When to consider logistic LASSO regression in multivariate analysis?

A nomogram for predicting early biliary complications in adult liver recipients of deceased donor grafts: Integrating artery resistive index and clinical risk factors

BACKGROUND

This study aimed to identify predictors of biliary complications within 90 days after liver transplantation in adult recipients of deceased donor grafts.

METHODS

The study retrospectively analyzed adult patients who underwent liver transplantation from January 2016 to December 2021 using deceased donor grafts in our center. Patients were randomly divided into training and validation cohorts (7:3 ratio). A nomogram was developed using least absolute shrinkage and selection operator logistic regression for feature selection, followed by a 2-way stepwise approach in multivariate logistic regression. Model performance was assessed with the C-index, receiver operating characteristic area under the curve, calibration curves, and decision curve analysis.

RESULTS

A total of 757 patients were included, of whom 76 developed early biliary complications. Least absolute shrinkage and selection operator binary logistic analysis showed that postoperative day 1 arterial resistance index, acute rejection, acute-on-chronic liver failure, hepatic artery thrombosis, recipient body mass index, and donor age were independent predictors of biliary complications within 90 days. A nomogram was established on the basis of these factors. The C-index for the final nomogram was 0.822. The area under the curve in the training cohort was 0.837 (95% confidence interval, 0.780-0.893) and 0.771 (95% confidence interval, 0.677-0.865) in the validation cohort. Calibration curves demonstrated good agreement between predicted and actual outcomes. Decision curve analysis confirmed the clinical utility of the nomogram.

CONCLUSION

Low arterial resistance index (≤0.57) on the first postoperative day is a predictor of biliary complications within 90 days after liver transplantation in adult recipients of deceased donor grafts. The nomogram provides a practical tool for predicting complications and guiding clinical decisions.

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.

Development and validation of a radiogenomics prognostic model integrating PET/CT radiomics and glucose metabolism-related gene signatures for non-small cell lung cancer

BACKGROUND

Non-small cell lung cancer (NSCLC) is a highly heterogeneous malignancy characterized by altered glucose metabolism. Integration of PET/CT radiomics with glucose metabolism-related genomic signatures could provide a more comprehensive approach for prognosis and treatment guidance.

METHODS

Radiomics features were extracted from PET/CT images of 156 NSCLC patients from The Cancer Imaging Archive (TCIA) database, and glucose metabolism-related gene signatures were obtained from TCGA and GEO databases. We developed a multimodal radiogenomics prognostic model (RGC-score) using least absolute shrinkage and selection operator (LASSO) regression, combining PET/CT radiomics, glucose metabolism-related genes (GMR-genes). Functional enrichment analysis, immune infiltration assessment, and drug sensitivity analysis were performed to investigate the biological significance of glucose metabolism-related genes (GMR-genes).

RESULTS

The RGC-score model effectively stratified NSCLC patients into distinct high- and low-risk groups with significant differences in survival outcomes (P < 0.001), demonstrating excellent predictive performance (1-year AUC = 0.907, 5-year AUC = 0.968).GMR-genes are mainly involved in the process of metabolic remodeling of tumors, which is closely related to the immune microenvironment (especially CD8+ T cell infiltration) and immune checkpoint molecule expression. Additionally, significant differences in drug sensitivity were identified between glucose metabolism subtypes.

CONCLUSION

The RGC-score robustly predicts NSCLC prognosis and informs metabolic-immune interactions for personalized therapy. Limitations include the retrospective design and modest validation cohort size, necessitating prospective multicenter trials.

Impact of intermittent fasting on physical activity: a national survey of Chinese residents aged 18-80 years

Objectives

This study aims to investigate the prevalence of intermittent fasting (IF) among Chinese residents aged 18-80 and assess its impact on physical activity (PA) levels.

Methods

Data were sourced from the Psychology and Behavior Investigation of Chinese Residents, a nationally representative cross-sectional survey conducted between June 20 and 31 August 2022. A multistage stratified cluster sampling method was used. Propensity score matching (PSM) was applied to compare PA levels between individuals practicing IF and those not practicing it. Multiple logistic regression and subgroup analysis were performed to explore associations between PA levels and relevant factors.

Results

IF was practiced by 9.78% of participants, with the highest prevalence (70.78%) among those aged 18-34. While there were no significant differences in baseline characteristics between the IF and non-IF groups, sleep duration differed. IF was significantly associated with reduced PA levels (OR = 0.769, 95%CI: 0.657-0.900), and subgroup analysis highlighted the effect of sleep patterns on PA.

Conclusion

IF is common among younger Chinese residents and correlates with lower PA levels, indicating a potential need for individualized health guidance to balance dietary strategies with PA.

A coagulation-related long non-coding RNA signature to predict prognosis and immune features of breast cancer

Breast cancer (BC) remains one of the most common malignancies among women worldwide, with persistently poor prognosis despite advancements in diagnostics and therapies. Long non-coding RNAs (lncRNAs) and coagulation-related genes (CRGs) are increasingly recognized for their roles in prognosis and immune modulation. Using transcriptomic data from 1,045 BC patients in TCGA, we identified CRG-associated lncRNAs via coexpression analysis (Pearson |R|> 0.4, p < 0.001) and constructed a prognostic model through univariate Cox analysis, LASSO regression with tenfold cross-validation (λ = 0.05), and multivariate Cox analysis. The model stratified patients into high- and low-risk groups with distinct overall survival (HR = 3.21, p < 0.001) and demonstrated robust predictive accuracy (AUC = 0.795 at 1 year). Functional enrichment revealed immune-related pathways (e.g., cytokine signaling, PD-L1 regulation), and high-risk patients exhibited elevated tumor mutational burden (TMB) and PD-L1 expression, suggesting enhanced immunotherapy responsiveness. Drug sensitivity analysis identified 5 targeted agents (e.g., BIBW2992) with differential efficacy between risk groups. This CRG-lncRNA signature provides a novel tool for prognosis prediction and personalized immunotherapy in BC, illuminating crosstalk between coagulation and immune pathways.

Multimodal Visualization and Explainable Machine Learning-Driven Markers Enable Early Identification and Prognosis Prediction for Symptomatic Aortic Stenosis and Heart Failure With Preserved Ejection Fraction After Transcatheter Aortic Valve Replacement: Multicenter Cohort Study

BACKGROUND

Currently, there is a paucity of literature addressing personalized risk stratification using multimodal data in patients with symptomatic aortic stenosis and heart failure with preserved ejection fraction (HFpEF) following transcatheter aortic valve replacement (TAVR).

OBJECTIVE

This study aimed to enhance the performance of risk assessment models in this patient population by developing a predictive model for adverse outcomes using various machine learning (ML) techniques.

METHODS

This multicenter cohort study included 326 patients diagnosed with severe AS and HFpEF who underwent TAVR between January 2017 and December 2023. Patients were allocated to training (n=195) and independent validation (n=131) sets based on hospital affiliation. A dual-phase feature selection process, combining least absolute shrinkage and selection operator logistic regression and the Boruta algorithm, was used to identify relevant variables from the multimodal dataset. A total of 5 ML model-decision trees, K-nearest neighbors, random forest, support vector machine, and extreme gradient boosting were used to construct a visualization and explainable predictive framework to elucidate model decision-making processes.

RESULTS

The primary features identified included age, N-terminal pro-brain natriuretic peptide, fasting blood glucose, triglyceride/high-density lipoprotein cholesterol ratio, triglyceride glucose index, triglyceride glucose-BMI index, atherogenic index of plasma index, and Apolipoprotein B. Among the 5 models, the support vector machine demonstrated the best predictive performance for major adverse cardiovascular and cerebrovascular events in patients with severe AS and HFpEF following TAVR, achieving an area under the curve of 0.756 (95% CI 0.631-0.881) in the independent validation set. The model exhibited good calibration and robust predictive power in both training and validation sets and demonstrated the highest net benefit in decision curve analysis compared to other models. To extract significant variables influencing the algorithm and ensure model appropriateness, we interpreted cohort and personalized model predictions using Shapley Additive Explanations values.

CONCLUSIONS

Our ML-based multimodal model, incorporating 8 readily accessible predictors, demonstrated robust predictive capability for 12 months of major adverse cardiovascular and cerebrovascular events risk. This model can be used to identify high-risk individuals with AS and HFpEF following TAVR, potentially aiding in risk stratification and personalized treatment strategies.

MultiOmics analysis of metabolic dysregulation and immune features in breast cancer

Metabolic disorders and diminished immune response are hallmark characteristics of tumors. However, limited studies have comprehensively integrated metabolic and immunological factors to evaluate or predict the prognosis of cancer patients. In this study, we utilized 72 metabolic pathway gene sets from the MsigDB database to conduct GSVA, univariate regression, and prognostic analyses on 247 breast cancer samples sourced from the TCGA and GEO databases. Consequently, five metabolic pathways with significant research value were identified. Based on these findings, unsupervised clustering was performed on the breast cancer samples to compare differences in gene expression, clinicopathological features, immune infiltration levels, and prognosis across different clusters. This process led to the identification of nine metabolism-related characteristic genes. Additionally, single-cell sequencing analysis was employed to assess the spatial expression patterns of these characteristic genes, revealing significantly higher expression indices in tumor cells compared to non-tumor cells. Subsequently, machine learning algorithms were applied to reconstruct metabolic risk models for evaluating the prognosis of breast cancer patients. The results indicated that the high metabolic risk group exhibited higher gene mutation scores, a greater proportion of unfavorable clinicopathological parameters, and lower chemokine and immune scores compared to the low-risk group. In conclusion, the metabolic risk model constructed using metabolism-related characteristic genes can accurately distinguish and predict the survival prognosis and immunotherapy outcomes of breast cancer patients, offering novel targets and insights for personalized treatment strategies.

Development and validation of a prediction model for the risk of citrate accumulation in critically ill patients with citrate anticoagulation for continuous renal replacement therapy: a retrospective cohort study based on MIMIC-IV database

BACKGROUND

Acute kidney injury (AKI) is a common clinical syndrome, especially in the intensive care unit (ICU), with an incidence of more than 50% and in-hospital mortality of 30%. Continuous renal replacement therapy (CRRT) is an important supportive treatment for patients with AKI (Patel in Trauma Surg Acute Care Open e001381, 2024). Citrate is the preferred anticoagulant for critically ill patients requiring CRRT. Unfortunately, such patients may be confronted with citrate accumulation during citrate anticoagulation.

METHODS

The MIMIC-IV2.2 database was used to extract data of patients undergoing CRRT who opted for citrate anticoagulation during ICU admission, including 883 critically ill patients. These 883 patients were randomized into training (n = 618) and Internal validation (n = 265) groups at a ratio of 7:3. Least Absolute Shrinkage and Selection Operator(LASSO)-logistic regression was utilized to screen the variables and construct the prediction model, followed by the plotting of the nomogram. Then, Utilizing the retrospective data from the ICU at Jiangbei Hospital in Nanjing, China, from 2014 to 2024 (n = 200) for external model validation, the model was evaluated with discriminant analysis, calibration curves, decision curve analysis, and rationality analysis.

RESULTS

A total of 883 critically ill patients undergoing CRRT were included, consisting of 542 males and 341 females, with a mean age of 65 ± 14 years. Additionally, there were 618 patients in the training set and 265 in the validation set. A total of 47 independent variables were obtained, among which 15 independent variables were screened with LASSO regression and included in the multivariate logistic analysis. The five risk factors ultimately included in the prediction model were height, hepatic insufficiency, mechanical ventilation, prefilter replacement rate, and albumin. The area under the receiver operating characteristic curve (ROC) of the model was 0.758 (0.701-0.816), 0.747 (0.678-0.817), and 0.714 (0.632-0.810) for the training set, internal validation set, and external validation set, respectively. The calibration curves in the training set and internal/external validation sets showed a high degree of consistency between predicted values and observed values (according to the Hosmer-Lemeshow test, the P-values were 0.7673, 0.2401, and 0.4512 for the training set, internal validation set, and external validation set, respectively). In addition, the Decision-Curve(DCA) revealed that the model had good clinical applicability. Nomo-score comparisons exhibited the rationality of the model.

CONCLUSION

The model developed based on LASSO-logistic regression can reliably predict the risk of citrate accumulation in critically ill patients with citrate anticoagulation for CRRT, providing valuable guidance for the application of early measures to prevent the occurrence of citrate accumulation and to improve the prognosis of patients.

A fall risk prediction model based on the CHARLS database for older individuals in China

BACKGROUND

Falls represent the second leading cause of injury-related mortality among older adults globally. The occurrence of falls is the consequence of the interaction of numerous complex risk factors. The objective of this study was to develop a validated fall risk prediction model for the Chinese older individuals.

METHODS

The study used data from the China Health and Retirement Longitudinal Study (CHARLS), a dataset representative of the Chinese population. Thirty-eight indicators including biological factors, behavioral factors and health status were analyzed in this study. The study cohort was randomly divided into the training set (70%) and the validation set (30%). Variables were screened using LASSO regression analysis, the best predictive model based on 10-fold cross-validation, logistic regression model was applied to explore the correlates of fall risk in the older individuals, a nomogram was constructed to develop the predictive model, calibration curves were applied to assess the accuracy of the nomogram model, and predictive performance was assessed by area under the receiver operating characteristic curve and decision curve analysis.

RESULT

A total of 4,913 cases from the 2015 CHARLS database for people aged 60 years and older were ultimately included, and a total of 1,082 (22.02%) of the older individuals had experienced a fall within two years. Multivariate logistic regression analysis showed that Sleeping time, Hearing, Grip strength, ADL score, Cognition, Depression, Health, KD, and Pain DRUG were predictors of fall risk in the older individuals. These factors were used to construct nomogram models that showed good agreement and accuracy. The AUC value for the predictive model was 0.644 (95% CI = 0.621-0.666), with a specificity of 0.695 and a sensitivity of 0.522. For the internal validation set, the AUC value was 0.644 (95% CI = 0.611-0.678), with a specificity of 0.629 and a sensitivity of 0.577. The Hosmer-Lemeshow test value of the model for the training set is p = 0.9368 and for the validation set is p = 0.8545 (both > 0.05). The calibration curves show a more significant agreement between the nomogram model and the actual observations. The ROC and DCA indicate a better predictive performance of the nomogram.

CONCLUSION

The comprehensive nomogram constructed in this study is a promising and convenient tool for assessing the risk of falls in the Chinese older individuals and to help older adults understand the risk level of falls, avoid and eliminate modifiable risk factors, and reduce the incidence of falls.

CLINICAL TRIAL NUMBER

Not applicable.

A predictive model of cognitive impairment risk in older adults with hypertension

BACKGROUND

Hypertension is one of the most common diseases in the world, impacting global life expectancy and associated with an increased risk of cognitive impairment.

OBJECTIVE

This study aimed to develop a nomogram that accurately predicts the risk of cognitive impairment in hypertensive patients using the National Health and Nutrition Examination Study (NHANES).

METHODS

A total of 1517 hypertensive patients from NHANES 2011-2014 were included in this study. The population was divided into two groups: 1065 cases (70 %) in the train set and 452 cases (30 %) in the test set. Lasso regression model and multivariate logistic regression analyses identified predictors significantly associated with cognitive impairment, and the nomogram was constructed using these predictors. The performance of the model was assessed using metrics such as area under the curve (AUC) of receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA).

RESULTS

The nomogram identified seven predictors, including sex, age, education, poverty income ratio (PIR), depression, vigorous work activity, and creatinine. A web-based dynamic nomogram (https://cognitive-impairment-in-hypertension.shinyapps.io/DynNomapp/) was constructed based on these factors. The AUC of the train set was 0.802 and the AUC of the test set was 0.756, indicating that the model had excellent discriminative ability. The calibration curve showed that the model was well-calibrated. The DCA indicated that early intervention for those at risk would result in a net benefit.

CONCLUSION

The model performed well and was clinically predictive, making it easy for clinicians to use and screen for possible cognitive impairment in elderly hypertensive patients.

A Simple Nomogram for Predicting Extended High-Frequency Hearing Loss in Pilots Despite Normal Audiometry: A Retrospective Study

BACKGROUND

The extended high-frequency (EHF; 0.9-16 kHz) region is sensitive to noise exposure and can indicate early noise-induced hearing loss. EHF hearing loss (EHFHL; >20 dB HL for EHF averages) may affect pilots' noise perception, impacting communication and response in flight. Early identification and monitoring of EHFHL are crucial for pilots' hearing health and flight safety. However, EHF is not included in routine medical assessments for pilots in China. This study aimed to develop a nomogram to predict EHFHL in pilots with normal audiograms (≤20 dB HL at each standard frequency), providing an early intervention tool.

METHODS

A total of 1091 pilots were randomly assigned to the training set (763) and validation set (328). Set characteristics were compared using univariate analysis. In the training set, least absolute shrinkage and selection operator regression identified key predictors, followed by multivariable binary logistic regression to construct a nomogram. The nomogram's performance was evaluated in both sets, assessing calibration, discrimination and clinical utility.

RESULTS

The nomogram incorporated four factors as follows: left-ear high-frequency audiometry threshold averages (HFAs: 3, 4, 6 and 8 kHz; odds ratio [OR] = 1.144; 95% confidence interval [CI] = 1.083-1.210), right-ear HFAs (OR = 1.186, 95% CI = 1.115-1.263), flight time (OR = 1.001, 95% CI = 1-1.001) and triglyceride (OR = 1.393, 95% CI = 1.038-1.885). The model's area under the curve was 0.819 (95% CI = 0.790-0.850) and 0.771 (95% CI = 0.712-0.830) during validation. The predictive model was well calibrated (Hosmer-Lemeshow test, χ2  =  10.77; P = 0.292). Decision curve analysis showed a net benefit for the training set between 4% and 88%, with similar benefits observed for the validation set from 12% to 100%.

CONCLUSION

This study developed and validated the first prediction model for EHFHL in Chinese pilots, demonstrating its reliability and clinical utility. The findings support early detection and personalised monitoring, with potential applications in hearing protection strategies and flight safety.

Development and Validation of a Nomogram for Evaluating the Incident Risk of Pain Catastrophizing Among Patients Who Have Severe Knee Osteoarthritis Awaiting Primary Total Knee Arthroplasty

BACKGROUND

It is clinically important to anticipate the likelihood of pain catastrophizing in patients who undergo total knee arthroplasty (TKA). Persistent pain and diminished physical function following TKA are independently associated with preoperative pain catastrophizing. The purpose of this study was to develop and validate a nomogram model to predict pain catastrophizing in patients who have severe osteoarthritis undergoing primary TKA.

METHODS

Data were collected from patients who have severe osteoarthritis undergoing primary TKA at four tertiary general hospitals in Changsha, China, from September to December 2023. The study cohort was randomly divided into a training group and a validation group in the proportion of 70 to 30%. Least absolute shrinkage and selection operator regression analysis was utilized to select the optimal predictive variables for the model. A nomogram model was created using independent risk factors that were identified through multivariate regression analysis. Their performance was assessed using the concordance index and calibration curves, and their clinical utility was analyzed using decision curve analysis.

RESULTS

A total of 416 patients were included, 291 in the training group and 125 in the validation group. There were 115 (27.6%) who had pain catastrophizing. The predictors contained in the nomogram were pain intensity during activity, anxiety and depression, body mass index, social support, and household. The area under the curve of the nomogram was 0.976 (95% confidence interval = 0.96 to 0.99) for the training group and 0.917 (95% confidence interval = 0.88 to 0.96) for the validation group. The calibration curves confirmed the nomogram's accuracy, and decision curve analysis showed its strong predictive performance.

CONCLUSIONS

The comprehensive nomogram generated in this study was a valid and easy-to-use tool for assessing the risk of pain catastrophizing in preoperative TKA patients, and helped healthcare professionals to screen the high-risk population.

Inferring Staphylococcus aureus host species and cross-species transmission from a genome-based model

BACKGROUND

Staphylococcus aureus is an important pathogen that can colonize humans and various animals. However, the host-associated determinants of S. aureus remain uncertain, which leads to difficulties in inferring its host species and cross-species transmission. We performed a 3-stage genome-wide association study (discovery, confirming, and validation) to compare genetic variation between pig and human S. aureus, aiming to elucidate the host-specific genetic elements (k-mers).

RESULTS

After 3-stage association analyses, we found a subset of 20 consensus-significant host-associated k-mers, which are significantly overrepresented in a specific host. Surprisingly for host prediction, both the final model with the top 5 k-mers and the simplest model with only the most important k-mer achieved a high classification accuracy of 98%, giving a simple target for predicting host species and cross-species transmission of S. aureus. The final classifier with the top 5 k-mers revealed that 97.5% of S. aureus isolates from livestock-exposed workers were predicted as pig origin, suggesting a high cross-species transmission risk. The time-based phylogeny inferred the cross-species transmission directions, indicating that ST9 can cross-species spread from animals to humans while ST59 can cross-species spread in the opposite direction.

CONCLUSION

Our findings provide novel insights into host-associated determinants and an accurate model for inferring S. aureus host species and cross-species transmission.

Machine learning can predict surgical indication: new clustering model from a large adult spine deformity database

PURPOSE

The choice of the best management for Adult Spine Deformity (ASD) is challenging. Health-related quality of life (HRQoL), comorbidities, symptoms and spine geometry, along with surgical risk and potential residual disability play a role, and a definite algorithm for patient management is lacking. Machine learning allows to analyse complex settings more efficiently than other available statistical tools. Aim of this study was to develop a machine-learning algorithm that, based on baseline data, would be able to predict whether an ASD patient would undergo surgery or not.

METHODS

Retrospective evaluation of prospectively collected data. Demographic data, HRQoL and radiographic parameters were collected. Two clustering methods were performed to differentiate groups of patients with similar characteristics. Three models were then used to identify the most relevant variables for management prediction.

RESULTS

Data from 1319 patients were available. Three clusters were identified: older subjects with sagittal imbalance and high PI, younger patients with greater coronal deformity and no sagittal imbalance, older patients with moderate sagittal imbalance and lower PI. The group of younger patients showed the highest error rate for the prediction (37%), which was lower for the other two groups (20-27%). For all groups, quality of life parameters such as the ODI and the SRS 22 and the Cobb angle of the major curve were the strongest predictors of surgical indication, albeit with different odds ratios in each group.

CONCLUSION

Three clusters could be identified along with the variables that, in each, are most likely to drive the choice of management.

Risk prediction models for intensive care unit-acquired weakness in critically ill patients: A systematic review

BACKGROUND

Intensive care unit (ICU)-acquired weakness (ICU-AW) is a critical complication that significantly worsens patient prognosis. It is widely thought that risk prediction models can be harnessed to guide preventive interventions. While the number of ICU-AW risk prediction models is increasing, the quality and applicability of these models in clinical practice remain unclear.

OBJECTIVE

The objective of this study was to systematically review published studies on risk prediction models for ICU-AW.

METHODS

We searched electronic databases (PubMed, Web of Science, The Cochrane Library, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), China National Knowledge Infrastructure (CNKI), China Science and Technology Periodical Database (VIP), and Wanfang Database) from inception to October 2023 for studies on ICU-AW risk prediction models. Two independent researchers screened the literature, extracted data, and assessed the risk of bias and applicability of the included studies.

RESULTS

A total of 2709 articles were identified. After screening, 25 articles were selected, encompassing 25 risk prediction models. The area under the curve for these models ranged from 0.681 to 0.926. Evaluation of bias risk indicated that all included models exhibited a high risk of bias, with three models demonstrating poor applicability. The top five predictors among these models were mechanical ventilation duration, age, Acute Physiology and Chronic Health Evaluation II score, blood lactate levels, and the length of ICU stay. The combined area under the curve of the ten validation models was 0.83 (95% confidence interval: 0.77-0.88), indicating a strong discriminative ability.

CONCLUSIONS

Overall, ICU-AW risk prediction models demonstrate promising discriminative ability. However, further optimisation is needed to address limitations, including data source heterogeneity, potential biases in study design, and the need for robust statistical validation. Future efforts should prioritise external validation of existing models or the development of high-quality predictive models with superior performance.

REGISTRATION

The protocol for this study is registered with the International Prospective Register of Systematic Reviews (registration number: CRD42023453187).

Multimodal Data-Driven Prognostic Model for Predicting Long-Term Prognosis in Patients With Ischemic Cardiomyopathy and Heart Failure With Preserved Ejection Fraction After Coronary Artery Bypass Grafting: A Multicenter Cohort Study

BACKGROUND

Limited data from the literature are available to assess the efficacy of coronary artery bypass grafting in patients with ischemic cardiomyopathy and heart failure with preserved ejection fraction. Therefore, our objective was to use machine learning techniques integrating clinical features, biomarker data, and echocardiography data to enhance comprehension and risk stratification in patients diagnosed with ischemic cardiomyopathy and heart failure with preserved ejection fraction who have undergone coronary artery bypass grafting surgery.

METHODS AND RESULTS

For this study, 294 patients with ischemic cardiomyopathy and heart failure with preserved ejection fraction who underwent coronary artery bypass grafting surgery were assigned to the development cohort (n=176) and the independent validation cohort (n=118). A total of 52 clinical variables were extracted for each patient. The principal clinical end point was the incidence of major adverse cardiovascular events, encompassing cardiac mortality, acute myocardial infarction, acute heart failure, and graft failure. From least absolute shrinkage and selection operator regression, 4 predictors were selected for the final prediction nomogram: diabetes, hypertension, the systemic immune-inflammation index, and NT-proBNP (N-terminal pro-B-type natriuretic peptide). The prediction nomogram achieved satisfactory prediction performance in both the development cohort (C index, 0.768 [95% CI, 0.701-0.835]) and independent validation cohort (C index, 0.633 [95% CI, 0.521-0.745]). Adequate calibration was noted for the likelihood of major adverse cardiovascular events in both the development and independent validation cohorts. Decision curve analysis confirmed the clinical usefulness of the established prediction nomogram.

CONCLUSIONS

A clinically feasible prognostic model, based on preoperative multimodal data, was developed for risk stratification of patients with ischemic heart and heart failure with preserved ejection fraction who receive coronary artery bypass grafting surgery.

REGISTRATION

https://www.chictr.org.cn; Unique identifier: ChiCTR2300074439.

Development and Validation of a Risk Prediction Model for Sarcopenia in Chinese Older Patients with Type 2 Diabetes Mellitus

Purpose

Sarcopenia is a common prevalent age-related disorder among older patients with type 2 diabetes mellitus (T2DM). This study aimed to develop and validate a nomogram model to assess the risk of incident sarcopenia among older patients with T2DM.

Patients and methods

A total of 1434 older patients (≥ 60 years) diagnosed with T2DM between May 2020 and November 2023 were recruited. The study cohort was randomly divided into a training set (n = 1006) and a validation set (n = 428) at the ratio of 7:3. The best-matching predictors of sarcopenia were incorporated into the nomogram model. The accuracy and applicability of the nomogram model were measured by using the area under the receiver operating characteristic curve (AUC), calibration curve, Hosmer-Lemeshow test, and decision curve analysis (DCA).

Results

571 out of 1434 participants (39.8%) had sarcopenia. Nine best-matching factors, including age, body mass index (BMI), diabetic duration, glycated hemoglobin A1c (HbA1c), 25 (OH)Vitamin D, nephropathy, neuropathy, nutrition status, and osteoporosis were selected to construct the nomogram prediction model. The AUC values for training and validation sets were 0.800 (95% CI = 0.773-0.828) and 0.796 (95% CI = 0.755-0.838), respectively. Furthermore, the agreement between predicted and actual clinical probability of sarcopenia was demonstrated by calibration curves, the Hosmer-Lemeshow test (P > 0.05), and DCA.

Conclusion

Sarcopenia was prevalent among older patients with T2DM. A visual nomogram prediction model was verified effectively to evaluate incident sarcopenia in older patients with T2DM, allowing targeted interventions to be implemented timely to combat sarcopenia in geriatric population with T2DM.

Depression screening model for middle-aged and elderly diabetic patients in China

Diabetes is a common global disease closely associated with an increased risk of depression. This study analyzed China Health and Retirement Longitudinal Study (CHARLS) data to examine depression in diabetic patients across China. using 29 variables including demographic, behavioral, health conditions, and mental health parameters. The dataset was randomly divided into a 70% training set and a 30% validation set. Predictive factors significantly associated with depression were identified using least absolute shrinkage and selection operator (LASSO) and logistic regression analysis. A nomogram model was constructed using these predictive factors. The model evaluation included the C-index, calibration curves, the Hosmer-Lemeshow test, and DCA. Depression prevalence was 39.1% among diabetic patients. Multifactorial logistic regression identified significant predictors including gender, permanent address, self-perceived health status, presence of lung disease, arthritis, memory disorders, life satisfaction, cognitive function score, ADL score, and social activity. The nomogram model showed high consistency and accuracy, with AUC values of 0.802 for the training set and 0.812 for the validation set. Both sets showed good model fit with Hosmer-Lemeshow P > 0.05. Calibration curves showed significant consistency between the nomogram model and actual observations. ROC and DCA indicated that the nomogram had a good predictive performance. The nomogram developed in this study effectively assesses depression risk in diabetic patients, helping clinicians identify high-risk individuals. This tool could potentially improve patient outcomes.

Feature genes identification and immune infiltration assessment in abdominal aortic aneurysm using WGCNA and machine learning algorithms

Objective

Abdominal aortic aneurysm (AAA) is a life-threatening vascular condition. This study aimed to discover new indicators for the early detection of AAA and explore the possible involvement of immune cell activity in its development.

Methods

Sourced from the Gene Expression Omnibus, the AAA microarray datasets GSE47472 and GSE57691 were combined to generate the training set. Additionally, a separate dataset (GSE7084) was designated as the validation set. Enrichment analyses were carried out to explore the underlying biological mechanisms using Disease Ontology, Kyoto Encyclopedia of Genes and Genomes, and Gene Ontology. We then utilized weighted gene co-expression network analysis (WGCNA) along with 3 machine learning techniques: least absolute shrinkage and selection operator, support vector machine-recursive feature elimination, and random forest, to identify feature genes for AAA. Moreover, data were validated using the receiver operating characteristic (ROC) curve, with feature genes defined as those having an area under the curve above 85% and a p-value below 0.05. Finally, the single sample gene set enrichment analysis algorithm was applied to probe the immune landscape in AAA and its connection to the selected feature genes.

Results

We discovered 72 differentially expressed genes (DEGs) when comparing healthy and AAA samples, including 36 upregulated and 36 downregulated genes. Functional enrichment analysis revealed that the DEGs associated with AAA are primarily involved in inflammatory regulation and immune response. By intersecting the result of 3 machine learning algorithms and WGCNA, 3 feature genes were identified, including MRAP2, PPP1R14A, and PLN genes. The diagnostic performance of all these genes was strong, as revealed by the ROC analysis. A significant increase in 15 immune cell types in AAA samples was observed, based on the analysis of immune cell infiltration. In addition, the 3 feature genes show a strong linkage with different types of immune cells.

Conclusion

Three feature genes (MRAP2, PPP1R14A, and PLN) related to the development of AAA were identified. These genes are linked to immune cell activity and the inflammatory microenvironment, providing potential biomarkers for early detection and a basis for further research into AAA progression.

Development and Validation of Machine Learning Models for Identifying Prediabetes and Diabetes in Normoglycemia

BACKGROUND

Prediabetes and diabetes are both abnormal states of glucose metabolism (AGM) that can lead to severe complications. Early detection of AGM is crucial for timely intervention and treatment. However, fasting blood glucose (FBG) as a mass population screening method may fail to identify some individuals who are actually AGM but with normoglycemia. This study aimed to develop and validate machine learning (ML) models to identify AGM among individuals with normoglycemia using routine health check-up indicators.

METHODS

According to the American Diabetes Association (ADA) criteria, participants with normoglycemia (FBG ≤ 5.6 mmol/L) were collected from 2019 to 2023, and then divided into AGM and Normal groups using glycosylated haemoglobin (HbA1c) 5.7% as the threshold. Data from 2019 to 2022 were divided into training and internal validation sets at a 7:3 ratio, while data from 2023 were used as the external validation set. Seven ML algorithms-including logistic regression (LR), random forest (RF), support vector machine (SVM), extreme gradient boosting machine, multilayer perceptron (MLP), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost)-were used to build models for identifying AGM in normoglycemia population. Model performance was evaluated using the area under the receiver operating characteristic curve (auROC) and the precision-recall curve (auPR). The feature contributions to the optimal model was visualised using the SHapley Additive exPlanations (SHAP). Finally, an intuitive and user-friendly interactive interface was developed.

RESULTS

A total of 59,259 participants were finally enroled in this study, and then divided into the training set of 32,810, the internal validation set of 14,060, and the external validation set of 12,389. The Catboost model outperformed the others with auROC of 0.806 and 0.794 for the internal and external validation set, respectively. Age was the most important feature influencing the performance of the CatBoost model, followed by fasting blood glucose, red blood cells, haemoglobin, body mass index, and triglyceride-glucose.

CONCLUSION

A well-performed ML model to identify AGM in the normoglycemia population was built, offering significant potential for early intervention and treatment of AGM that would otherwise have been missed.

Development and validation of a risk prediction model for older adults with social isolation in China

BACKGROUND

Older adults are vulnerable to social isolation due to declining physical and cognitive function, decreased interpersonal interactions, and reduced outdoor activities after retirement. This study aimed to develop and validate a predictive model to assess the risk of social isolation among older adults in China.

METHODS

Using data from the 2011 China Health and Retirement Longitudinal Study (CHARLS). The study cohort was randomly divided into training and validation groups in a 70:30 ratio. We used least absolute shrinkage and selection operator (LASSO) regression analysis with tenfold cross-validation to identify optimal predictive factors and examined the correlates of social isolation using logistic regression. A nomogram was constructed for the predictive model, and its accuracy was assessed using calibration curves. The predictive performance of the model was assessed using area under the receiver operating characteristic (ROC) curve and decision curve analysis (DCA).

RESULTS

From the 2011 CHARLS database, 4,747 older adults were included in the final analysis, of whom 1,654 (34.8%) experienced social isolation. Multifactorial logistic regression identified educational level, marital status, gender, physical activity, physical self -maintenance ability, and number of children as predictive factors for social isolation. The predictive model achieved an AUC of 0.739 (95%CI = 0.722-0.756) in the training set and 0.708 (95%CI = 0.681-0.735) in the validation set. The Hosmer-Lemeshow test yielded P values of 0.111 and 0.324, respectively (both P > 0.05), indicating significant agreement between the nomogram and observed outcomes. The nomogram showed excellent predictive ability according to ROC and DCA.

CONCLUSIONS

The predictive model developed to assess the risk of social isolation in the Chinese older adults shows promising utility for early screening and intervention by clinical healthcare professionals.

A dynamic model to predict early occurrence of acute kidney injury in ICU hospitalized cirrhotic patients: a MIMIC database analysis

BACKGROUND

This study aimed to develop a tool for predicting the early occurrence of acute kidney injury (AKI) in ICU hospitalized cirrhotic patients.

METHODS

Eligible patients with cirrhosis were identified from the Medical Information Mart for Intensive Care database. Demographic data, laboratory examinations, and interventions were obtained. After splitting the population into training and validation cohorts, the least absolute shrinkage and selection operator regression model was used to select factors and construct the dynamic online nomogram. Calibration and discrimination were used to assess nomogram performance, and clinical utility was evaluated by decision curve analysis (DCA).

RESULTS

A total of 1254 patients were included in the analysis, and 745 developed AKI. The mean arterial pressure, white blood cell count, total bilirubin level, Glasgow Coma Score, creatinine, heart rate, platelet count and albumin level were identified as predictors of AKI. The developed model had a good ability to differentiate AKI from non-AKI, with AUCs of 0.797 and 0.750 in the training and validation cohorts, respectively. Moreover, the nomogram model showed good calibration. DCA showed that the nomogram had a superior overall net benefit within wide and practical ranges of threshold probabilities.

CONCLUSIONS

The dynamic online nomogram can be an easy-to-use tool for predicting the early occurrence of AKI in critically ill patients with cirrhosis.

Early Identification of Metabolic Syndrome in Adults of Jiaxing, China: Utilizing a Multifactor Logistic Regression Model

Purpose

The purpose of this study is to develop and validate a clinical prediction model for diagnosing Metabolic Syndrome (MetS) based on indicators associated with its occurrence.

Patients and Methods

This study included a total of 26,637 individuals who underwent health examinations at the Jiaxing First Hospital Health Examination Center from January 19, 2022, to December 31, 2022. They were randomly divided into training (n = 18645) and validation (n = 7992) sets in a 7:3 ratio. Firstly, the Least Absolute Shrinkage and Selection Operator (LASSO) regression algorithm was employed for variable selection. Subsequently, a multifactor Logistic regression analysis was conducted to establish the predictive model, accompanied by nomograms. Thirdly, model validation was performed using Receiver Operating Characteristic (ROC) curves, Harrell's concordance index (C-index), calibration plots, and Decision Curve Analysis (DCA), followed by internal validation.

Results

In this study, six predictive indicators were selected, including Body Mass Index, Triglycerides, Blood Pressure, High-Density Lipoprotein Cholesterol, Low-Density Lipoprotein Cholesterol, and Fasting Blood Glucose. The model demonstrated excellent predictive performance, with an AUC of 0.978 (0.976-0.980) for the training set and 0.977 (0.974-0.980) for the validation set in the nomogram. Calibration curves indicated that the model possessed good calibration ability (Training set: Emax 0.081, Eavg 0.005, P = 0.580; Validation set: Emax 0.062, Eavg 0.007, P = 0.829). Furthermore, decision curve analysis suggested that applying the nomogram for diagnosis is more beneficial when the threshold probability of MetS is less than 89%, compared to either treating-all or treating-none at all.

Conclusion

We developed and validated a nomogram based on MetS risk factors, which can effectively predict the occurrence of MetS. The proposed nomogram demonstrates significant discriminative ability and clinical applicability. It can be utilized to identify variables and risk factors for diagnosing MetS at an early stage.

Development and validation of a risk prediction model for motoric cognitive risk syndrome in older adults

OBJECTIVE

The objective of this study was to develop a risk prediction model for motoric cognitive risk syndrome (MCR) in older adults.

METHODS

Participants were selected from the 2015 China Health and Retirement Longitudinal Study database and randomly assigned to the training group and the validation group, with proportions of 70% and 30%, respectively. LASSO regression analysis was used to screen the predictors. Then, identified predictors were included in multivariate logistic regression analysis and used to construct model nomogram. The performance of the model was evaluated by area under the receiver operating characteristic (ROC) curve (AUC), calibration curves and decision curve analysis (DCA).

RESULTS

528 out of 3962 participants (13.3%) developed MCR. Multivariate logistic regression analysis showed that weakness, chronic pain, limb dysfunction score, visual acuity score and Five-Times-Sit-To-Stand test were predictors of MCR in older adults. Using these factors, a nomogram model was constructed. The AUC values for the training and validation sets of the predictive model were 0.735 (95% CI = 0.708-0.763) and 0.745 (95% CI = 0.705-0.785), respectively.

CONCLUSION

The nomogram constructed in this study is a useful tool for assessing the risk of MCR in older adults, which can help clinicians identify individuals at high risk.

Development and validation of a model for predicting the early occurrence of RF in ICU-admitted AECOPD patients: a retrospective analysis based on the MIMIC-IV database

BACKGROUND

This study aims to construct a model predicting the probability of RF in AECOPD patients upon hospital admission.

METHODS

This study retrospectively extracted data from MIMIC-IV database, ultimately including 3776 AECOPD patients. The patients were randomly divided into a training set (n = 2643) and a validation set (n = 1133) in a 7:3 ratio. First, LASSO regression analysis was used to optimize variable selection by running a tenfold k-cyclic coordinate descent. Subsequently, a multifactorial Cox regression analysis was employed to establish a predictive model. Thirdly, the model was validated using ROC curves, Harrell's C-index, calibration plots, DCA, and K-M curve.

RESULT

Eight predictive indicators were selected, including blood urea nitrogen, prothrombin time, white blood cell count, heart rate, the presence of comorbid interstitial lung disease, heart failure, and the use of antibiotics and bronchodilators. The model constructed with these 8 predictors demonstrated good predictive capabilities, with ROC curve areas under the curve (AUC) of 0.858 (0.836-0.881), 0.773 (0.746-0.799), 0.736 (0.701-0.771) within 3, 7, and 14 days in the training set, respectively and the C-index was 0.743 (0.723-0.763). Additionally, calibration plots indicated strong consistency between predicted and observed values. DCA analysis demonstrated favorable clinical utility. The K-M curve indicated the model's good reliability, revealed a significantly higher RF occurrence probability in the high-risk group than that in the low-risk group (P < 0.0001).

CONCLUSION

The nomogram can provide valuable guidance for clinical practitioners to early predict the probability of RF occurrence in AECOPD patients, take relevant measures, prevent RF, and improve patient outcomes.

Development and Validation of a Clinic Machine Learning Classifier for the Prediction of Risk Stratifications of Prostate Cancer Bone Metastasis Progression to Castration Resistance

Objective

To explore the predictive factors and predictive model construction for the progression of prostate cancer bone metastasis to castration resistance.

Methods

Clinical data of 286 patients diagnosed with prostate cancer with bone metastasis, initially treated with endocrine therapy, and progressing to metastatic castration resistant prostate cancer (mCRPC) were collected. By comparing the differences in various factors between different groups with fast and slow occurrence of castration-resistant prostate cancer (CRPC). Kaplan-Meier survival analysis and COX multivariate risk proportional regression model were used to compare the differences in the time to progression to CRPC in different groups. The COX multivariate risk proportional regression model was used to evaluate the impact of candidate factors on the time to progression to CRPC and establish a predictive model. The accuracy of the model was then tested using receiver operating characteristic (ROC) curves and decision curve analysis (DCA).

Results

The median time for 286 mCRPC patients to progress to CRPC was 17 (9.5-28.0) months. Multivariate analysis showed that the lowest value of PSA (PSA nadir), the time when PSA dropped to its lowest value (timePSA), and the number of BM, and LDH were independent risk factors for rapid progression to CRPC. Based on the four independent risk factors mentioned above, a prediction model was established, with the optimal prediction model being a random forest with area under curve (AUC) of 0.946[95% CI: 0.901-0.991] and 0.927[95% CI: 0.864-0.990] in the training and validation cohort, respectively.

Conclusion

After endocrine therapy, the PSA nadir, timePSA, the number of BM, and LDH are the main risk factors for rapid progression to mCRPC in patients with prostate cancer bone metastases. Establishing a CRPC prediction model is helpful for early clinical intervention decision-making.

Multimodal data integration for enhanced longitudinal prediction for cardiac and cerebrovascular events following initial diagnosis of obstructive sleep apnea syndrome

Background

Obstructive sleep apnea syndrome (OSAS), a prevalent condition, often coexists with intricate metabolic issues and is frequently associated with negative cardiovascular outcomes. We developed a longitudinal prediction model integrating multimodal data for cardiovascular risk stratification of patients with an initial diagnosis of OSAS.

Methods

We reviewed the data of patients with new-onset OSAS who underwent diagnostic polysomnography between 2018-19. Patients were treated using standard treatment regimens according to clinical practice guidelines.

Results

Over a median follow-up of 32 months, 98/729 participants (13.4%) experienced our composite outcome. At a ratio of 7:3, cases were randomly divided into development (n = 510) and validation (n = 219) cohorts. A prediction nomogram was created using six clinical factors - sex, age, diabetes mellitus, history of coronary artery disease, triglyceride-glucose index, and apnea-hypopnea index. The prediction nomogram showed excellent discriminatory power, based on Harrell's C-index values of 0.826 (95% confidence interval (CI) = 0.779-0.873) for the development cohort and 0.877 (95% CI = 0.824-0.93) for the validation cohort. Moreover, comparing the predicted and observed major adverse cardiac and cerebrovascular events in both development and validation cohorts indicated that the prediction nomogram was well-calibrated. Decision curve analysis demonstrated the good clinical applicability of the prediction nomogram.

Conclusions

Our findings demonstrated the construction of an innovative visualisation tool that utilises various types of data to predict poor outcomes in Chinese patients diagnosed with OSAS, providing accurate and personalised therapy.

Registration

Chinese Clinical Trial Registry ChiCTR2300075727.

Validation of mitochondrial biomarkers and immune dynamics in polycystic ovary syndrome

PROBLEM

Polycystic ovary syndrome (PCOS), a prevalent endocrine-metabolic disorder, presents considerable therapeutic challenges due to its complex and elusive pathophysiology.

METHOD OF STUDY

We employed three machine learning algorithms to identify potential biomarkers within a training dataset, comprising GSE138518, GSE155489, and GSE193123. The diagnostic accuracy of these biomarkers was rigorously evaluated using a validation dataset using area under the curve (AUC) metrics. Further validation in clinical samples was conducted using PCR and immunofluorescence techniques. Additionally, we investigate the complex interplay among immune cells in PCOS using CIBERSORT to uncover the relationships between the identified biomarkers and various immune cell types.

RESULTS

Our analysis identified ACSS2, LPIN1, and NR4A1 as key mitochondria-related biomarkers associated with PCOS. A notable difference was observed in the immune microenvironment between PCOS patients and healthy controls. In particular, LPIN1 exhibited a positive correlation with resting mast cells, whereas NR4A1 demonstrated a negative correlation with monocytes in PCOS patients.

CONCLUSION

ACSS2, LPIN1, and NR4A1 emerge as PCOS-related diagnostic biomarkers and potential intervention targets, opening new avenues for the diagnosis and management of PCOS.

Cancer-associated fibroblast-secreted FGF7 as an ovarian cancer progression promoter

BACKGROUND

Ovarian cancer (OC) is distinguished by its aggressive nature and the limited efficacy of current treatment strategies. Recent studies have emphasized the significant role of cancer-associated fibroblasts (CAFs) in OC development and progression.

METHODS

Employing sophisticated machine learning techniques on bulk transcriptomic datasets, we identified fibroblast growth factor 7 (FGF7), derived from CAFs, as a potential oncogenic factor. We investigated the relationship between FGF7 expression and various clinical parameters. A series of in vitro experiments were undertaken to evaluate the effect of CAFs-derived FGF7 on OC cell activities, such as proliferation, migration, and invasion. Single-cell transcriptomic analysis was also conducted to elucidate the interaction between FGF7 and its receptor. Detailed mechanistic investigations sought to clarify the pathways through which FGF7 fosters OC progression.

RESULTS

Our findings indicate that higher FGF7 levels correlate with advanced tumor stages, increased vascular invasion, and poorer prognosis. CAFs-derived FGF7 significantly enhanced OC cell proliferation, migration, and invasion. Single-cell analysis and in vitro studies revealed that CAFs-derived FGF7 inhibits the ubiquitination and degradation of hypoxia-inducible factor 1 alpha (HIF-1α) via FGFR2 interaction. Activation of the FGF7/HIF-1α pathway resulted in the upregulation of mesenchymal markers and downregulation of epithelial markers. Importantly, in vivo treatment with neutralizing antibodies targeting CAFs-derived FGF7 substantially reduced tumor growth.

CONCLUSION

Neutralizing FGF7 in the medium or inhibiting HIF-1α signaling reversed the effects of FGF7-mediated EMT, emphasizing the dependence of FGF7-mediated EMT on HIF-1α activation. These findings suggest that targeting the FGF7/HIF-1α/EMT axis may offer new therapeutic opportunities to intervene in OC progression.

Predicting feature genes correlated with immune infiltration in patients with abdominal aortic aneurysm based on machine learning algorithms

Abdominal aortic aneurysm (AAA) is a condition characterized by a pathological and progressive dilatation of the infrarenal abdominal aorta. The exploration of AAA feature genes is crucial for enhancing the prognosis of AAA patients. Microarray datasets of AAA were downloaded from the Gene Expression Omnibus database. A total of 43 upregulated differentially expressed genes (DEGs) and 32 downregulated DEGs were obtained. Function, pathway, disease, and gene set enrichment analyses were performed, in which enrichments were related to inflammation and immune response. AHR, APLNR, ITGA10 and NR2F6 were defined as feature genes via machine learning algorithms and a validation cohort, which indicated high diagnostic abilities by the receiver operating characteristic curves. The cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) method was used to quantify the proportions of immune infiltration in samples of AAA and normal tissues. We have predicted AHR, APLNR, ITGA10 and NR2F6 as feature genes of AAA. CD8 + T cells and M2 macrophages correlated with these genes may be involved in the development of AAA, which have the potential to be developed as risk predictors and immune interventions.

Nomogram for soiling prediction in postsurgery hirschsprung children: a retrospective study

PURPOSE

The aim of this study was to develop a nomogram for predicting the probability of postoperative soiling in patients aged greater than 1 year operated for Hirschsprung disease (HSCR).

MATERIALS AND METHODS

The authors retrospectively analyzed HSCR patients with surgical therapy over 1 year of age from January 2000 and December 2019 at our department. Eligible patients were randomly categorized into the training and validation set at a ratio of 7:3. By integrating the least absolute shrinkage and selection operator [LASSO] and multivariable logistic regression analysis, crucial variables were determined for establishment of the nomogram. And, the performance of nomogram was evaluated by C-index, area under the receiver operating characteristic curve, calibration curves, and decision curve analysis. Meanwhile, a validation set was used to further assess the model.

RESULTS

This study enrolled 601 cases, and 97 patients suffered from soiling. Three risk factors, including surgical history, length of removed bowel, and surgical procedures were identified as predictive factors for soiling occurrence. The C-index was 0.871 (95% CI: 0.821-0.921) in the training set and 0.878 (95% CI: 0.811-0.945) in the validation set, respectively. And, the AUC was found to be 0.896 (95% CI: 0.855-0.929) in the training set and 0.866 (95% CI: 0.767-0.920) in the validation set. Additionally, the calibration curves displayed a favorable agreement between the nomogram model and actual observations. The decision curve analysis revealed that employing the nomogram to predict the risk of soiling occurrence would be advantageous if the threshold was between 1 and 73% in the training set and 3-69% in the validation set.

CONCLUSION

This study represents the first efforts to develop and validate a model capable of predicting the postoperative risk of soiling in patients aged greater than 1 year operated for HSCR. This model may assist clinicians in determining the individual risk of soiling subsequent to HSCR surgery, aiding in personalized patient care and management.

Development and validation of a cognitive dysfunction risk prediction model for the abdominal obesity population

Objectives

This study was aimed to develop a nomogram that can accurately predict the likelihood of cognitive dysfunction in individuals with abdominal obesity by utilizing various predictor factors.

Methods

A total of 1490 cases of abdominal obesity were randomly selected from the National Health and Nutrition Examination Survey (NHANES) database for the years 2011-2014. The diagnostic criteria for abdominal obesity were as follows: waist size ≥ 102 cm for men and waist size ≥ 88 cm for women, and cognitive function was assessed by Consortium to Establish a Registry for Alzheimer's Disease (CERAD), Word Learning subtest, Delayed Word Recall Test, Animal Fluency Test (AFT), and Digit Symbol Substitution Test (DSST). The cases were divided into two sets: a training set consisting of 1043 cases (70%) and a validation set consisting of 447 cases (30%). To create the model nomogram, multifactor logistic regression models were constructed based on the selected predictors identified through LASSO regression analysis. The model's performance was assessed using several metrics, including the consistency index (C-index), the area under the receiver operating characteristic (ROC) curve (AUC), calibration curves, and decision curve analysis (DCA) to assess the clinical benefit of the model.

Results

The multivariate logistic regression analysis revealed that age, sex, education level, 24-hour total fat intake, red blood cell folate concentration, depression, and moderate work activity were significant predictors of cognitive dysfunction in individuals with abdominal obesity (p < 0.05). These predictors were incorporated into the nomogram. The C-indices for the training and validation sets were 0.814 (95% CI: 0.875-0.842) and 0.805 (95% CI: 0.758-0.851), respectively. The corresponding AUC values were 0.814 (95% CI: 0.875-0.842) and 0.795 (95% CI: 0.753-0.847). The calibration curves demonstrated a satisfactory level of agreement between the nomogram model and the observed data. The DCA indicated that early intervention for at-risk populations would provide a net benefit, as indicated by the line graph.

Conclusion

Age, sex, education level, 24-hour total fat intake, red blood cell folate concentration, depression, and moderate work activity were identified as predictive factors for cognitive dysfunction in individuals with abdominal obesity. In conclusion, the nomogram model developed in this study can effectively predict the clinical risk of cognitive dysfunction in individuals with abdominal obesity.

A predictive model for social participation of middle-aged and older adult stroke survivors: the China Health and Retirement Longitudinal Study

Objective

This study aims to develop and validate a prediction model for evaluating the social participation in the community middle-aged and older adult stroke survivors.

Methods

The predictive model is based on data from the China Health and Retirement Longitudinal Study (CHARLS), which focused on individuals aged 45 years or older. The study utilized subjects from the CHARLS 2015 and 2018 wave, eighteen factors including socio-demographic variables, behavioral and health status, mental health parameters, were analyzed in this study. To ensure the reliability of the model, the study cohort was randomly split into a training set (70%) and a validation set (30%). The Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis was used to identify the most effective predictors of the model through a 10-fold cross-validation. The logistic regression model was employed to investigate the factors associated with social participation in stroke patients. A nomogram was constructed to develop a prediction model. Calibration curves were used to assess the accuracy of the nomogram model. The model's performance was evaluated using the area under the curve (AUC) and decision curve analysis (DCA).

Result

A total of 1,239 subjects with stroke from the CHARLS database collected in 2013 and 2015 wave were eligible in the final analysis. Out of these, 539 (43.5%) subjects had social participation. The model considered nineteen factors, the LASSO regression selected eleven factors, including age, gender, residence type, education level, pension, insurance, financial dependence, physical function (PF), self-reported healthy,cognition and satisfaction in the prediction model. These factors were used to construct the nomogram model, which showed a certain extent good concordance and accuracy. The AUC values of training and internal validation sets were 0.669 (95%CI 0.631-0.707) and 0.635 (95% CI 0.573-0.698), respectively. Hosmer-Lemeshow test values were p = 0.588 and p = 0.563. Calibration curves showed agreement between the nomogram model and actual observations. ROC and DCA indicated that the nomogram had predictive performance.

Conclusion

The nomogram constructed in this study can be used to evaluate the probability of social participation in middle-aged individuals and identify those who may have low social participation after experiencing a stroke.

Imaging phenotyping using 18F-FDG PET/CT radiomics to predict micropapillary and solid pattern in lung adenocarcinoma

OBJECTIVES

To develop and validate a machine learning model using 18F-FDG PET/CT radiomics signature and clinical features to predict the presence of micropapillary and solid (MP/S) components in lung adenocarcinoma.

METHODS

Eight hundred and forty-six patients who underwent preoperative PET/CT with pathologically confirmed adenocarcinoma were enrolled. After segmentation, 1688 radiomics features were extracted from PET/CT and selected to construct predictive models. Then, we developed a nomogram based on PET/CT radiomics integrated with clinical features. Receiver operating curves, calibration curves, and decision curve analysis (DCA) were performed for diagnostics assessment and test of the developed models for distinguishing patients with MP/S components from the patients without.

RESULTS

PET/CT radiomics-clinical combined model could well distinguish patients with MP/S components from those without MP/S components (AUC = 0.87), which performed better than PET (AUC = 0.829, p < 0.05) or CT (AUC = 0.827, p < 0.05) radiomics models in the training cohort. In test cohorts, radiomics-clinical combined model outperformed the PET radiomics model in test cohort 1 (AUC = 0.859 vs 0.799, p < 0.05) and the CT radiomics model in test cohort 2 (AUC = 0.880 vs 0.829, p < 0.05). Calibration curve indicated good coherence between all model prediction and the actual observation in training and test cohorts. DCA revealed PET/CT radiomics-clinical model exerted the highest clinical benefit.

CONCLUSION

18F-FDG PET/CT radiomics signatures could achieve promising prediction efficiency to identify the presence of MP/S components in adenocarcinoma patients to help the clinician decide on personalized treatment and surveillance strategies. The PET/CT radiomics-clinical combined model performed best. CRITICAL RELEVANCE STATEMENT: 18F-FDG PET/CT radiomics signatures could achieve promising prediction efficiency to identify the presence of micropapillary and solid components in adenocarcinoma patients to help the clinician decide on personalized treatment and surveillance strategies.

Identification of hub genes and their correlation with immune infiltrating cells in membranous nephropathy: an integrated bioinformatics analysis

BACKGROUND

Membranous nephropathy (MN) is a chronic glomerular disease that leads to nephrotic syndrome in adults. The aim of this study was to identify novel biomarkers and immune-related mechanisms in the progression of MN through an integrated bioinformatics approach.

METHODS

The microarray data were downloaded from the Gene Expression Omnibus (GEO) database. The differentially expressed genes (DEGs) between MN and normal samples were identified and analyzed by the Gene Ontology analysis, the Kyoto Encyclopedia of Genes and Genomes analysis and the Gene Set Enrichment Analysis (GSEA) enrichment. Hub The hub genes were screened and identified by the weighted gene co-expression network analysis (WGCNA) and the least absolute shrinkage and selection operator (LASSO) algorithm. The receiver operating characteristic (ROC) curves evaluated the diagnostic value of hub genes. The single-sample GSEA analyzed the infiltration degree of several immune cells and their correlation with the hub genes.

RESULTS

We identified a total of 574 DEGs. The enrichment analysis showed that metabolic and immune-related functions and pathways were significantly enriched. Four co-expression modules were obtained using WGCNA. The candidate signature genes were intersected with DEGs and then subjected to the LASSO analysis, obtaining a total of 6 hub genes. The ROC curves indicated that the hub genes were associated with a high diagnostic value. The CD4+ T cells, CD8+ T cells and B cells significantly infiltrated in MN samples and correlated with the hub genes.

CONCLUSIONS

We identified six hub genes (ZYX, CD151, N4BP2L2-IT2, TAPBP, FRAS1 and SCARNA9) as novel biomarkers for MN, providing potential targets for the diagnosis and treatment.

Prognostic Model and Tumor Immune Microenvironment Analysis of Complement-Related Genes in Gastric Cancer

Introduction

The complement system is integral to the innate and adaptive immune response, helping antibodies eliminate pathogens. However, the potential role of complement and its modulators in the tumor microenvironment (TME) of gastric cancer (GC) remains unclear.

Methods

This study assessed the expression, frequency of somatic mutations, and copy number variations of complement family genes in GC derived from The Cancer Genome Atlas (TCGA). Lasso and Cox regression analyses were conducted to develop a prognostic model based on the complement genes family, with the training and validation sets taken from the TCGA-GC cohort (n=371) and the International Gene Expression Omnibus (GEO) cohort (n=433), correspondingly. The nomogram assessment model was used to predict patient outcomes. Additionally, the link between immune checkpoints, immune cells, and the prognostic model was investigated.

Results

In contrast to patients at low risk, those at high risk had a less favorable outcome. The prognostic model-derived risk score was shown to serve as a prognostic marker of GC independently, as per the multivariate Cox analysis. Nomogram assessment showed that the model had high reliability for predicting the survival of patients with GC in the 1, 3, 5 years. Additionally, the risk score was positively linked to the expression of immune checkpoints, notably CTLA4, LAG3, PDCD1, and CD274, according to an analysis of immune processes. The core gene C5aR1 in the prognostic model was found to be upregulated in GC tissues in contrast to adjoining normal tissues, and patients with elevated expressed levels of C5aR1 had lower 10-year overall survival (OS) rates.

Conclusion

Our work reveals that complement genes are associated with the diversity and complexity of TME. The complement prognosis model help improves our understanding of TME infiltration characteristics and makes immunotherapeutic strategies more effective.

Prediction of Diabetic Kidney Disease in Newly Diagnosed Type 2 Diabetes Mellitus

Background

Diabetic kidney disease (DKD), a common microvascular complication of diabetes mellitus (DM), is always asymptomatic until it develops to the advanced stage. Thus, we aim to develop a nomogram prediction model for progression to DKD in newly diagnosed type 2 diabetes mellitus (T2DM).

Methods

This was a single-center analysis of prospective data collected from 521 newly diagnosed patients with T2DM. All related clinical records were incorporated, including the triglyceride-glucose index (TyG index). The least absolute shrinkage and selection operator (LASSO) was used to build a prediction model. In addition, discrimination, calibration, and clinical practicality of the nomogram were evaluated.

Results

In this study, 156 participants were incorporated as the validation set, while the remaining 365 were incorporated into the training set. The predictive factors included in the individualized nomogram prediction model included 5 variables. The area under the curve (AUC) for the prediction model was 0.826 (95% CI 0.775 to 0.876), indicating excellent discrimination performance. The model performed exceptionally well in terms of predictive accuracy and clinical applicability, according to calibration curves and decision curve analysis.

Conclusion

The predictive nomogram for the risk of DKD in newly diagnosed T2DM patients had outstanding discrimination and calibration, which could help in clinical practice.

A risk nomogram for predicting prolonged intensive care unit stays in patients with chronic obstructive pulmonary disease

BACKGROUND

Providing intensive care is increasingly expensive, and the aim of this study was to construct a risk column line graph (nomograms)for prolonged length of stay (LOS) in the intensive care unit (ICU) for patients with chronic obstructive pulmonary disease (COPD).

METHODS

This study included 4,940 patients, and the data set was randomly divided into training (n = 3,458) and validation (n = 1,482) sets at a 7:3 ratio. First, least absolute shrinkage and selection operator (LASSO) regression analysis was used to optimize variable selection by running a tenfold k-cyclic coordinate descent. Second, a prediction model was constructed using multifactorial logistic regression analysis. Third, the model was validated using receiver operating characteristic (ROC) curves, Hosmer-Lemeshow tests, calibration plots, and decision-curve analysis (DCA), and was further internally validated.

RESULTS

This study selected 11 predictors: sepsis, renal replacement therapy, cerebrovascular disease, respiratory failure, ventilator associated pneumonia, norepinephrine, bronchodilators, invasive mechanical ventilation, electrolytes disorders, Glasgow Coma Scale score and body temperature. The models constructed using these 11 predictors indicated good predictive power, with the areas under the ROC curves being 0.826 (95%CI, 0.809-0.842) and 0.827 (95%CI, 0.802-0.853) in the training and validation sets, respectively. The Hosmer-Lemeshow test indicated a strong agreement between the predicted and observed probabilities in the training (χ2 = 8.21, p = 0.413) and validation (χ2 = 0.64, p = 0.999) sets. In addition, decision-curve analysis suggested that the model had good clinical validity.

CONCLUSION

This study has constructed and validated original and dynamic nomograms for prolonged ICU stay in patients with COPD using 11 easily collected parameters. These nomograms can provide useful guidance to medical and nursing practitioners in ICUs and help reduce the disease and economic burdens on patients.

Development and validation of a radiopathomic model for predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer patients

BACKGROUND

Neoadjuvant chemotherapy (NAC) has become the standard therapeutic option for early high-risk and locally advanced breast cancer. However, response rates to NAC vary between patients, causing delays in treatment and affecting the prognosis for patients who do not sensitive to NAC.

MATERIALS AND METHODS

In total, 211 breast cancer patients who completed NAC (training set: 155, validation set: 56) were retrospectively enrolled. we developed a deep learning radiopathomics model(DLRPM) by Support Vector Machine (SVM) method based on clinicopathological features, radiomics features, and pathomics features. Furthermore, we comprehensively validated the DLRPM and compared it with three single-scale signatures.

RESULTS

DLRPM had favourable performance for the prediction of pathological complete response (pCR) in the training set (AUC 0.933[95% CI 0.895-0.971]), and in the validation set (AUC 0.927 [95% CI 0.858-0.996]). In the validation set, DLRPM also significantly outperformed the radiomics signature (AUC 0.821[0.700-0.942]), pathomics signature (AUC 0.766[0.629-0.903]), and deep learning pathomics signature (AUC 0.804[0.683-0.925]) (all p < 0.05). The calibration curves and decision curve analysis also indicated the clinical effectiveness of the DLRPM.

CONCLUSIONS

DLRPM can help clinicians accurately predict the efficacy of NAC before treatment, highlighting the potential of artificial intelligence to improve the personalized treatment of breast cancer patients.

Correlation between the triglyceride-glucose index and the onset of atrial fibrillation in patients with non-alcoholic fatty liver disease

BACKGROUND

Non-alcoholic fatty liver disease (NAFLD) is associated with atrial fibrillation (AF). Insulin resistance (IR) is the main cause of the high prevalence of AF in NAFLD patients. The triglyceride-glucose index (TyG) is a novel IR-related indicator implicated in the incidence and severity of NAFLD. However, the role of TyG in determining the risk for AF in patients with NAFLD remains unclear.

METHODS

A retrospective study was conducted on 912 patients diagnosed with NAFLD via ultrasonography. These patients were divided into two groups: (1) NAFLD+ AF and (2) NAFLD+ non-AF. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used to assess the correlation between the TyG index and the high risk for AF. A receiver operating characteristic (ROC) curve was constructed to evaluate the predictive value for the TyG index for AF. Restricted cubic splines (RCS) were used to test the linear correlation between TyG and the risk for AF.

RESULTS

A total of 204 patients with AF and 708 patients without AF were included in this study. The LASSO logistic regression analysis showed that TyG was an independent risk factor for AF (odds ratio [OR] = 4.84, 95% confidence interval [CI] 2.98-7.88, P < 0.001). The RCS showed that the risk for AF increased linearly with TyG over the entire TyG range; this risk was also evident when the patients were analyzed based on sex (P for nonlinear > 0.05). In addition, the correlation between TyG and AF was a consistent finding in subgroup analysis. Furthermore, ROC curve analysis showed that TyG levels combined with traditional risk factors improved the predictive value for atrial fibrillation.

CONCLUSION

The TyG index is useful in assessing the risk for atrial fibrillation in patients with NAFLD. Patients with NAFLD and increased TyG indices have higher risks for atrial fibrillation. Therefore, TyG indices should be assessed when managing patients with NAFLD.

Comprehensive analyses reveal the prognosis and biological function roles of chromatin regulators in lung adenocarcinoma

The present study explored the prognosis and biological function roles of chromatin regulators (CRs) in patients with lung adenocarcinoma (LUAD). Using transcriptome profile and clinical follow-up data of LUAD dataset, we explored the molecular classification, developed, and validated a CR prognostic model, built an individual risk scoring system in LUAD, and compared the clinical and molecular characteristics between different subtypes and risk stratifications. We investigated the chemotherapy sensitivity and predicted potential immunotherapy response. Lastly, we collected the clinical samples and validated the prognosis and potential function role of NAPS2. Our study indicated that LUAD patients could be classified into two subtypes that had obviously different clinical background and molecular features. We constructed a prognostic model with eight CR genes, which was well validated in several other population cohort. We built high- and low-risk stratifications for LUAD patients. Patients from high-risk group were totally different from low-risk groups in clinical, biological function, gene mutation, microenvironment, and immune infiltration levels. We idented several potential molecular compounds for high-risk group treatment. We predicted that high-risk group may have poor immunotherapy response. We finally found that Neuronal PAS Domain Protein 2 (NPAS2) involved in the progression of LUAD via regulating cell adhesion. Our study indicated that CR involved in the progression of LUAD and affect their prognosis. Different therapeutic strategies should be developed for different molecular subtypes and risk stratifications. Our comprehensive analyses uncover specific determinants of CRs in LUAD and provides implications for investigating disease-associated CRs.

Screening marker genes of type 2 diabetes mellitus in mouse lacrimal gland by LASSO regression

Type 2 diabetes mellitus (T2DM) is characterized by insulin resistance and a relative deficiency of insulin. This study aims to screen T2DM-related maker genes in the mouse extraorbital lacrimal gland (ELG) by LASSO regression.C57BLKS/J strain with leptin db/db homozygous mice (T2DM, n = 20) and wild-type mice (WT, n = 20) were used to collect data. The ELGs were collected for RNA sequencing. LASSO regression was conducted to screen marker genes with the training set. Five genes were selected from 689 differentially expressed genes by LASSO regression, including Synm, Elovl6, Glcci1, Tnks and Ptprt. Expression of Synm was downregulated in ELGs of T2DM mice. Elovl6, Glcci1, Tnks, and Ptprt were upregulated in T2DM mice. Area under receiver operating curve of the LASSO model was 1.000(1.000-1.000) and 0.980(0.929-1.000) in the training set and the test set, respectively. The C-index and the robust C-index of the LASSO model were 1.000 and 0.999, respectively, in the training set, and 1.000 and 0.978, respectively, in the test set. In the lacrimal gland of db/db mice, Synm, Elovl6, Glcci1, Tnks and Ptprt can be used as marker genes of T2DM. Abnormal expression of marker genes is related to lacrimal gland atrophy and dry eye in mice.

Machine learning-based in-hospital mortality risk prediction tool for intensive care unit patients with heart failure

OBJECTIVE

Risk stratification of patients with congestive heart failure (HF) is vital in clinical practice. The aim of this study was to construct a machine learning model to predict the in-hospital all-cause mortality for intensive care unit (ICU) patients with HF.

METHODS

eXtreme Gradient Boosting algorithm (XGBoost) was used to construct a new prediction model (XGBoost model) from the Medical Information Mart for Intensive Care IV database (MIMIC-IV) (training set). The eICU Collaborative Research Database dataset (eICU-CRD) was used for the external validation (test set). The XGBoost model performance was compared with a logistic regression model and an existing model (Get with the guideline-Heart Failure model) for mortality in the test set. Area under the receiver operating characteristic cure and Brier score were employed to evaluate the discrimination and the calibration of the three models. The SHapley Additive exPlanations (SHAP) value was applied to explain XGBoost model and calculate the importance of its features.

RESULTS

The total of 11,156 and 9,837 patients with congestive HF from the training set and test set, respectively, were included in the study. In-hospital all-cause mortality occurred in 13.3% (1,484/11,156) and 13.4% (1,319/9,837) of patients, respectively. In the training set, of 17 features with the highest predictive value were selected into the models with LASSO regression. Acute Physiology Score III (APS III), age and Sequential Organ Failure Assessment (SOFA) were strongest predictors in SHAP. In the external validation, the XGBoost model performance was superior to that of conventional risk predictive methods, with an area under the curve of 0.771 (95% confidence interval, 0.757-0.784) and a Brier score of 0.100. In the evaluation of clinical effectiveness, the machine learning model brought a positive net benefit in the threshold probability of 0%-90%, prompting evident competitiveness compare to the other two models. This model has been translated into an online calculator which is accessible freely to the public (https://nkuwangkai-app-for-mortality-prediction-app-a8mhkf.streamlit.app).

CONCLUSION

This study developed a valuable machine learning risk stratification tool to accurately assess and stratify the risk of in-hospital all-cause mortality in ICU patients with congestive HF. This model was translated into a web-based calculator which access freely.

Development and validation of a risk prediction model for frailty in patients with diabetes

BACKGROUND

Frailty is the third most common complication of diabetes after macrovascular and microvascular complications. The aim of this study was to develop a validated risk prediction model for frailty in patients with diabetes.

METHODS

The research used data from the China Health and Retirement Longitudinal Study (CHARLS), a dataset representative of the Chinese population. Twenty-five indicators, including socio-demographic variables, behavioral factors, health status, and mental health parameters, were analyzed in this study. The study cohort was randomly divided into a training set and a validation set at a ratio of 70 to 30%. LASSO regression analysis was used to screen the variables for the best predictors of the model based on a 10-fold cross-validation. The logistic regression model was applied to explore the associated factors of frailty in patients with diabetes. A nomogram was constructed to develop the prediction model. Calibration curves were applied to evaluate the accuracy of the nomogram model. The area under the receiver operating characteristic curve and decision curve analysis were conducted to assess predictive performance.

RESULTS

One thousand four hundred thirty-six patients with diabetes from the CHARLS database collected in 2013 (n = 793) and 2015 (n = 643) were included in the final analysis. A total of 145 (10.9%) had frailty symptoms. Multivariate logistic regression analysis showed that marital status, activities of daily living, waist circumference, cognitive function, grip strength, social activity, and depression as predictors of frailty in people with diabetes. These factors were used to construct the nomogram model, which showed good concordance and accuracy. The AUC values of the predictive model and the internal validation set were 0.912 (95%CI 0.887-0.937) and 0.881 (95% CI 0.829-0.934). Hosmer-Lemeshow test values were P = 0.824 and P = 0.608 (both > 0.05). Calibration curves showed significant agreement between the nomogram model and actual observations. ROC and DCA indicated that the nomogram had a good predictive performance.

CONCLUSIONS

Comprehensive nomogram constructed in this study was a promising and convenient tool to evaluate the risk of frailty in patients with diabetes, and contributed clinicians to screening the high-risk population.

Machine learning-based prediction of candidate gene biomarkers correlated with immune infiltration in patients with idiopathic pulmonary fibrosis

Objective

This study aimed to identify candidate gene biomarkers associated with immune infiltration in idiopathic pulmonary fibrosis (IPF) based on machine learning algorithms.

Methods

Microarray datasets of IPF were extracted from the Gene Expression Omnibus (GEO) database to screen for differentially expressed genes (DEGs). The DEGs were subjected to enrichment analysis, and two machine learning algorithms were used to identify candidate genes associated with IPF. These genes were verified in a validation cohort from the GEO database. Receiver operating characteristic (ROC) curves were plotted to assess the predictive value of the IPF-associated genes. The cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) algorithm was used to evaluate the proportion of immune cells in IPF and normal tissues. Additionally, the correlation between the expression of IPF-associated genes and the infiltration levels of immune cells was examined.

Results

A total of 302 upregulated and 192 downregulated genes were identified. Functional annotation, pathway enrichment, Disease Ontology and gene set enrichment analyses revealed that the DEGs were related to the extracellular matrix and immune responses. COL3A1, CDH3, CEBPD, and GPIHBP1 were identified as candidate biomarkers using machine learning algorithms, and their predictive value was verified in a validation cohort. Additionally, ROC analysis revealed that the four genes had high predictive accuracy. The infiltration levels of plasma cells, M0 macrophages and resting dendritic cells were higher and those of resting natural killer (NK) cells, M1 macrophages and eosinophils were lower in the lung tissues of patients with IPF than in those of healthy individuals. The expression of the abovementioned genes was correlated with the infiltration levels of plasma cells, M0 macrophages and eosinophils.

Conclusion

COL3A1, CDH3, CEBPD, and GPIHBP1 are candidate biomarkers of IPF. Plasma cells, M0 macrophages and eosinophils may be involved in the development of IPF and may serve as immunotherapeutic targets in IPF.

Nomogram model for the risk of insulin resistance in obese children and adolescents based on anthropomorphology and lipid derived indicators

OBJECTIVE

This study aims to screen for measures and lipid-derived indicators associated with insulin resistance (IR) in obese children and adolescents and develop a nomogram model for predicting the risk of insulin resistance.

METHODS

A total of 404 eligible obese children and adolescents aged 10-17 years were recruited for this study from a summer camp between 2019 and 2021. The risk factors were screened using the least absolute shrinkage and selection operator (LASSO)-logistic regression model, and a nomogram model was developed. The diagnostic value of the model was evaluated by plotting the receiver operator characteristic curve and calculating the area under the curve. Internal validation was performed using the Bootstrap method, with 1000 self-samples to evaluate the model stability. The clinical applicability of the model was assessed by plotting the clinical decision curve.

RESULTS

On the basis of the LASSO regression analysis results, three lipid-related derivatives, TG/HDL-c, TC/HDL-c, and LDL-c/HDL-c, were finally included in the IR risk prediction model. The nomogram model AUC was 0.804 (95% CI: 0.760 to 0.849). Internal validation results show a C-Index of 0.799, and the mean absolute error between the predicted and actual risks of IR was 0.015. The results of the Hosmer-Lemeshow goodness-of-fit test show a good model prediction (χ² = 9.523, P = 0.300).

CONCLUSION

Three early warning factors, TG/HDL-c, TC/HDL-c, and LDL-c/HDL-c, were screened, which can effectively predict the risk of developing IR in obese children and adolescents, and the nomogram model has an eligible diagnostic value.

Quantitative analysis of early-stage EEG reactivity predicts awakening and recovery of consciousness in patients with severe brain injury

BACKGROUND

Decisions of withdrawal of life-sustaining therapy for patients with severe brain injury are often based on prognostic evaluations such as analysis of electroencephalography (EEG) reactivity (EEG-R). However, EEG-R usually relies on visual assessment, which requires neurophysiological expertise and is prone to inter-rater variability. We hypothesised that quantitative analysis of EEG-R obtained 3 days after patient admission can identify new markers of subsequent awakening and consciousness recovery.

METHODS

In this prospective observational study of patients with severe brain injury requiring mechanical ventilation, quantitative EEG-R was assessed using standard 11-lead EEG with frequency-based (power spectral density) and functional connectivity-based (phase-lag index) analyses. Associations between awakening in the intensive care unit (ICU) and reactivity to auditory and nociceptive stimulations were assessed with logistic regression. Secondary outcomes included in-ICU mortality and 3-month Coma Recovery Scale-Revised (CRS-R) score.

RESULTS

Of 116 patients, 86 (74%) awoke in the ICU. Among quantitative EEG-R markers, variation in phase-lag index connectivity in the delta frequency band after noise stimulation was associated with awakening (adjusted odds ratio=0.89, 95% confidence interval: 0.81-0.97, P=0.02 corrected for multiple tests), independently of age, baseline severity, and sedation. This new marker was independently associated with improved 3-month CRS-R (adjusted β=-0.16, standard error 0.075, P=0.048), but not with mortality (adjusted odds ratio=1.08, 95% CI: 0.99-1.18, P=0.10).

CONCLUSIONS

An early-stage quantitative EEG-R marker was independently associated with awakening and 3-month level of consciousness in patients with severe brain injury. This promising marker based on functional connectivity will need external validation before potential integration into a multimodal prognostic model.

Predicting the Risk of Incorrect Inhalation Technique in Patients with Chronic Airway Diseases by a New Predictive Nomogram

Purpose

To develop and internally validate a nomogram for predicting the risk of incorrect inhalation techniques in patients with chronic airway diseases.

Methods

A total of 206 patients with chronic airway diseases treated with inhaled medications were recruited in this study. Patients were divided into correct (n=129) and incorrect (n=77) cohorts based on their mastery of inhalation devices, which were assessed by medical professionals. Data were collected on the basis of questionnaires and medical records. The least absolute shrinkage and selection operator method (LASSO) and multivariate logistic regression analyses were conducted to identify the risk factors of incorrect inhalation techniques. Then, calibration curve, Harrell's C-index, area under the receiver operating characteristic curve (AUC), decision curve analysis (DCA) and bootstrapping validation were applied to assess the apparent performance, clinical validity and internal validation of the predicting model, respectively.

Results

Seven risk factors including age, education level, drug cognition, self-evaluation of curative effect, inhalation device use instruction before treatment, post-instruction evaluation and evaluation at return visit were finally determined as the predictors of the nomogram prediction model. The ROC curve obtained by this model showed that the AUC was 0.814, with a sensitivity of 0.78 and specificity of 0.75. In addition, the C-index was 0.814, with a Z value of 10.31 (P<0.001). It was confirmed to be 0.783 by bootstrapping validation, indicating that the model had good discrimination and calibration. Furthermore, analysis of DCA showed that the nomogram had good clinical validity.

Conclusion

The application of the developed nomogram to predict the risk of incorrect inhalation techniques during follow-up visits is feasible.

Construction and validation of a risk prediction model for aromatase inhibitor-associated bone loss

Purpose

To establish a high-risk prediction model for aromatase inhibitor-associated bone loss (AIBL) in patients with hormone receptor-positive breast cancer.

Methods

The study included breast cancer patients who received aromatase inhibitor (AI) treatment. Univariate analysis was performed to identify risk factors associated with AIBL. The dataset was randomly divided into a training set (70%) and a test set (30%). The identified risk factors were used to construct a prediction model using the eXtreme gradient boosting (XGBoost) machine learning method. Logistic regression and least absolute shrinkage and selection operator (LASSO) regression methods were used for comparison. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of the model in the test dataset.

Results

A total of 113 subjects were included in the study. Duration of breast cancer, duration of aromatase inhibitor therapy, hip fracture index, major osteoporotic fracture index, prolactin (PRL), and osteocalcin (OC) were found to be independent risk factors for AIBL (p < 0.05). The XGBoost model had a higher AUC compared to the logistic model and LASSO model (0.761 vs. 0.716, 0.691).

Conclusion

The XGBoost model outperformed the logistic and LASSO models in predicting the occurrence of AIBL in patients with hormone receptor-positive breast cancer receiving aromatase inhibitors.

Validation of a Disease-Free Survival Prediction Model Using UBE2C and Clinical Indicators in Breast Cancer Patients

Objective

To explore the validation of a disease-free survival (DFS) model for predicting disease progression based on the combination of ubiquitin-conjugating enzyme E2 C (UBE2C) levels and clinical indicators in breast cancer patients.

Methods

We enrolled 121 patients with breast cancer, collected their baseline characteristics and follow-up data, and analyzed the UBE2C levels in tumor tissues. We studied the relationship between UBE2C expression in tumor tissues and disease progression events of patients. We used the Kaplan-Meier method for identifying the disease-free survival rate of patients, and the multivariate Cox regression analysis to study the risk factors affecting the prognosis of patients. We sought to develop and validate a model for predicting disease progression.

Results

We found that the level of expression of UBE2C could effectively distinguish the prognosis of patients. In the Receiver Operating Characteristic (ROC) curve analysis, the Area under the ROC Curve (AUC) = 0.826 (0.714-0.938) indicating that high levels of UBE2C was a high-risk factor for poor prognosis. After evaluating different models using the ROC curve, Concordance index (C-index), calibration curve, Net Reclassification Index (NRI), Integrated Discrimination Improvement Index (IDI), and other methods, we finally developed a model for the expression of Tumor-Node (TN) staging using Ki-67 and UBE2C, which had an AUC=0.870, 95% CI of 0.786-0.953. The traditional TN model had an AUC=0.717, and 95% CI of 0.581-0.853. Decision Curve Analysis (DCA) and Clinical Impact Curve (CIC) analysis indicated that the model had good clinical benefits and it was relatively simple to use.

Conclusion

We found that high levels of UBE2C was a high-risk factor for poor prognosis. The use of UBE2C in addition to other breast cancer-related indicators effectively predicted the possible disease progression, thus providing a reliable basis for clinical decision-making.