated with a 1.8% and 1.6% higher risk of 30-day and 90-day all-cause mortality, respectively (P < 0.001), indicating a linear relationship, and when treated as a categorical variable, the Quartile 4 group demonstrated a significantly higher mortality risk. Boruta feature selection also displayed that the NMa ratio possessed a higher Z score, and the models established utilizing the Cox and Random Forest algorithm identified excellent predictive performance (area under the curve (AUC) = 0.72, AUC = 0.74, respectively).

CONCLUSION

The NMa ratio is strongly and linearly associated with 30-day and 90-day all-cause mortality, with higher levels significantly increasing mortality risk, even after adjusting for potential confounders. Predictive models using Cox regression and Random Forest algorithms showed strong performance, indicating that the NMa ratio could function as a predictor of negative prognosis in patients with sepsis.

Applications of machine learning approaches for pediatric asthma exacerbation management: a systematic review

BACKGROUND

Pediatric asthma is a common chronic respiratory disease worldwide, and its acute exacerbation events significantly impact children's health and quality of life. Machine learning, an advanced data analysis technique, has shown great potential in healthcare applications in recent years. This systematic review aims to assess the application of ML techniques in pediatric asthma exacerbation and explore their effectiveness and potential value.

METHODS

Studies from four electronic databases, including PubMed, EBSCO, Elsevier, and Web of Science, from Jan 2000 to Jan 2025, were searched. Studies applying the ML methods for pediatric asthma exacerbation and published in English were eligible. The risk of bias and applicability of the included studies was assessed using the Effective Public Health Practice Project (EPHPP) quality assessment tool.

RESULTS

A total of 23 studies were selected for inclusion in this review, covering different ML models such as decision trees, neural networks, and support vector machines. These studies focused on analyzing risk factors for asthma exacerbation, diagnosing and predicting, optimizing and allocating healthcare resources, and comprehensive asthma management. The results show that ML techniques have significant advantages in the application of pediatric asthma exacerbation and in the provision of personalized health care.

CONCLUSIONS

ML techniques show great promise for application in pediatric asthma exacerbations. With further research and clinical validation, these techniques are expected to provide strong support for diagnosis, personalized treatment, and long-term management of pediatric asthma exacerbation.

CLINICAL TRIAL NUMBER

Not applicable, Prospero registration number CRD42024559232.

Early prediction of sepsis associated encephalopathy in elderly ICU patients using machine learning models: a retrospective study based on the MIMIC-IV database

Background

Sepsis associated encephalopathy (SAE) is prevalent among elderly patients in the ICU and significantly affects patient prognosis. Due to the symptom similarity with other neurological disorders and the absence of specific biomarkers, early clinical diagnosis remains challenging. This study aimed to develop a predictive model for SAE in elderly ICU patients.

Methods

The data of elderly sepsis patients were extracted from the MIMIC IV database (version 3.1) and divided into training and test sets in a 7:3 ratio. Feature variables were selected using the LASSO-Boruta combined algorithm, and five machine learning (ML) models, including Extreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost),Light Gradient Boosting Machine(LGBM), Multilayer Perceptron (MLP), and Support Vector Machines (SVM), were subsequently developed using these variables. A comprehensive set of performance metrics was used to assess the predictive accuracy, calibration, and clinical applicability of these models. For the machine learning model with the best performance, we employed the SHapley Additive Explanations(SHAP) method to visualize the model.

Results

Based on strict inclusion and exclusion criteria, a total of 3,156 elderly sepsis patients were enrolled in the study, with an SAE incidence rate of 48.7%. The mortality rate of elderly sepsis patients who developed SAE was significantly higher than that of patients in the non-SAE group (28.78% vs. 12.59%, P < 0.001). A total of 18 feature variables were selected for the construction of the ML model using the LASSO-Boruta combined algorithm. Compared to the other four models and traditional scoring systems, the XGBoost model demonstrated the best overall predictive performance, with Area Under the Curve(AUC)=0.898, accuracy=0.830, recall=0.819, F1-Score=0.820, specificity=0.840, and Precision=0.821. Furthermore, the results from the Decision Curve Analysis (DCA) and calibration curves demonstrated that the XGBoost model has significant clinical value and stable predictive performance. The ten-fold cross-validation method further confirmed the robustness and generalizability of the model. In addition, we simplified the model based on the SHAP feature importance ranking, and the results indicated that the simplified XGBoost model retains excellent predictive ability (AUC=0.858).

Conclusions

The XGBoost model effectively predicts SAE in elderly ICU patients and may serve as a reliable tool for clinicians to identify high-risk patients.

Risk Factors for Lymph Node Metastasis in Stage pT1 Invasive Lung Adenocarcinoma

OBJECTIVE

To analyze the risk factors for lymph node metastasis (LNM) in patients with stage pT1 lung adenocarcinoma to select a more appropriate surgical option.

METHODS

In this retrospective study, 294 patients with postoperative pathologically confirmed stage pT1 invasive lung adenocarcinoma were collected and divided into two groups according to whether they had mediastinal or hilar LNM. Patient tumor imaging, pathological features and gene mutations were analyzed, and risk factors that might predict LNM were derived via univariate and multivariate logistic analyses. LNM-related variables were screened by Boruta and least absolute shrinkage and selection operator regression analysis.

RESULTS

Among the 294 patients, 45 (15.3%) had positive mediastinal or hilar lymph nodes. There were no significant differences between the two groups in terms of sex, age, or underlying disease. The difference in the percentage of solidity between the two groups was significant, with the higer percentage group showing a more significant difference. The results of multivariate logistic analysis revealed that a high percentage of solid components and wild-type epidermal growth factor receptor (EGFR) were risk factors for LNM. The nomogram for predicting LNM included the consolidation tumor ratio, tumor size, micropapillary and EGFR, with an area under the curve of 93.4% (95% CI: 88.7-99.1) in the derivation cohort and 92.3% (95% CI: 84.6-99.9) in the validation cohort.

CONCLUSIONS

A high proportion of solid components and wild-type EGFR were risk factors for pT1 stage lung adenocarcinoma, suggesting that the choice of lung segmentectomy needs to be evaluated and selected more cautiously.

A Predictive Model for Acute Kidney Injury Based on Leukocyte-Related Indicators in Hepatocellular Carcinoma Patients Admitted to the Intensive Care Unit

Background: This study aimed to develop and validate a straightforward clinical risk model utilizing white blood cell (WBC) counts to predict acute kidney injury (AKI) in critically sick patients with hepatocellular carcinoma (HCC). Methods: Data were taken from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database for the training cohort. Data for an internal validation cohort were obtained from the eICU Collaborative Research Database (eICU-CRD), while patients from our hospital were utilized for external validation. A risk model was created utilizing significant indicators identified through multivariate logistic regression, following logistic regression analysis to determine the primary predictors of WBC-related biomarkers for AKI prediction. The Kaplan-Meier curve was employed to evaluate the prognostic efficacy of the new risk model. Results: A total of 1628 critically sick HCC patients were enrolled. Among these, 23 (23.2%) patients at our hospital, 84 (17.9%) patients in the eICU-CRD database, and 379 (35.8%) patients in the MIMIC-IV database developed AKI. A unique risk model was developed based on leukocyte-related indicators following the multivariate logistic regression analysis, incorporating white blood cell to neutrophil ratio (WNR), white blood cell to monocyte ratio (WMR), white blood cell to hemoglobin ratio (WHR), and platelet to lymphocyte ratio (PLR). This risk model exhibited robust predictive capability for AKI, in-hospital mortality, and ICU mortality across the training set, internal validation set, and external validation set. Conclusion: This risk model seems to have practical consequences as an innovative and accessible tool for forecasting the prognosis of critically ill HCC patients, which may, to some degree, aid in identifying equitable risk assessments and treatment strategies.

A machine learning-based analysis for the definition of an optimal renal biopsy for kidney cancer

OBJECTIVE

Renal Tumor biopsy (RTB) can assist clinicians in determining the most suitable approach for treatment of renal cancer. However, RTB's limitations in accurately determining histology and grading have hindered its broader adoption and data on the concordance rate between RTB results and final pathology after surgery are unavailable. Therefore, we aimed to develop a machine learning algorithm to optimize RTB technique and to investigate the degree of concordance between RTB and surgical pathology reports.

MATERIALS AND METHODS

Within a prospectively maintained database, patients with indeterminate renal masses who underwent RTB at a single tertiary center were identified. We recorded and analyzed the approach (US vs. CT), the number of biopsy cores (NoC), and total core tissue length (LoC) to evaluate their impact on diagnostic outcomes. The K-Nearest Neighbors (KNN), a non-parametric supervised machine learning model, predicted the probability of obtaining pathological characterization and grading. In surgical patients, final pathology reports were compared with RTB results.

RESULTS

Overall, 197 patients underwent RTB. Overall, 89.8% (n=177) and 44.7% (n=88) of biopsies were informative in terms of histology and grading, respectively. The discrepancy rate between the pathology results from renal tissue biopsy (RTB) and the final pathology report following surgery was 3.6% (n=7) for histology and 5.0% (n=10) for grading. According to the machine learning model, a minimum of 2 cores providing at least 0.8 cm of total tissue should be obtained to achieve the best accuracy in characterizing the cancer. Alternatively, in cases of RTB with more than two cores, no specific minimum tissue threshold is required.

CONCLUSIONS

The discordance rates between RTB pathology and final surgical pathology are notably minimal. We defined an optimal renal biopsy strategy based on at least 2 cores and at least 0.8 cm of tissue or at least 3 cores and no minimum tissue threshold.

PATIENTS SUMMARY

RTB is a useful test for kidney cancer, but it's not always perfect. Our study shows that it usually matches up well with what doctors find during surgery. Using machine learning can make RTB even better by helping doctors know how many samples to take. This helps doctors treat kidney cancer more accurately.

Integration of single-cell and bulk RNA sequencing data using machine learning identifies oxidative stress-related genes LUM and PCOLCE2 as potential biomarkers for heart failure

Oxidative stress (OS) is a pivotal mechanism driving the progression of cardiovascular diseases, particularly heart failure (HF). However, the comprehensive characterisation of OS-related genes in HF remains largely unexplored. In the present study, we analysed single-cell RNA sequencing datasets from the Gene Expression Omnibus and OS gene sets from GeneCards. We identified 167 OS-related genes potentially linked to HF by applying algorithms, such as AUCell, UCell, singscore, ssgsea, and AddModuleScore, combined with correlation analysis. Subsequently, we used feature selection algorithms, including least absolute shrinkage and selection operator, XGBoost, Boruta, random forest, gradient boosting machines, decision trees, and support vector machine recursive feature elimination, to identify lumican (LUM) and procollagen C-endopeptidase enhancer 2 (PCOLCE2) as key biomarker candidates with significant diagnostic potential. Bulk RNA-sequencing confirmed their elevated expression in patients with HF, highlighting their predictive utility. Single-cell analysis further revealed their upregulation primarily in fibroblasts, emphasising their cell-specific role in HF. To validate these findings, we developed a transverse aortic constriction-induced HF mouse model that showed enhanced cardiac OS activity and significant PCOLCE2 upregulation in the HF group. These results provide strong evidence of the involvement of OS-related mechanisms in HF. Herein, we propose a diagnostic strategy that provides novel insights into the molecular mechanisms underlying HF. However, further studies are required to validate its clinical utility and ensure its application in the diagnosis of HF.

Prediction Model for Risk of Death in Elderly Critically Ill Patients with Kidney Failure

Background and Objectives: Kidney failure (KF) is associated with high mortality, especially among critically ill patients in the intensive care unit (ICU). Conversely, age is an independent risk factor for the development of KF. Therefore, understanding the mortality risk profile of elderly critically ill patients with KF can help clinicians in implementing appropriate measures to improve patients' prognosis. The aim of this study was to construct high-performance mortality risk prediction models for elderly ICU patients with KF using machine learning methods. Materials and Methods: Elderly (≥65 years) ICU patients diagnosed with KF were selected and relevant information (including demographic details, vital signs, laboratory tests, etc.) was collected. They were randomly divided into training, validation, and test sets in a 6:2:2 ratio. Logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost) methods were employed to develop prediction models for the risk of death in these elderly KF patients. The model's performance was evaluated by the receiver operating characteristic curve, precision rate, recall rate, and decision curve analysis. Finally, breakdown plots were utilized to analyze the mortality risk of elderly KF patients. Results: A total of 8010 elderly ICU patients with KF were included in this study, among whom 1385 patients died. Mortality prediction models were constructed using various methods, with the areas under the curve (AUC) for the different models being 0.835 (LR model), 0.839 (RF model), 0.784 (SVM model), and 0.851 (XGBoost model), respectively. The integrated Brier score (IBS) for these models were 0.206 (LR model), 0.158 (RF model), 0.217 (SVM model), and 0.102 (XGBoost model), indicating that the XGBoost model and RF model exhibited superior differentiation and calibration capacity. Further analysis revealed that the XGBoost model outperformed the others in terms of both prediction accuracy and stability. Finally, based on the ranking of important features, the primary influencing factors for elderly KF patients were identified as urine output, metastatic solid tumor, body weight, body temperature, and severity score. Conclusions: Several high-performing predictive models for mortality risk in elderly ICU patients with KF have been developed using various machine learning algorithms, with the XGBoost model demonstrating the best performance.

Machine learning for risk prediction of acute kidney injury in patients with diabetes mellitus combined with heart failure during hospitalization

This study aimed to develop a machine learning (ML) model for predicting the risk of acute kidney injury (AKI) in diabetic patients with heart failure (HF) during hospitalization. Using data from 1,457 patients in the MIMIC-IV database, the study identified twenty independent risk factors for AKI through LASSO regression and logistic regression. Six ML algorithms were evaluated, including LightGBM, random forest, and neural networks. The LightGBM model demonstrated superior performance with the highest prediction accuracy, with AUC values of 0.973 and 0.804 in the training and validation sets, respectively. The Shapley additive explanations algorithm was used to visualize the model and identify the most relevant features for AKI risk. Clinical impact curves further confirmed the strong discriminatory ability and generalizability of the LightGBM model. This study highlights the potential of ML models, particularly LightGBM, to effectively predict AKI risk in diabetic patients with HF, enabling early identification of high-risk patients and timely interventions to improve prognosis.

Validation of a cancer population derived AKI machine learning algorithm in a general critical care scenario

PURPOSE

Acute Kidney Injury (AKI) is the sudden onset of kidney damage. This damage usually comes without warning and can lead to increased mortality and inpatient costs and is of particular significance to patients undergoing cancer treatment. In previous work, we developed a machine learning algorithm to predict AKI up to 30 days prior to the event, trained on cancer patient data. Here, we validate this model on non-cancer data.

METHODS/PATIENTS

Medical Information Mart for Intensive Care (MIMIC) is a large, freely available database containing de-identified data from patients who were admitted to the critical care units of the Beth Israel Deaconess Medical Center. Data from 28,498 MIMIC patients were used to validate our algorithm, non-availability of Total Protein measure being the largest removal criterion.

RESULTS AND CONCLUSIONS

Applying our algorithm to MIMIC data generated an AUROC of 0.821 (95% CI 0.820-0.821) per blood test. Our cancer derived algorithm compares positively with other AKI models derived and/or tested on MIMIC, with our model predicting AKI at the longest time frame of up to 30 days. This suggests that our model can achieve a good performance on patient cohorts very different to those from which it was derived, demonstrating the transferability and applicability for implementation in a clinical setting.

Advancing risk factor identification for pediatric lobar pneumonia: the promise of machine learning technologies

Background

Community-acquired pneumonia (CAP) is a prevalent pediatric condition, and lobar pneumonia (LP) is considered a severe subtype. Early identification of LP is crucial for appropriate management. This study aimed to develop and compare machine learning models to predict LP in children with CAP.

Methods

A total of 25 clinical and laboratory variables were collected. Missing data (<2%) were imputed, and the dataset was split into training (60%) and validation (40%) sets. Univariable logistic regression and Boruta feature selection were used to identify significant predictors. Four machine learning algorithms-Logistic Regression (LR), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Decision Tree (DT)-were compared using area under the curve (AUC), balanced accuracy, sensitivity, specificity, and F1 score. SHAP analysis was performed to interpret the best-performing model.

Results

A total of 278 patients with CAP were included in this study, of whom 65 were diagnosed with LP. The XGBoost model demonstrated the best performance with an AUC of 0.880 (95% CI: 0.807-0.934) in the training set and 0.746 (95% CI: 0.664-0.843) in the validation set. SHAP analysis identified age, CRP, CD64 index, lymphocyte percentage, and ALB as the top five predictive factors.

Conclusion

The XGBoost model showed superior performance in predicting LP in children with CAP. The model enabled early diagnosis and risk assessment of LP, thereby facilitating appropriate clinical decision-making.

Machine Learning for In-hospital Mortality Prediction in Critically Ill Patients With Acute Heart Failure: A Retrospective Analysis Based on the MIMIC-IV Database

BACKGROUND

The incidence, mortality, and readmission rates for acute heart failure (AHF) are high, and the in-hospital mortality for AHF patients in the intensive care unit (ICU) is higher. However, there is currently no method to accurately predict the mortality of AHF patients.

METHODS

The Medical Information Mart for Intensive Care Ⅳ (MIMIC-Ⅳ) database was used to perform a retrospective study. Patients meeting the inclusion criteria were identified from the MIMIC-Ⅳ database and randomly divided into a training set (n = 3,580, 70%) and a validation set (n = 1,534, 30%). The variates collected include demographic data, vital signs, comorbidities, laboratory test results, and treatment information within 24 hours of ICU admission. By using the least absolute shrinkage and selection operator (LASSO) regression model in the training set, variates that affect the in-hospital mortality of AHF patients were screened. Subsequently, in the training set, five common machine learning (ML) algorithms were applied to construct models using variates selected by LASSO to predict the in-hospital mortality of AHF patients. The predictive ability of the models was evaluated for sensitivity, specificity, accuracy, the area under the curve of receiver operating characteristics, and clinical net benefit in the validation set. To obtain a model with the best predictive ability, the predictive ability of common scoring systems was compared with the best ML model.

RESULTS

Among the 5,114 patients, in-hospital mortality was 12.5%. Comparing the area under the curve, the XGBoost model had the best predictive ability among all ML models, and the XGBoost model was chosen as the final model for its higher net benefit. Its predictive ability was superior to common scoring systems.

CONCLUSIONS

The XGBoost model can effectively predict the in-hospital mortality of AHF patients admitted to the ICU, which may assist clinicians in precise management and early intervention for patients with AHF to reduce mortality.

Research hotspots and frontiers of machine learning in renal medicine: a bibliometric and visual analysis from 2013 to 2024

BACKGROUND

The kidney, an essential organ of the human body, can suffer pathological damage that can potentially have serious adverse consequences on the human body and even affect life. Furthermore, the majority of kidney-induced illnesses are frequently not readily identifiable in their early stages. Once they have progressed to a more advanced stage, they impact the individual's quality of life and burden the family and broader society. In recent years, to solve this challenge well, the application of machine learning techniques in renal medicine has received much attention from researchers, and many results have been achieved in disease diagnosis and prediction. Nevertheless, studies that have conducted a comprehensive bibliometric analysis of the field have yet to be identified.

OBJECTIVES

This study employs bibliometric and visualization analyses to assess the progress of the application of machine learning in the renal field and to explore research trends and hotspots in the field.

METHODS

A search was conducted using the Web of Science Core Collection database, which yielded articles and review articles published from the database's inception to May 12, 2024. The data extracted from these articles and review articles were then analyzed. A bibliometric and visualization analysis was conducted using the VOSviewer, CiteSpace, and Bibliometric (R-Tool of R-Studio) software.

RESULTS

2,358 papers were retrieved and analyzed for this topic. From 2013 to 2024, the number of publications and the frequency of citations in the relevant research areas have exhibited a consistent and notable increase annually. The data set comprises 3734 institutions in 91 countries and territories, with 799 journals publishing the results. The total number of authors contributing to the data set is 14,396. China and the United States have the highest number of published papers, with 721 and 525 papers, respectively. Harvard University and the University of California System exert the most significant influence at the institutional level. Regarding authors, Cheungpasitporn, Wisit, and Thongprayoon Charat of the Mayo Clinic organization were the most prolific researchers, with 23 publications each. It is noteworthy that researcher Breiman I had the highest co-citation frequency. The journal with the most published papers was "Scientific Reports," while "PLoS One" had the highest co-citation frequency. In this field of machine learning applied to renal medicine, the article "A Clinically Applicable Approach to Continuous Prediction of Future Acute Kidney Injury" by Tomasev N et al., published in NATURE in 2019, emerged as the most influential article with the highest co-citation frequency. A keyword and reference co-occurrence analysis reveals that current research trends and frontiers in nephrology are the management of patients with renal disease, prediction and diagnosis of renal disease, imaging of renal disease, and development of personalized treatment plans for patients with renal disease. "Acute kidney injury," "chronic kidney disease," and "kidney tumors" are the most discussed diseases in medical research.

CONCLUSIONS

The field of renal medicine is witnessing a surge in the application of machine learning. On one hand, this study offers a novel perspective on applying machine learning techniques to kidney-related diseases based on bibliometric analysis. This analysis provides a comprehensive overview of the current status and emerging research areas in the field, as well as future trends and frontiers. Conversely, this study furnishes data on collaboration and exchange between countries, regions, institutions, journals, authors, keywords, and reference co-citations. This information can facilitate the advancement of future research endeavors, which aim to enhance interdisciplinary collaboration, optimize data sharing and quality, and further advance the application of machine learning in the renal field.

Relationship between atherogenic index of plasma and length of stay in critically ill patients with atherosclerotic cardiovascular disease: a retrospective cohort study and predictive modeling based on machine learning

BACKGROUND

The atherogenic index of plasma (AIP) is considered an important marker of atherosclerosis and cardiovascular risk. However, its potential role in predicting length of stay (LOS), especially in patients with atherosclerotic cardiovascular disease (ASCVD), remains to be explored. We investigated the effect of AIP on hospital LOS in critically ill ASCVD patients and explored the risk factors affecting LOS in conjunction with machine learning.

METHODS

Using data from the Medical Information Mart for Intensive Care (MIMIC)-IV. AIP was calculated as the logarithmic ratio of TG to HDL-C, and patients were stratified into four groups based on AIP values. We investigated the association between AIP and two key clinical outcomes: ICU LOS and total hospital LOS. Multivariate logistic regression models were used to evaluate these associations, while restricted cubic spline (RCS) regressions assessed potential nonlinear relationships. Additionally, machine learning (ML) techniques, including logistic regression (LR), decision tree (DT), random forest (RF), extreme gradient boosting (XGB), and light gradient boosting machine (LGB), were applied, with the Shapley additive explanation (SHAP) method used to determine feature importance.

RESULTS

The study enrolled a total of 2423 patients with critically ill ASCVD, predominantly male (54.91%), and revealed that higher AIP values were independently associated with longer ICU and hospital stays. Specifically, for each unit increase in AIP, the odds of prolonged ICU and hospital stays were significantly higher, with adjusted odds ratios (OR) of 1.42 (95% CI, 1.11-1.81; P = 0.006) and 1.73 (95% CI, 1.34-2.24; P < 0.001), respectively. The RCS regression demonstrated a linear relationship between increasing AIP and both ICU LOS and hospital LOS. ML models, specifically LGB (ROC:0.740) and LR (ROC:0.832) demonstrated superior predictive accuracy for these endpoints, identifying AIP as a vital component of hospitalization duration.

CONCLUSION

AIP is a significant predictor of ICU and hospital LOS in patients with critically ill ASCVD. AIP could serve as an early prognostic tool for guiding clinical decision-making and managing patient outcomes.

A novel real-time model for predicting acute kidney injury in critically ill patients within 12 hours

BACKGROUND

A major challenge in the prevention and early treatment of acute kidney injury (AKI) is the lack of high-performance predictors in critically ill patients. Therefore, we innovatively constructed U-AKIpredTM for predicting AKI in critically ill patients within 12 h of panel measurement.

METHODS

The prospective cohort study included 680 patients in the training set and 249 patients in the validation set. After performing inclusion and exclusion criteria, 417 patients were enrolled in the training set and 164 patients were enrolled in the validation set. AKI was diagnosed by Kidney Disease: Improving Global Outcomes (KDIGO) criteria.

RESULTS

Twelve urinary kidney injury biomarkers (mALB, IgG, TRF, α1MG, NAG, NGAL, KIM-1, L-FABP, TIMP2, IGFBP7, CAF22, and IL-18) exhibited good predictive performance for AKI within 12 h in critically ill patients. U-AKIpredTM, combined with three crucial biomarkers (α1MG, L-FABP, and IGFBP7) by multivariate logistic regression analysis, exhibited better predictive performance for AKI in critically ill patients within 12 h than the other 12 kidney injury biomarkers. The area under the curve (AUC) of the U-AKIpredTM, as a predictor of AKI within 12 h, was 0.802 (95% CI: 0.771-0.833, P < .001) in the training set and 0.844 (95% CI: 0.792-0.896, P < .001) in the validation cohort. A nomogram based on the results of the training and validation sets of U-AKIpredTM was developed that showed optimal predictive performance for AKI. The fitting effect and prediction accuracy of U-AKIpredTM was evaluated by multiple statistical indicators. To provide a more flexible predictive tool, the dynamic nomogram (https://www.xsmartanalysis.com/model/U-AKIpredTM) was constructed using a web calculator. Decision curve analysis and a clinical impact curve were used to reveal that U-AKIpredTM with the three crucial biomarkers had a higher net benefit than these 12 kidney injury biomarkers, respectively. The net reclassification index and integrated discrimination index were used to improve the significant risk reclassification of AKI compared with the 12 kidney injury biomarkers. The predictive efficiency of U-AKIpredTM was better than the NephroCheck® when testing for AKI and severe AKI.

CONCLUSION

U-AKIpredTM is an excellent predictive model of AKI in critically ill patients within 12 h and would assist clinicians in identifying those at high risk of AKI.

Predicting the risk of acute kidney injury in patients with acute pancreatitis complicated by sepsis using a stacked ensemble machine learning model: a retrospective study based on the MIMIC database

OBJECTIVE

This study developed and validated a stacked ensemble machine learning model to predict the risk of acute kidney injury in patients with acute pancreatitis complicated by sepsis.

DESIGN

A retrospective study based on patient data from public databases.

PARTICIPANTS

This study analysed 1295 patients with acute pancreatitis complicated by septicaemia from the US Intensive Care Database.

METHODS

From the MIMIC database, data of patients with acute pancreatitis and sepsis were obtained to construct machine learning models, which were internally and externally validated. The Boruta algorithm was used to select variables. Then, eight machine learning algorithms were used to construct prediction models for acute kidney injury (AKI) occurrence in intensive care unit (ICU) patients. A new stacked ensemble model was developed using the Stacking ensemble method. Model evaluation was performed using area under the receiver operating characteristic curve (AUC), precision-recall (PR) curve, accuracy, recall and F1 score. The Shapley additive explanation (SHAP) method was used to explain the models.

MAIN OUTCOME MEASURES

AKI in patients with acute pancreatitis complicated by sepsis.

RESULTS

The final study included 1295 patients with acute pancreatitis complicated by sepsis, among whom 893 cases (68.9%) developed acute kidney injury. We established eight base models, including Logit, SVM, CatBoost, RF, XGBoost, LightGBM, AdaBoost and MLP, as well as a stacked ensemble model called Multimodel. Among all models, Multimodel had an AUC value of 0.853 (95% CI: 0.792 to 0.896) in the internal validation dataset and 0.802 (95% CI: 0.732 to 0.861) in the external validation dataset. This model demonstrated the best predictive performance in terms of discrimination and clinical application.

CONCLUSION

The stack ensemble model developed by us achieved AUC values of 0.853 and 0.802 in internal and external validation cohorts respectively and also demonstrated excellent performance in other metrics. It serves as a reliable tool for predicting AKI in patients with acute pancreatitis complicated by sepsis.

Occurrence rate and risk factors for acute kidney injury after lung transplantation: a systematic review and meta-analysis

Background

Compared with other solid organ transplantation, the morbidity rate of acute kidney injury is higher in lung transplantation. Our research was designed to examine the occurrence rate and risk factors for acute kidney injury after lung transplantation through a systematic review and meta-analysis.

Methodology

We conducted a database search for case-control studies and cohort studies on the occurrence rate and risk factors for acute kidney injury after lung transplantation up to August 19, 2023. Stata 15.0 was used for data analysis.

Results

Nineteen case-control or cohort studies were included, involving 1,755 cases of acute kidney injury after lung transplantation and 1,404 cases of non-acute kidney injury after lung transplantation. Based on the meta-analysis, the risk factors for acute kidney injury after lung transplantation included pulmonary fibrosis (OR, 1.34; CI [1.09-1.65]), hypertension (OR, 1.30; CI [1.07-1.58]), pre-op mechanical ventilation (OR, 3.30; CI [1.84-5.90]), pre-op extracorporeal membrane oxygenation (OR, 3.70; CI [2.51-5.45]), double lung transplantation (OR, 1.91; CI [1.45-2.53]), cardiopulmonary bypass support (OR, 1.82; CI [1.38-2.40]), cardiovascular events (OR, 1.50; CI [1.15-1.96]), intra-op hypotension (OR, 2.70; CI [1.42-5.14]), post-op extracorporeal membrane oxygenation (OR, 1.90; CI [1.20-3.01]), sepsis (OR, 3.20; CI [2.16-4.73]), dialysis (OR, 12.79; CI [6.11-26.8]).

Conclusions

Based on the existing evidence, clinical professionals can implement early detection, diagnosis and treatment of patients with acute kidney injury after lung transplantation, to improve the quality of life of these patients.

Association between the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio and mortality among hypertension patients

The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) reflects the balance between pro- and anti-atherogenic lipoproteins. This study aims to explore the relationship between NHHR and mortality among hypertension patients. Data from 17,075 hypertensive adults in the National Health and Nutrition Examination Survey (NHANES) were analyzed. Multivariate Cox regression and restricted cubic splines were used to assess the correlation between NHHR and mortality. A segmented Cox model evaluated threshold effects, and sensitivity analyses confirmed result robustness. Machine learning algorithms were used to establish a prediction model. Over a median follow-up of 84 months, 3625 deaths occurred. A U-shaped association was observed between NHHR and both all-cause and cardiovascular mortality, with threshold values at 2.32 and 2.65. Below these thresholds, NHHR was negatively associated with mortality, while values above the thresholds were positively associated. NHHR was classified as an important variable in the prediction model, with the random survival forest (rsf) algorithm showing superior performance. This study identified a U-shaped association between NHHR and mortality in hypertension patients, with threshold points at NHHR values of 2.32 and 2.65, indicating that NHHR is a potential predictor of mortality in patients with hypertension.

Development and validation of the PHM-CPA model to predict in-hospital mortality for cirrhotic patients with acute kidney injury

BACKGROUND

The presence of acute kidney injury (AKI) significantly increases in-hospital mortality risk for cirrhotic patients. Early prognosis prediction for these patients is crucial. We aimed to develop and validate a machine learning model for in-hospital mortality prediction for cirrhotic patients with AKI.

METHODS

Data from cirrhotic patients with AKI hospitalized at the First Affiliated Hospital of Zhejiang University between January 1, 2013, and December 31, 2020 were used to train and validate an extreme Gradient Boosting model to predict in-hospital mortality risk. The Boruta algorithm was used for variable selection. The optimal model was selected and named as PHM-CPA (Prediction of in-Hospital Mortality for Cirrhotic Patients with AKI). The PHM-CPA model was then externally validated in patients from eICU Collaborative Research Database (eICU-CRD) and Medical Information Mart for Intensive Care III dataset (MIMIC). The predictive performance of PHM-CPA model was compared with that of logistic regression (LR) model and 25 previously reported models.

RESULTS

A total of 519 cirrhotic patients with AKI were enrolled in model training cohort, of whom 118 (23%) died during hospitalization. Fifteen variables from common laboratory tests were selected to develop the PHM-CPA model. The PHM-CPA model achieved an AUROC of 0.816 (95% CI, 0.763-0.861) in the internal validation cohort and 0.787 (95% CI, 0.745-0.830) in the external validation cohort. The PHM-CPA model consistently outperformed the LR model and 25 previously reported models.

CONCLUSION

We developed and validated the PHM-CPA model, comprising readily available clinical variables, which demonstrated superior performance and calibration in predicting in-hospital mortality for cirrhotic patients with AKI.

COMPREHENSIVE CHARACTERIZATION OF CYTOKINES IN PATIENTS UNDER EXTRACORPOREAL MEMBRANE OXYGENATION: EVIDENCE FROM INTEGRATED BULK AND SINGLE-CELL RNA SEQUENCING DATA USING MULTIPLE MACHINE LEARNING APPROACHES

ABSTRACT

Background : Extracorporeal membrane oxygenation (ECMO) is an effective technique for providing short-term mechanical support to the heart, lungs, or both. During ECMO treatment, the inflammatory response, particularly involving cytokines, plays a crucial role in pathophysiology. However, the potential effects of cytokines on patients receiving ECMO are not comprehensively understood. Methods : We acquired three ECMO support datasets, namely two bulk and one single-cell RNA sequencing (RNA-seq), from the Gene Expression Omnibus (GEO) combined with hospital cohorts to investigate the expression pattern and potential biological processes of cytokine-related genes (CRGs) in patients under ECMO. Subsequently, machine learning approaches, including support vector machine (SVM), random forest (RF), modified Lasso penalized regression, extreme gradient boosting (XGBoost), and artificial neural network (ANN), were applied to identify hub CRGs, thus developing a prediction model called CRG classifier. The predictive and prognostic performance of the model was comprehensively evaluated in GEO and hospital cohorts. Finally, we mechanistically analyzed the relationship between hub cytokines, immune cells, and pivotal molecular pathways. Results : Analyzing bulk and single-cell RNA-seq data revealed that most CRGs were significantly differentially expressed; the enrichment scores of cytokine and cytokine-cytokine receptor (CCR) interaction were significantly higher during ECMO. Based on multiple machine learning algorithms, nine key CRGs (CCL2, CCL4, IFNG, IL1R2, IL20RB, IL31RA, IL4, IL7, and IL7R) were used to develop the CRG classifier. The CRG classifier exhibited excellent prognostic values (AUC > 0.85), serving as an independent risk factor. It performed better in predicting mortality and yielded a larger net benefit than other clinical features in GEO and hospital cohorts. Additionally, IL1R2, CCL4, and IL7R were predominantly expressed in monocytes, NK cells, and T cells, respectively. Their expression was significantly positively correlated with the relative abundance of corresponding immune cells. Gene set variation analysis (GSVA) revealed that para-inflammation, complement and coagulation cascades, and IL6/JAK/STAT3 signaling were significantly enriched in the subgroup that died after receiving ECMO. Spearman correlation analyses and Mantel tests revealed that the expression of hub cytokines (IL1R2, CCL4, and IL7R) and pivotal molecular pathways scores (complement and coagulation cascades, IL6/JAK/STAT3 signaling, and para-inflammation) were closely related. Conclusion : A predictive model (CRG classifier) comprising nine CRGs based on multiple machine learning algorithms was constructed, potentially assisting clinicians in guiding individualized ECMO treatment. Additionally, elucidating the underlying mechanistic pathways of cytokines during ECMO will provide new insights into its treatment.

The development and validation of a prediction model for post-AKI outcomes of pediatric inpatients

Background

Acute kidney injury (AKI) is common in hospitalized children. A post-AKI outcomes prediction model is important for the early detection of important clinical outcomes associated with AKI so that early management of pediatric AKI patients can be initiated.

Methods

Three retrospective cohorts were set up based on two pediatric hospitals in China, in which 8205 children suffered AKI during hospitalization. Two clinical outcomes were evaluated, i.e. hospital mortality and dialysis within 28 days after AKI occurrence. A Genetic Algorithm was used for feature selection, and a Random Forest model was built to predict clinical outcomes. Subsequently, a temporal validation set and an external validation set were used to evaluate the performance of the prediction model. Finally, the stratification ability of the prediction model for the risk of mortality was compared with a commonly used mortality risk score, the pediatric critical illness score (PCIS).

Results

The prediction model performed well for the prediction of hospital mortality with an area under the receiver operating curve (AUROC) of 0.854 [95% confidence interval (CI) 0.816-0.888], and the AUROC was >0.850 for both temporal and external validation. For the prediction of dialysis, the AUROC was 0.889 (95% CI 0.871-0.906). In addition, the AUROC of the prediction model for hospital mortality was superior to that of PCIS (P < .0001 in both temporal and external validation).

Conclusions

The new proposed post-AKI outcomes prediction model shows potential applicability in clinical settings.

Development and validation of a novel risk-predicted model for early sepsis-associated acute kidney injury in critically ill patients: a retrospective cohort study

OBJECTIVES

This study aimed to develop a prediction model for the detection of early sepsis-associated acute kidney injury (SA-AKI), which is defined as AKI diagnosed within 48 hours of a sepsis diagnosis.

DESIGN

A retrospective study design was employed. It is not linked to a clinical trial. Data for patients with sepsis included in the development cohort were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The least absolute shrinkage and selection operator regression method was used to screen the risk factors, and the final screened risk factors were constructed into four machine learning models to determine an optimal model. External validation was performed using another single-centre intensive care unit (ICU) database.

SETTING

Data for the development cohort were obtained from the MIMIC-IV 2.0 database, which is a large publicly available database that contains information on patients admitted to the ICUs of Beth Israel Deaconess Medical Center in Boston, Massachusetts, USA, from 2008 to 2019. The external validation cohort was generated from a single-centre ICU database from China.

PARTICIPANTS

A total of 7179 critically ill patients with sepsis were included in the development cohort and 269 patients with sepsis were included in the external validation cohort.

RESULTS

A total of 12 risk factors (age, weight, atrial fibrillation, chronic coronary syndrome, central venous pressure, urine output, temperature, lactate, pH, difference in alveolar-arterial oxygen pressure, prothrombin time and mechanical ventilation) were included in the final prediction model. The gradient boosting machine model showed the best performance, and the areas under the receiver operating characteristic curve of the model in the development cohort, internal validation cohort and external validation cohort were 0.794, 0.725 and 0.707, respectively. Additionally, to aid interpretation and clinical application, SHapley Additive exPlanations techniques and a web version calculation were applied.

CONCLUSIONS

This web-based clinical prediction model represents a reliable tool for predicting early SA-AKI in critically ill patients with sepsis. The model was externally validated using another ICU cohort and exhibited good predictive ability. Additional validation is needed to support the utility and implementation of this model.

[Predicting Intensive Care Unit Mortality in Patients With Heart Failure Combined With Acute Kidney Injury Using an Interpretable Machine Learning Model: A Retrospective Cohort Study]

Objective

Heart failure (HF) complicated by acute kidney injury (AKI) significantly impacts patient outcomes, and it is crucial to make early predictions of short-term mortality. This study is focused on developing an interpretable machine learning model to enhance early prediction accuracy in such clinical scenarios.

Methods

This retrospective cohort study utilized data from the Medical Information Mart for Intensive Care Ⅳ (MIMIC-Ⅳ, version 2.0) database. Data from the first 24 hours after admission to the ICU were extracted and divided into a training set (70%) and a validation set (30%). We utilized the SHapley Additive exPlanation (SHAP) method to interpret the workings of an extreme gradient boosting (XGBoost) model and identify key prognostic factors. The XGBoost model's predictive ability was evaluated against three other machine learning models using the area under the curve (AUC) metric, and its interpretation was enhanced using the SHAP method.

Results

The study included 8028 patients with HF complicated by AKI. The XGBoost model outperformed the other models, achieving an AUC of 0.93 (95% confidence interval [CI]: 0.78-0.94; accuracy = 0.89), while neural network model showed the worst performance (AUC = 0.79, 95% CI: 0.77-0.82; accuracy = 0.82). Decision curve analysis showed the superior net benefit of the XGBoost model within the 9% to 60% threshold probabilities. SHAP analysis was performed to identify the top 20 predictors, with age (mean SHAP value 1.29) and Glasgow Coma Scale score (mean SHAP value 1.24) emerging as significant factors.

Conclusions

Our interpretable model offers an enhanced ability to predict mortality risk in HF patients with AKI in ICUs. This model can be used to assist in formulating effective treatment plans and optimizing resource allocation.

Artificial intelligence for the detection of acute myeloid leukemia from microscopic blood images; a systematic review and meta-analysis

Background

Leukemia is the 11th most prevalent type of cancer worldwide, with acute myeloid leukemia (AML) being the most frequent malignant blood malignancy in adults. Microscopic blood tests are the most common methods for identifying leukemia subtypes. An automated optical image-processing system using artificial intelligence (AI) has recently been applied to facilitate clinical decision-making.

Aim

To evaluate the performance of all AI-based approaches for the detection and diagnosis of acute myeloid leukemia (AML).

Methods

Medical databases including PubMed, Web of Science, and Scopus were searched until December 2023. We used the "metafor" and "metagen" libraries in R to analyze the different models used in the studies. Accuracy and sensitivity were the primary outcome measures.

Results

Ten studies were included in our review and meta-analysis, conducted between 2016 and 2023. Most deep-learning models have been utilized, including convolutional neural networks (CNNs). The common- and random-effects models had accuracies of 1.0000 [0.9999; 1.0001] and 0.9557 [0.9312, and 0.9802], respectively. The common and random effects models had high sensitivity values of 1.0000 and 0.8581, respectively, indicating that the machine learning models in this study can accurately detect true-positive leukemia cases. Studies have shown substantial variations in accuracy and sensitivity, as shown by the Q values and I2 statistics.

Conclusion

Our systematic review and meta-analysis found an overall high accuracy and sensitivity of AI models in correctly identifying true-positive AML cases. Future research should focus on unifying reporting methods and performance assessment metrics of AI-based diagnostics.

Systematic review registration

https://www.crd.york.ac.uk/prospero/#recordDetails, CRD42024501980.

Albumin corrected anion gap and clinical outcomes in elderly patients with acute kidney injury caused or accompanied by sepsis: a MIMIC-IV retrospective study

BACKGROUND

Elderly acute kidney injury (AKI) occurring in the intensive care unit (ICU), particularly when caused or accompanied by sepsis, is linked to extended hospital stays, increased mortality rates, heightened prevalence of chronic diseases, and diminished quality of life. This study primarily utilizes a comprehensive critical care database to examine the correlation of albumin corrected anion gap (ACAG) levels with short-term prognosis in elderly patients with AKI caused or accompanied by sepsis, thus assisting physicians in early identification of high-risk patients.

METHODS

This study utilized data from the Medical Information Mart for Intensive Care IV (MIMIC-IV, v2.0) database. The patient population was divided into death and survival groups based on a 14-day prognosis. Subsequently, the entire population was further categorized into a normal ACAG group (12-20 mmol/L) and a high ACAG group (> 20 mmol/L) based on ACAG levels. The LASSO regression cross-validation method was employed to identify significant risk factors for inclusion in multivariate Cox regression analyses. A restricted cubic spline (RCS) was then employed to visually represent the correlation between ACAG levels and the risk of mortality in patients. Kaplan-Meier curves were utilized to plot the cumulative survival rates at 14 and 30 days for both patient groups. The robustness of the findings was subsequently evaluated through subgroup analyses.

RESULTS

Our study identified a total of 3741 eligible subjects, revealing higher all-cause mortality rates at both 14-day and 30-day intervals in the high ACAG group compared to the normal ACAG group (χ2 = 87.023, P < 0.001; χ2 = 90.508, P < 0.001). Cox regression analysis further demonstrated that an elevated ACAG on ICU admission independently posed a risk factor for both 14- and 30-day prognosis within this population. In addition, the analysis conducted using RCS revealed a non-linear association between the levels of ACAG and the risk of mortality at both 14 and 30 days in the patient cohort (χ2 = 18.220, P < 0.001; χ2 = 18.360, P < 0.001). The application of Kaplan-Meier analysis demonstrated a statistically significant decrease in cumulative survival rates among individuals with high ACAG levels (P < 0.001). Subgroup analyses indicated that ACAG levels interacted with cerebrovascular disease and acute pancreatitis on 14-day mortality (P < 0.05 for interaction).

CONCLUSION

Elevated ACAG levels at ICU admission are an independent risk factor for poor short-term prognosis, correlating with increased all-cause mortality at 14 and 30 days in elderly patients with AKI caused or accompanied by sepsis. This highlights the importance of monitoring ACAG in critically ill patients to identify those at higher risk of adverse outcomes early.

Utilizing deep learning-based causal inference to explore vancomycin's impact on continuous kidney replacement therapy necessity in blood culture-positive intensive care unit patients

Patients with positive blood cultures in the intensive care unit (ICU) are at high risk for septic acute kidney injury requiring continuous kidney replacement therapy (CKRT), especially when treated with vancomycin. This study developed a machine learning model to predict CKRT and examined vancomycin's impact using deep learning-based causal inference. We analyzed ICU patients with positive blood cultures, utilizing the Medical Information Mart for Intensive Care III data set. The primary outcome was defined as the initiation of CKRT during the ICU stay. The machine learning models were developed to predict the outcome. The deep learning-based causal inference model was utilized to quantitatively demonstrate the impact of vancomycin on the probability of CKRT initiation. Logistic regression was performed to analyze the relationship between the variables and the susceptibility of vancomycin. A total of 1,318 patients were included in the analysis, with 41 requiring CKRT. The Random Forest and Light Gradient Boosting Machine exhibited the best performance, with Area Under Curve of Receiver Operating Characteristic Curve values of 0.905 and 0.886, respectively. The deep learning-based causal inference model demonstrated an average 7.7% increase in the probability of CKRT occurrence when administrating vancomycin in total data set. Additionally, that younger age, lower diastolic blood pressure, higher heart rate, higher baseline creatinine, and lower bicarbonate levels sensitized the probability of CKRT application in response to vancomycin treatment. Deep learning-based causal inference models showed that vancomycin administration increases CKRT risk, identifying specific patient characteristics associated with higher susceptibility.IMPORTANCEThis study assesses the impact of vancomycin on the risk of continuous kidney replacement therapy (CKRT) in intensive care unit (ICU) patients with blood culture-positive infections. Utilizing deep learning-based causal inference and machine learning models, the research quantifies how vancomycin administration increases CKRT risk by an average of 7.7%. Key variables influencing susceptibility include baseline creatinine, diastolic blood pressure, heart rate, and bicarbonate levels. These findings offer insights into managing vancomycin-induced kidney risk and may inform patient-specific treatment strategies in ICU settings.

Harnessing artificial intelligence in sepsis care: advances in early detection, personalized treatment, and real-time monitoring

Sepsis remains a leading cause of morbidity and mortality worldwide due to its rapid progression and heterogeneous nature. This review explores the potential of Artificial Intelligence (AI) to transform sepsis management, from early detection to personalized treatment and real-time monitoring. AI, particularly through machine learning (ML) techniques such as random forest models and deep learning algorithms, has shown promise in analyzing electronic health record (EHR) data to identify patterns that enable early sepsis detection. For instance, random forest models have demonstrated high accuracy in predicting sepsis onset in intensive care unit (ICU) patients, while deep learning approaches have been applied to recognize complications such as sepsis-associated acute respiratory distress syndrome (ARDS). Personalized treatment plans developed through AI algorithms predict patient-specific responses to therapies, optimizing therapeutic efficacy and minimizing adverse effects. AI-driven continuous monitoring systems, including wearable devices, provide real-time predictions of sepsis-related complications, enabling timely interventions. Beyond these advancements, AI enhances diagnostic accuracy, predicts long-term outcomes, and supports dynamic risk assessment in clinical settings. However, ethical challenges, including data privacy concerns and algorithmic biases, must be addressed to ensure fair and effective implementation. The significance of this review lies in addressing the current limitations in sepsis management and highlighting how AI can overcome these hurdles. By leveraging AI, healthcare providers can significantly enhance diagnostic accuracy, optimize treatment protocols, and improve overall patient outcomes. Future research should focus on refining AI algorithms with diverse datasets, integrating emerging technologies, and fostering interdisciplinary collaboration to address these challenges and realize AI's transformative potential in sepsis care.

Predicting the risk of gastroparesis in critically ill patients after CME using an interpretable machine learning algorithm - a 10-year multicenter retrospective study

Background

Gastroparesis following complete mesocolic excision (CME) can precipitate a cascade of severe complications, which may significantly hinder postoperative recovery and diminish the patient's quality of life. In the present study, four advanced machine learning algorithms-Extreme Gradient Boosting (XGBoost), Random Forest (RF), Support Vector Machine (SVM), and k-nearest neighbor (KNN)-were employed to develop predictive models. The clinical data of critically ill patients transferred to the intensive care unit (ICU) post-CME were meticulously analyzed to identify key risk factors associated with the development of gastroparesis.

Methods

We gathered 34 feature variables from a cohort of 1,097 colon cancer patients, including 87 individuals who developed gastroparesis post-surgery, across multiple hospitals, and applied a range of machine learning algorithms to construct the predictive model. To assess the model's generalization performance, we employed 10-fold cross-validation, while the receiver operating characteristic (ROC) curve was utilized to evaluate its discriminative capacity. Additionally, calibration curves, decision curve analysis (DCA), and external validation were integrated to provide a comprehensive evaluation of the model's clinical applicability and utility.

Results

Among the four predictive models, the XGBoost algorithm demonstrated superior performance. As indicated by the ROC curve, XGBoost achieved an area under the curve (AUC) of 0.939 in the training set and 0.876 in the validation set, reflecting exceptional predictive accuracy. Notably, in the k-fold cross-validation, the XGBoost model exhibited robust consistency across all folds, underscoring its stability. The calibration curve further revealed a favorable concordance between the predicted probabilities and the actual outcomes of the XGBoost model. Additionally, the DCA highlighted that patients receiving intervention under the XGBoost model experienced significantly greater clinical benefit.

Conclusion

The onset of postoperative gastroparesis in colon cancer patients remains an elusive challenge to entirely prevent. However, the prediction model developed in this study offers valuable assistance to clinicians in identifying key high-risk factors for gastroparesis, thereby enhancing the quality of life and survival outcomes for these patients.

Identification of macrophage polarisation and mitochondria-related biomarkers in diabetic retinopathy

BACKGROUND

The activation of macrophages or microglia in patients' whole body or local eyes play significant roles in diabetic retinopathy (DR). Mitochondrial function regulates the inflammatory polarization of macrophages. Therefore, the common mechanism of mitochondrial related genes (MRGs) and macrophage polarisation related genes (MPRGs) in DR is explored in our study to illustrate the pathophysiology of DR.

METHODS

In this study, using common transcriptome data, differentially expressed genes (DEGs) were firstly analysed for GSE221521, while module genes related to MPRGs were obtained by weighted gene co-expression network analysis (WGCNA), intersections of DEGs with MRGs were taken, intersections of DEGs with module genes of the MPRGs were taken. After that, correlation analyses were performed to obtain candidate genes. Key genes were obtained by Mendelian randomisation (MR) analysis, then biomarkers were obtained by machine learning combined with receiver operating characteristic (ROC) and expression validation between DR and control cohorts in GSE221521 and GSE160306 to obtain biomarkers. Finally, biomarkers were subjected to immune infiltration analysis, gene set enrichment analysis (GSEA), and gene-gene interaction (GGI) analysis.

RESULTS

A number of 784 of DEGs were taken to intersect with 1136 MRGs and 782 MPRGs, respectively, after which 89 genes with correlation were taken as candidate genes. MR analysis yielded 13 key genes with clear causal links to DR. The expression trends of PTAR1 and SLC25A34 were consistent and notable between DR cohort and control cohort in GSE221521 and GSE160306. So PTAR1 and SLC25A34 were used as biomarkers. Immune infiltration analysis showed that activated NK cell and Monocyte were notably different between DR cohort and control cohorts, and PTAR1 showed the strongest positive correlations with activated NK cell. Both biomarkers were enriched in lysosome and insulin signaling pathway. The GGI network showed that biomarkers associated with prenyltransferase activity and prenylation function.

CONCLUSION

This study identified two biomarkers (PTAR1 and SLC25A34) which explore the pathogenesis of DR and provide reference targets for drug development.