XGBoost Machine Learning Algorithm for Prediction of Outcome in Aneurysmal Subarachnoid Hemorrhage

Predicting lack of clinical improvement following varicose vein ablation using machine learning

OBJECTIVE

Varicose vein ablation is generally indicated in patients with active/healed venous ulcers. However, patient selection for intervention in individuals without venous ulcers is less clear. Tools that predict lack of clinical improvement (LCI) after vein ablation may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year LCI after varicose vein ablation.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent endovenous or surgical varicose vein treatment for Clinical-Etiological-Anatomical-Pathophysiological (CEAP) C2 to C4 disease between 2014 and 2024. We identified 226 predictive features (111 preoperative [demographic/clinical], 100 intraoperative [procedural], and 15 postoperative [immediate postoperative course/complications]). The primary outcome was 1-year LCI, defined as a preoperative Venous Clinical Severity Score (VCSS) minus postoperative VCSS of ≤0, indicating no clinical improvement after vein ablation. The data were divided into training (70%) and test (30%) sets. Six ML models were trained using preoperative features with 10-fold cross-validation (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). The algorithm with the best performance was further trained using intraoperative and postoperative features. The focus was on preoperative features, whereas intraoperative and postoperative features were of secondary importance, because preoperative predictions offer the most potential to mitigate risk, such as deciding whether to proceed with intervention. Model calibration was assessed using calibration plots, and the accuracy of probabilistic predictions was evaluated with Brier scores. Performance was evaluated across subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, prior ipsilateral varicose vein ablation, location of primary vein treated, and treatment type.

RESULTS

Overall, 33,924 patients underwent varicose vein treatment (30,602 endovenous [90.2%] and 3322 surgical [9.8%]) during the study period and 5619 (16.6%) experienced 1-year LCI. Patients who developed the primary outcome were older, more likely to be socioeconomically disadvantaged, and less likely to use compression therapy routinely. They also had less severe disease as characterized by lower preoperative VCSS, Varicose Vein Symptom Questionnaire scores, and CEAP classifications. The best preoperative prediction model was XGBoost, achieving an AUROC of 0.94 (95% confidence interval [CI], 0.93-0.95). In comparison, logistic regression had an AUROC of 0.71 (95% CI, 0.70-0.73). The XGBoost model had marginally improved performance at the intraoperative and postoperative stages, both achieving an AUROC of 0.97 (95% CI, 0.96-0.98). Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.12 (preoperative), 0.11 (intraoperative), and 0.10 (postoperative). Of the top 10 predictors, 7 were preoperative features including VCSS, Varicose Vein Symptom Questionnaire score, CEAP classification, prior varicose vein ablation, thrombus in the greater saphenous vein, and reflux in the deep veins. Model performance remained robust across all subgroups.

CONCLUSIONS

We developed ML models that can accurately predict outcomes after endovenous and surgical varicose vein treatment for CEAP C2 to C4 disease, performing better than logistic regression. These algorithms have potential for important utility in guiding patient counseling and perioperative risk mitigation strategies to prevent LCI after varicose vein ablation.

Construction of a risk prediction model for occupational noise-induced hearing loss using routine blood and biochemical indicators in Shenzhen, China: a predictive modelling study

OBJECTIVES

Occupational noise-induced hearing loss (ONIHL) represents a prevalent occupational health condition, traditionally necessitating multiple pure-tone audiometry assessments. We have developed and validated a machine learning model leveraging routine haematological and biochemical parameters, thereby offering novel insights into the risk prediction of ONIHL.

DESIGN, SETTING AND PARTICIPANTS

This study analysed data from 3297 noise-exposed workers in Shenzhen, including 160 ONIHL cases, with the data set divided into D1 (2868 samples, 107 ONIHL cases) and D2 (429 samples, 53 ONIHL cases). The inclusion criteria were formulated based on the GBZ49-2014 Diagnosis of Occupational Noise-Induced Hearing Loss. Model training was performed using D1, and model validation was conducted using D2. Routine blood and biochemical indicators were extracted from the case data, and a range of machine learning algorithms including extreme gradient boosting (XGBoost) were employed to construct predictive models. The model underwent refinement to identify the most representative variables, and decision curve analysis was conducted to evaluate the net benefit of the model across various threshold levels.

PRIMARY OUTCOME MEASURES

Model creation data set and validation data sets: ONIHL.

RESULTS

The prediction model, developed using XGBoost, demonstrated exceptional performance, achieving an area under the receiver operating characteristic curve (AUC) of 0.942, a sensitivity of 0.875 and a specificity of 0.936 on the validation data set. On the test data set, the model achieved an AUC of 0.990. After implementing feature selection, the model was refined to include only 16 features, while maintaining strong performance on a newly acquired independent data set, with an AUC of 0.872, a balanced accuracy of 0.798, a sensitivity of 0.755 and a specificity of 0.840. The analysis of feature importance revealed that serum albumin (ALB), platelet distribution width (PDW), coefficient of variation in red cell distribution width (RDW-CV), serum creatinine (Scr) and lymphocyte percentage (LYMPHP) are critical factors for risk stratification in patients with ONIHL.

CONCLUSION

The analysis of feature importance identified ALB, PDW, RDW-CV, Scr and LYMPHP as pivotal factors for risk stratification in patients with ONIHL. The machine learning model, using XGBoost, effectively distinguishes patients with ONIHLamong individuals exposed to noise, thereby facilitating early diagnosis and intervention.

Predicting outcomes following open abdominal aortic aneurysm repair using machine learning

Patients undergoing open surgical repair of abdominal aortic aneurysm (AAA) have a high risk of post-operative complications. However, there are no widely used tools to predict surgical risk in this population. We used machine learning (ML) techniques to develop automated algorithms that predict 30-day outcomes following open AAA repair. The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent elective, non-ruptured open AAA repair between 2011 and 2021. Input features included 35 pre-operative demographic/clinical variables. The primary outcome was 30-day major adverse cardiovascular event (MACE; composite of myocardial infarction, stroke, or death). We split our data into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 ML models were trained using pre-operative features with logistic regression as the baseline comparator. Overall, 3,620 patients were included. Thirty-day MACE occurred in 311 (8.6%) patients. The best performing prediction model was XGBoost, achieving an AUROC (95% CI) of 0.90 (0.89-0.91). Comparatively, logistic regression had an AUROC (95% CI) of 0.66 (0.64-0.68). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.03. Our automated ML algorithm can help guide risk-mitigation strategies for patients being considered for open AAA repair to improve outcomes.

High-resolution vessel wall imaging-driven radiomic analysis for the precision prediction of intracranial aneurysm rupture risk: a promising approach

Objective

This study aimed to extract the radiomic features of intracranial aneurysm (IA) and parent artery (PA) walls from high-resolution vessel wall imaging (HR-VWI) images and construct and validate machine learning (ML) predictive models by comparing them with the radiomics score (Rad-score).

Methods

In this study, 356 IAs from 306 patients were retrospectively analyzed at Yuzhong Center and randomly divided into training and test cohorts in an 8:2 ratio. Additionally, 66 IAs from 58 patients were used at Jiangnan Center to validate the predictive model. Radiomic features of the IA and PA walls were extracted from the contrast-enhanced HR-VWI images. Univariate and least absolute shrinkage and selection operator (LASSO) regression analyses were performed on the training cohort features to identify optimal rupture-associated features. The Rad-score model was constructed by calculating the total score derived from the weighted sum of optimal radiomic features, and three ML models were built using the XGBoost, LightGBM, and CART algorithms, and evaluated using both the test and external validation cohorts.

Results

Eight optimal IA wall features and four PA wall features were identified. The Rad-score model demonstrated an area under the curve (AUC) of 0.858, 0.800, and 0.770 for the training, test, and external validation cohorts, respectively. Among the three ML models, the XGBoost model performed best across all cohorts, with AUC values of 0.983, 0.891, and 0.864, respectively. Compared to the Rad-score model, the XGBoost model exhibited superior AUC values (p < 0.05), better calibration curve Brier scores, and greater net clinical benefit.

Conclusion

The radiomic features extracted from HR-VWI images demonstrated robust predictive utility for IA rupture risk in both the Rad-score and ML models. The XGBoost-based ML model outperformed the Rad-score model in efficacy and performance, and proved to be a noninvasive, efficient, and accurate tool for identifying high-risk IA patients.

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

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

Predictive modeling of pregnancy outcomes utilizing multiple machine learning techniques for in vitro fertilization-embryo transfer

OBJECTIVE

This study aims to investigate the influencing factors of pregnancy outcomes during in vitro fertilization and embryo transfer (IVF-ET) procedures in clinical practice. Several prediction models were constructed to predict pregnancy outcomes and models with higher accuracy were identified for potential implementation in clinical settings.

METHODS

The clinical data and pregnancy outcomes of 2625 women who underwent fresh cycles of IVF-ET between 2016 and 2022 at the Reproductive Center of the Affiliated Hospital of Inner Mongolia Medical University were enrolled to establish a comprehensive dataset. The observed features were preprocessed and analyzed. A predictive model for pregnancy outcomes of IVF-ET treatment was constructed based on the processed data. The dataset was divided into a training set and a test set in an 8:2 ratio. Predictive models for clinical pregnancy and clinical live births were developed. The ROC curve was plotted, and the AUC was calculated and the prediction model with the highest accuracy rate was selected from multiple models. The key features and main aspects of IVF-ET treatment outcome prediction were further analyzed.

RESULTS

The clinical pregnancy outcome was categorized into pregnancy and live birth. The XGBoost model exhibited the highest AUC for predicting pregnancy, achieving a validated AUC of 0.999 (95% CI: 0.999-1.000). For predicting live births, the LightGBM model exhibited the highest AUC of 0.913 (95% CI: 0.895-0.930).

CONCLUSION

The XGBoost model predicted the possibility of pregnancy with an accuracy of up to 0.999. While the LightGBM model predicted the possibility of live birth with an accuracy of up to 0.913.

Using Machine Learning to Predict Outcomes Following Thoracic and Complex Endovascular Aortic Aneurysm Repair

BACKGROUND

Thoracic endovascular aortic repair (TEVAR) and complex endovascular aneurysm repair (EVAR) are complex procedures that carry a significant risk of complications. While risk prediction tools can aid in clinical decision making, they remain limited. We developed machine learning algorithms to predict outcomes following TEVAR and complex EVAR.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent elective TEVAR and complex EVAR for noninfrarenal aortic aneurysms between 2012 and 2023. We extracted 172 features from the index hospitalization, including 93 preoperative (demographic/clinical), 46 intraoperative (procedural), and 33 postoperative (in-hospital course/complications) variables. The primary outcome was 1-year thoracoabdominal aortic aneurysm life-altering event, defined as new permanent dialysis, new permanent paralysis, stroke, or death. The data were split into training (70%) and test (30%) sets. We trained 6 machine learning models using preoperative features with 10-fold cross-validation. Model robustness was evaluated using calibration plots and Brier scores.

RESULTS

Overall, 10 738 patients underwent TEVAR or complex EVAR, with 1485 (13.8%) experiencing 1-year thoracoabdominal aortic aneurysm life-altering event. Extreme Gradient Boosting was the best preoperative prediction model, achieving an area under the receiver operating characteristic curve of 0.96 (95% CI, 0.95-0.97), compared with 0.70 (95% CI, 0.68-0.72) for logistic regression. The Extreme Gradient Boosting model maintained excellent performance at the intra- and postoperative stages, with areas under the receiver operating characteristic curves of 0.97 (95% CI, 0.96-0.98) and 0.98 (95% CI, 0.97-0.99), respectively. Calibration plots indicated good agreement between predicted/observed event probabilities, with Brier scores of 0.09 (preoperative), 0.08 (intraoperative), and 0.05 (postoperative).

CONCLUSIONS

Machine learning models can accurately predict 1-year outcomes following TEVAR and complex EVAR, performing better than logistic regression.

Construction and evaluation of machine learning-based predictive models for early-onset preeclampsia

OBJECTIVE

To analyze the influencing factors of early-onset preeclampsia (EOPE). And to construct and validate the prediction model of EOPE using machine learning algorithm.

STUDY DESIGN

Based on Python system, the data profile of 1040 pregnant women was divided into 80% training set and 20% test set. Logistic regression algorithm, XGBoost algorithm, random forest algorithm, support vector machine algorithm and artificial neural network algorithm were used to construct the EOPE prediction model, respectively, and the resulting model was validated by resampling method. Accuracy, sensitivity, specificity, F1 score, and area under the ROC curve were used to evaluate the resulting models and screen the optimal models.

MAIN OUTCOME MEASURES

EOPE in pregnant women.

RESULTS

The results of binary logistic regression showed that the influencing factors of EOPE included six indicators: pre-pregnancy BMI, number of pregnancies, mean arterial pressure, smoking, alpha-fetoprotein, and methods of conception. Among them, the prediction model of EOPE constructed based on the XGBoost algorithm performed the best in the training and test sets, with an F1 score of 0.554 ± 0.068 and an AUC of 0.963 (95 % CI: 0.943 ∼ 0.983) in the training set, and an F1 score of 0.488 ± 0.082 and an AUC of 0.936 (95 % CI: 0.887 ∼ 0.983).

CONCLUSION

Our prediction model for EOPE constructed based on the XGBoost algorithm has superior disease prediction ability and can provide assistance in predicting the disease risk of EOPE.

Microbiota and metabolomic profiling coupled with machine learning to identify biomarkers and drug targets in nasopharyngeal carcinoma

Background

Nasopharyngeal carcinoma (NPC) is a prevalent malignancy in certain regions, with radiotherapy as the standard treatment. However, resistance to radiotherapy remains a critical challenge, necessitating the identification of novel biomarkers and therapeutic targets. The tumor-associated microbiota and metabolites have emerged as potential modulators of radiotherapy outcomes.

Methods

This study included 22 NPC patients stratified into radiotherapy-responsive (R, n = 12) and radiotherapy-non-responsive (NR, n = 10) groups. Tumor tissue and fecal samples were subjected to 16S rRNA sequencing to profile microbiota composition and targeted metabolomics to quantify short-chain fatty acids (SCFAs). The XGBoost algorithm was applied to identify microbial taxa associated with radiotherapy response, and quantitative PCR (qPCR) was used to validate key findings. Statistical analyses were conducted to assess differences in microbial diversity, relative abundance, and metabolite levels between the groups.

Results

Significant differences in alpha diversity at the species level were observed between the R and NR groups. Bacteroides acidifaciens was enriched in the NR group, while Propionibacterium acnes and Clostridium magna were more abundant in the R group. Machine learning identified Acidosoma, Propionibacterium acnes, and Clostridium magna as key predictors of radiotherapy response. Metabolomic profiling revealed elevated acetate levels in the NR group, implicating its role in tumor growth and immune evasion. Validation via qPCR confirmed the differential abundance of these microbial taxa in both tumor tissue and fecal samples.

Discussion

Our findings highlight the interplay between microbiota and metabolite profiles in influencing radiotherapy outcomes in NPC. These results suggest that targeting the microbiota-metabolite axis may enhance radiotherapy efficacy in NPC.

Comparison of prediction for short-term and long-term outcomes in patients with aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis

Despite extensive research on prediction models for outcomes in aneurysmal subarachnoid hemorrhage (aSAH) patients, the distinction between models for short- and long-term outcomes remains insufficiently explored. This study aims to compare these models, identify the risk factors of poor outcomes, summarize the predictors of outcomes, and assess the performance of the prediction models for short- and long-term outcomes in aSAH patients. PubMed, Web of Science, the Cochrane Library, and Embase were searched to identify studies investigating risk factors for developed and/or validated prediction models for short-term (< 12 months) and long-term (≥ 12 months) outcomes in aSAH patients. The main outcome was neurological function, defined as poor if the Glasgow Outcome Scale (GOS) score was ≤ 3, or if the modified Rankin Scale (mRS) score was ≥ 3. Fifty-six studies reporting 61 models with 36,879 aSAH patients were included. A total of 93 predictors were examined and categorized into six domains including demographic factors, scoring systems, clinical factors, aneurysm characteristics, laboratory examinations, and imaging features. Among these, laboratory examinations were included in 57.45% (27/47) of models predicting short-term outcomes, while only 14.29% (2/14) of long-term prediction models incorporated them. An mFisher score of 3-4 [OR = 1.95, 95%CI (1.43, 2.64), P < 0.01] and the presence of multiple aneurysms [OR = 1.56, 95% CI (1.25, 1.94), P < 0.01] were identified as risk factors for poor short-term outcomes, however, this association was weakened in predicting poor long-term outcomes. All studies were found to have a high risk of bias, primarily due to inappropriate data sources and inadequate reporting of the analysis domain. This review suggested that aSAH patients with poor clinical scores and hypertension are at a higher risk of poor outcomes. The majority of the included prediction models perform well, but generally lack reporting in the analysis domain, which may hinder their clinical applicability.

Machine learning for predicting poor outcomes in aneurysmal subarachnoid hemorrhage: A systematic review and meta-analysis involving 8445 participants

Early prediction of poor outcomes in patients impacted with aneurysmal subarachnoid hemorrhage (aSAH) is crucial for timely intervention and effective management. This systematic review and meta-analysis aimed to evaluate the performance of machine learning (ML) algorithms in predicting poor outcomes in patients with aSAH, assessing their sensitivity, specificity, and other algorithm metrics. A comprehensive search of PubMed, Scopus, Embase, Web of science and Cochrane library conducted to identify eligible studies. We extracted data on sensitivity, specificity, accuracy, precision, F1score and area under the curve (AUC) from the included studies. Out of 2238 studies screened, 12 met our eligibility criteria involving ML approaches in patients with confirmed aSAH. ML algorithms, particularly XGBoost and CatBoost, offer promising performance for predicting poor outcomes in aSAH patients. Meta-analysis was performed on 12 studies resulted in a pooled sensitivity of 0.88 [95 % CI: 0.76-0.94], specificity of 0.78 [95 % CI 0.66-0.86], positive DLR of 3.91 [95 % CI: 2.42-6.30], negative DLR of 0.16 [95 % CI: 0.07-0.34], diagnostic odds ratio of 24.9 [95 % CI: 7.97-77.82], the diagnostic score of 3.21[95 % CI: 2.08-4.35], and the area AUC was 0.82, indicating substantial diagnostic performance. However, conventional LR showed slightly superior predictive function compared to ML algorithms. These findings underscore the potential of ML algorithms to significantly advance the predictability of poor outcomes in patients with aSAH, suggesting that ML can play a critical role in enhancing clinical decision-making.

Using machine learning to predict outcomes following transcarotid artery revascularization

Transcarotid artery revascularization (TCAR) is a relatively new and technically challenging procedure that carries a non-negligible risk of complications. Risk prediction tools may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year outcomes following TCAR. The Vascular Quality Initiative (VQI) database was used to identify patients who underwent TCAR between 2016 and 2023. We identified 115 features from the index hospitalization (82 pre-operative [demographic/clinical], 14 intra-operative [procedural], and 19 post-operative [in-hospital course/complications]). The primary outcome was 1-year post-procedural stroke or death. The data was divided into training (70%) and test (30%) sets. Six ML models were trained using pre-operative features with tenfold cross-validation. Overall, 38,325 patients were included (mean age 73.1 [SD 9.0] years, 14,248 [37.2%] female) and 2,672 (7.0%) developed 1-year stroke or death. The best pre-operative prediction model was XGBoost, achieving an AUROC of 0.91 (95% CI 0.90-0.92). In comparison, logistic regression had an AUROC of 0.68 (95% CI 0.66-0.70). The XGBoost model maintained excellent performance at the intra- and post-operative stages, with AUROC's (95% CI's) of 0.92 (0.91-0.93) and 0.94 (0.93-0.95), respectively. Our ML algorithm has potential for important utility in guiding peri-operative risk-mitigation strategies to prevent adverse outcomes following TCAR.

Interpretable machine learning and radiomics in hip MRI diagnostics: comparing ONFH and OA predictions to experts

Purpose

Distinguishing between Osteonecrosis of the femoral head (ONFH) and Osteoarthritis (OA) can be subjective and vary between users with different backgrounds and expertise. This study aimed to construct and evaluate several Radiomics-based machine learning models using MRI to differentiate between those two disorders and compare their efficacies to those of medical experts.

Methods

140 MRI scans were retrospectively collected from the electronic medical records. They were split into training and testing sets in a 7:3 ratio. Handcrafted radiomics features were harvested following the careful manual segmentation of the regions of interest (ROI). After thoroughly selecting these features, various machine learning models have been constructed. The evaluation was carried out using receiver operating characteristic (ROC) curves. Then NaiveBayes (NB) was selected to establish our final Radiomics-model as it performed the best. Three users with different expertise and backgrounds diagnosed and labeled the dataset into either OA or ONFH. Their results have been compared to our Radiomics-model.

Results

The amount of handcrafted radiomics features was 1197 before processing; after the final selection, only 12 key features were retained and used. User 1 had an AUC of 0.632 (95% CI 0.4801-0.7843), User 2 recorded an AUC of 0.565 (95% CI 0.4102-0.7196); while User 3 was on top with an AUC of 0.880 (95% CI 0.7753-0.9843). On the other hand, the Radiomics model attained an AUC of 0.971 (95% CI 0.9298-1.0000); showing greater efficacy than all other users. It also demonstrated a sensitivity of 0.937 and a specificity of 0.885. DCA (Decision Curve Analysis displayed that the radiomics-model had a greater clinical benefit in differentiating OA and ONFH.

Conclusion

We have successfully constructed and evaluated an interpretable radiomics-based machine learning model that could distinguish between OA and ONFH. This method has the ability to aid both junior and senior medical professionals to precisely diagnose and take prompt treatment measures.

A prediction study on the occurrence risk of heart disease in older hypertensive patients based on machine learning

OBJECTIVE

Constructing a predictive model for the occurrence of heart disease in elderly hypertensive individuals, aiming to provide early risk identification.

METHODS

A total of 934 participants aged 60 and above from the China Health and Retirement Longitudinal Study with a 7-year follow-up (2011-2018) were included. Machine learning methods (logistic regression, XGBoost, DNN) were employed to build a model predicting heart disease risk in hypertensive patients. Model performance was comprehensively assessed using discrimination, calibration, and clinical decision curves.

RESULTS

After a 7-year follow-up of 934 older hypertensive patients, 243 individuals (26.03%) developed heart disease. Older hypertensive patients with baseline comorbid dyslipidemia, chronic pulmonary diseases, arthritis or rheumatic diseases faced a higher risk of future heart disease. Feature selection significantly improved predictive performance compared to the original variable set. The ROC-AUC for logistic regression, XGBoost, and DNN were 0.60 (95% CI: 0.53-0.68), 0.64 (95% CI: 0.57-0.71), and 0.67 (95% CI: 0.60-0.73), respectively, with logistic regression achieving optimal calibration. XGBoost demonstrated the most noticeable clinical benefit as the threshold increased.

CONCLUSION

Machine learning effectively identifies the risk of heart disease in older hypertensive patients based on data from the CHARLS cohort. The results suggest that older hypertensive patients with comorbid dyslipidemia, chronic pulmonary diseases, and arthritis or rheumatic diseases have a higher risk of developing heart disease. This information could facilitate early risk identification for future heart disease in older hypertensive patients.

Prognostic prediction models for postoperative patients with stage I to III colorectal cancer based on machine learning

BACKGROUND

Colorectal cancer (CRC) is characterized by high heterogeneity, aggressiveness, and high morbidity and mortality rates. With machine learning (ML) algorithms, patient, tumor, and treatment features can be used to develop and validate models for predicting survival. In addition, important variables can be screened and different applications can be provided that could serve as vital references when making clinical decisions and potentially improving patient outcomes in clinical settings.

AIM

To construct prognostic prediction models and screen important variables for patients with stage I to III CRC.

METHODS

More than 1000 postoperative CRC patients were grouped according to survival time (with cutoff values of 3 years and 5 years) and assigned to training and testing cohorts (7:3). For each 3-category survival time, predictions were made by 4 ML algorithms (all-variable and important variable-only datasets), each of which was validated via 5-fold cross-validation and bootstrap validation. Important variables were screened with multivariable regression methods. Model performance was evaluated and compared before and after variable screening with the area under the curve (AUC). SHapley Additive exPlanations (SHAP) further demonstrated the impact of important variables on model decision-making. Nomograms were constructed for practical model application.

RESULTS

Our ML models performed well; the model performance before and after important parameter identification was consistent, and variable screening was effective. The highest pre- and postscreening model AUCs 95% confidence intervals in the testing set were 0.87 (0.81-0.92) and 0.89 (0.84-0.93) for overall survival, 0.75 (0.69-0.82) and 0.73 (0.64-0.81) for disease-free survival, 0.95 (0.88-1.00) and 0.88 (0.75-0.97) for recurrence-free survival, and 0.76 (0.47-0.95) and 0.80 (0.53-0.94) for distant metastasis-free survival. Repeated cross-validation and bootstrap validation were performed in both the training and testing datasets. The SHAP values of the important variables were consistent with the clinicopathological characteristics of patients with tumors. The nomograms were created.

CONCLUSION

We constructed a comprehensive, high-accuracy, important variable-based ML architecture for predicting the 3-category survival times. This architecture could serve as a vital reference for managing CRC patients.

Explainable Thyroid Cancer Diagnosis Through Two-Level Machine Learning Optimization with an Improved Naked Mole-Rat Algorithm

Modern technologies, particularly artificial intelligence methods such as machine learning, hold immense potential for supporting doctors with cancer diagnostics. This study explores the enhancement of popular machine learning methods using a bio-inspired algorithm-the naked mole-rat algorithm (NMRA)-to assess the malignancy of thyroid tumors. The study utilized a novel dataset released in 2022, containing data collected at Shengjing Hospital of China Medical University. The dataset comprises 1232 records described by 19 features. In this research, 10 well-known classifiers, including XGBoost, LightGBM, and random forest, were employed to evaluate the malignancy of thyroid tumors. A key innovation of this study is the application of the naked mole-rat algorithm for parameter optimization and feature selection within the individual classifiers. Among the models tested, the LightGBM classifier demonstrated the highest performance, achieving a classification accuracy of 81.82% and an F1-score of 86.62%, following two-level parameter optimization and feature selection using the naked mole-rat algorithm. Additionally, explainability analysis of the LightGBM model was conducted using SHAP values, providing insights into the decision-making process of the model.

Machine learning-guided differential gene expression analysis identifies a highly-connected seven-gene cluster in triple-negative breast cancer

Background

One of the most challenging cancers is triple-negative breast cancer, which is subdivided into many molecular subtypes. Due to the high degree of heterogeneity, the role of precision medicine remains challenging. With the use of machine learning (ML)-guided gene selection, the differential gene expression analysis can be optimized, and eventually, the process of precision medicine can see great advancement through biomarker discovery.

Purpose

Enhancing precision medicine in the oncology field by identification of the most representative differentially-expressed genes to be used as biomarkers or as novel drug targets.

Methods

By utilizing data from the Gene Expression Omnibus (GEO) repository and The Cancer Genome Atlas (TCGA), we identified the differentially expressed genes using the linear model for microarray analysis (LIMMA) and edgeR algorithms, and applied ML-based feature selection using several algorithms.

Results

A total of 27 genes were selected by merging features identified with both LIMMA and ML-based feature selection methods. The models with the highest area under the curve (AUC) are CatBoost, Extreme Gradient Boosting (XGBoost), Random Forest, and Multi-Layer Perceptron classifiers. ESR1, FOXA1, GATA3, XBP1, GREB1, AR, and AGR2 were identified as hub genes in a highly interconnected cluster.

Conclusion

ML-based gene selection shows a great impact on the identification of hub genes. The ML models built can improve precision oncology in diagnosis and prognosis. The identified hub genes can serve as biomarkers and warrant further research for potential drug target development.

Implementation of a Machine Learning Approach Evaluating Risk Factors for Complications after Single-Stage Augmentation Mastopexy

BACKGROUND

Single-stage mastopexy augmentation is a much-debated intervention due to its complexity and the associated relatively high complication rates. This study aimed to reevaluate the risk factors for these complications using a novel approach based on artificial intelligence and to demonstrate its possible limitations.

PATIENTS AND METHODS

Complete datasets of patients who underwent single-staged augmentation mastopexy during 2014-2023 at one institution by a single surgeon were collected retrospectively. These were subsequently processed and analyzed by CART, RF and XGBoost algorithms.

RESULTS

A total of 342 patients were included in the study, of which 43 (12.57%) reported surgery-associated complications, whereby capsular contracture (n = 19) was the most common. BMI represented the most important variable for the development of complications (FIS = 0.44 in CART). 2.9% of the patients expressed the desire for implant change in the course, with absence of any complications. A statistically significant correlation between smoking and the desire for implant change (p < 0.001) was revealed.

CONCLUSION

The importance of implementing artificial intelligence into clinical research could be underpinned by this study, as risk variables can be reclassified based on factors previously considered less or even irrelevant. Thereby we encountered limitations using ML approaches. Further studies will be needed to investigate the association between smoking, BMI and the current implant size with the desire for implant change without any complications. Moreover, we could show that the procedure can be performed safely without high risk of developing major complications.

LEVEL OF EVIDENCE IV

This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.

Identification and analysis of a cell communication prognostic signature for oral squamous cell carcinoma at bulk and single-cell levels

Head and neck squamous cancer (HNSC) is a heterogenous malignant tumour disease with poor prognosis and has become the current major public health concern worldwide. Oral squamous cell carcinoma (OSCC) is the majority of HNSC. It is still in lack of comprehensive tumour immune microenvironment analysis and prognostic model development for OSCC's clinic practice. Single-cell sequencing data analysis was conducted to identify immune cell subtypes and illustrate cell-cell interaction status in OSCC via R package 'Seurat', 'Harmony', 'elldex' and 'CellChat'. Base on the bulk sequencing data, WGCNA analysis was employed to identify the CD8+ T cell related gene module. XGBoost was used to construct the gene prognostic model for OSCC. Validation sets and immunotherapy data sets were analysed to further evaluate the model's effectiveness and immunotherapy responsiveness predicting potential. siRNA was used to down regulate FCRL4 expression. Real-time PCR and Western blot were used to validate target gene expression. The effects of FCRL4 on OSCC cells were detected by wound healing, Trans well and clone formation assays. Communication between epithelial cells and tissue stem cells may be the potential key regulators for OSCC progression. By integrating single-cell sequencing data analysis and bulk sequencing data analysis, we constructed a novel immune-related gene prognostic model. The model can effectively predict the prognosis and immunotherapy responsiveness of OSCC patients. In addition, the effects of FCRL4 on OSCC cells were validated. We comprehensively interpreted the immune microenvironment pattern of OSCC based on the single-cell sequencing data and bulk sequencing data analysis. A robust immune feature-based prognostic model was developed for the precise treatment and prognosis evaluation of OSCC.

Predicting inferior vena cava filter complications using machine learning

OBJECTIVE

Inferior vena cava (IVC) filter placement is associated with important long-term complications. Predictive models for filter-related complications may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year IVC filter complications using preoperative data.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent IVC filter placement between 2013 and 2024. We identified 77 preoperative demographic and clinical features from the index hospitalization when the filter was placed. The primary outcome was 1-year filter-related complications (composite of filter thrombosis, migration, angulation, fracture, and embolization or fragmentation, vein perforation, new caval or iliac vein thrombosis, new pulmonary embolism, access site thrombosis, or failed retrieval). The data were divided into training (70%) and test (30%) sets. Six ML models were trained using preoperative features with 10-fold cross-validation (Extreme Gradient Boosting, random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was assessed using calibration plot and Brier score. Performance was evaluated across subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, planned duration of filter, landing site of filter, and presence of prior IVC filter placement.

RESULTS

Overall, 14,476 patients underwent IVC filter placement and 584 (4.0%) experienced 1-year filter-related complications. Patients with a primary outcome were younger (59.3 ± 16.7 years vs 63.8 ± 16.0 years; P < .001) and more likely to have thrombotic risk factors including thrombophilia, prior venous thromboembolism (VTE), and family history of VTE. The best prediction model was Extreme Gradient Boosting, achieving an AUROC of 0.93 (95% confidence interval, 0.92-0.94). In comparison, logistic regression had an AUROC of 0.63 (95% confidence interval, 0.61-0.65). Calibration plot showed good agreement between predicted/observed event probabilities with a Brier score of 0.07. The top 10 predictors of 1-year filter-related complications were (1) thrombophilia, (2) prior VTE, (3) antiphospholipid antibodies, (4) factor V Leiden mutation, (5) family history of VTE, (6) planned duration of IVC filter (temporary), (7) unable to maintain therapeutic anticoagulation, (8) malignancy, (9) recent or active bleeding, and (10) age. Model performance remained robust across all subgroups.

CONCLUSIONS

We developed ML models that can accurately predict 1-year IVC filter complications, performing better than logistic regression. These algorithms have potential to guide patient selection for filter placement, counselling, perioperative management, and follow-up to mitigate filter-related complications and improve outcomes.

Detection of Subarachnoid Hemorrhage Using CNN with Dynamic Factor and Wandering Strategy-Based Feature Selection

Objectives: Subarachnoid Hemorrhage (SAH) is a serious neurological emergency case with a higher mortality rate. An automatic SAH detection is needed to expedite and improve identification, aiding timely and efficient treatment pathways. The existence of noisy and dissimilar anatomical structures in NCCT images, limited availability of labeled SAH data, and ineffective training causes the issues of irrelevant features, overfitting, and vanishing gradient issues that make SAH detection a challenging task. Methods: In this work, the water waves dynamic factor and wandering strategy-based Sand Cat Swarm Optimization, namely DWSCSO, are proposed to ensure optimum feature selection while a Parametric Rectified Linear Unit with a Stacked Convolutional Neural Network, referred to as PRSCNN, is developed for classifying grades of SAH. The DWSCSO and PRSCNN surpass current practices in SAH detection by improving feature selection and classification accuracy. DWSCSO is proposed to ensure optimum feature selection, avoiding local optima issues with higher exploration capacity and avoiding the issue of overfitting in classification. Firstly, in this work, a modified region-growing method was employed on the patient Non-Contrast Computed Tomography (NCCT) images to segment the regions affected by SAH. From the segmented regions, the wide range of patterns and irregularities, fine-grained textures and details, and complex and abstract features were extracted from pre-trained models like GoogleNet, Visual Geometry Group (VGG)-16, and ResNet50. Next, the PRSCNN was developed for classifying grades of SAH which helped to avoid the vanishing gradient issue. Results: The DWSCSO-PRSCNN obtained a maximum accuracy of 99.48%, which is significant compared with other models. The DWSCSO-PRSCNN provides an improved accuracy of 99.62% in CT dataset compared with the DL-ICH and GoogLeNet + (GLCM and LBP), ResNet-50 + (GLCM and LBP), and AlexNet + (GLCM and LBP), which confirms that DWSCSO-PRSCNN effectively reduces false positives and false negatives. Conclusions: the complexity of DWSCSO-PRSCNN was acceptable in this research, for while simpler approaches appeared preferable, they failed to address problems like overfitting and vanishing gradients. Accordingly, the DWSCSO for optimized feature selection and PRSCNN for robust classification were essential for handling these challenges and enhancing the detection in different clinical settings.

Early prediction of ventricular peritoneal shunt dependency in aneurysmal subarachnoid haemorrhage patients by recurrent neural network-based machine learning using routine intensive care unit data

Aneurysmal subarachnoid haemorrhage (aSAH) can lead to complications such as acute hydrocephalic congestion. Treatment of this acute condition often includes establishing an external ventricular drainage (EVD). However, chronic hydrocephalus develops in some patients, who then require placement of a permanent ventriculoperitoneal (VP) shunt. The aim of this study was to employ recurrent neural network (RNN)-based machine learning techniques to identify patients who require VP shunt placement at an early stage. This retrospective single-centre study included all patients who were diagnosed with aSAH and treated in the intensive care unit (ICU) between November 2010 and May 2020 (n = 602). More than 120 parameters were analysed, including routine neurocritical care data, vital signs and blood gas analyses. Various machine learning techniques, including RNNs and gradient boosting machines, were evaluated for their ability to predict VP shunt dependency. VP-shunt dependency could be predicted using an RNN after just one day of ICU stay, with an AUC-ROC of 0.77 (CI: 0.75-0.79). The accuracy of the prediction improved after four days of observation (Day 4: AUC-ROC 0.81, CI: 0.79-0.84). At that point, the accuracy of the prediction was 76% (CI: 75.98-83.09%), with a sensitivity of 85% (CI: 83-88%) and a specificity of 74% (CI: 71-78%). RNN-based machine learning has the potential to predict VP shunt dependency on Day 4 after ictus in aSAH patients using routine data collected in the ICU. The use of machine learning may allow early identification of patients with specific therapeutic needs and accelerate the execution of required procedures.

Using Machine Learning to Predict Outcomes Following Transfemoral Carotid Artery Stenting

BACKGROUND

Transfemoral carotid artery stenting (TFCAS) carries important perioperative risks. Outcome prediction tools may help guide clinical decision-making but remain limited. We developed machine learning algorithms that predict 1-year stroke or death following TFCAS.

METHODS AND RESULTS

The VQI (Vascular Quality Initiative) database was used to identify patients who underwent TFCAS for carotid artery stenosis between 2005 and 2024. We identified 112 features from the index hospitalization (82 preoperative [demographic/clinical], 13 intraoperative [procedural], and 17 postoperative [in-hospital course/complications]). The primary outcome was 1-year postprocedural stroke or death. The data were divided into training (70%) and test (30%) sets. Six machine learning models were trained using preoperative features with 10-fold cross-validation. The primary model evaluation metric was area under the receiver operating characteristic curve. The algorithm with the best performance was further trained using intra- and postoperative features. Model robustness was assessed using calibration plots and Brier scores. Overall, 35 214 patients underwent TFCAS during the study period and 3257 (9.2%) developed 1-year stroke or death. The best preoperative prediction model was extreme gradient boosting, achieving an area under the receiver operating characteristic curve of 0.94 (95% CI, 0.93-0.95). In comparison, logistic regression had an AUROC of 0.65 (95% CI, 0.63-0.67). The extreme gradient boosting model maintained excellent performance at the intra- and postoperative stages, with area under the receiver operating characteristic curve values of 0.94 (95% CI, 0.93-0.95) and 0.98 (95% CI, 0.97-0.99), respectively. Calibration plots showed good agreement between predicted/observed event probabilities with Brier scores of 0.11 (preoperative), 0.11 (intraoperative), and 0.09 (postoperative).

CONCLUSIONS

Machine learning can accurately predict 1-year stroke or death following TFCAS, performing better than logistic regression.

Improve the efficiency and accuracy of ophthalmologists' clinical decision-making based on AI technology

BACKGROUND

As global aging intensifies, the prevalence of ocular fundus diseases continues to rise. In China, the tense doctor-patient ratio poses numerous challenges for the early diagnosis and treatment of ocular fundus diseases. To reduce the high risk of missed or misdiagnosed cases, avoid irreversible visual impairment for patients, and ensure good visual prognosis for patients with ocular fundus diseases, it is particularly important to enhance the growth and diagnostic capabilities of junior doctors. This study aims to leverage the value of electronic medical record data to developing a diagnostic intelligent decision support platform. This platform aims to assist junior doctors in diagnosing ocular fundus diseases quickly and accurately, expedite their professional growth, and prevent delays in patient treatment. An empirical evaluation will assess the platform's effectiveness in enhancing doctors' diagnostic efficiency and accuracy.

METHODS

In this study, eight Chinese Named Entity Recognition (NER) models were compared, and the SoftLexicon-Glove-Word2vec model, achieving a high F1 score of 93.02%, was selected as the optimal recognition tool. This model was then used to extract key information from electronic medical records (EMRs) and generate feature variables based on diagnostic rule templates. Subsequently, an XGBoost algorithm was employed to construct an intelligent decision support platform for diagnosing ocular fundus diseases. The effectiveness of the platform in improving diagnostic efficiency and accuracy was evaluated through a controlled experiment comparing experienced and junior doctors.

RESULTS

The use of the diagnostic intelligent decision support platform resulted in significant improvements in both diagnostic efficiency and accuracy for both experienced and junior doctors (P < 0.05). Notably, the gap in diagnostic speed and precision between junior doctors and experienced doctors narrowed considerably when the platform was used. Although the platform also provided some benefits to experienced doctors, the improvement was less pronounced compared to junior doctors.

CONCLUSION

The diagnostic intelligent decision support platform established in this study, based on the XGBoost algorithm and NER, effectively enhances the diagnostic efficiency and accuracy of junior doctors in ocular fundus diseases. This has significant implications for optimizing clinical diagnosis and treatment.

Artificial Intelligence and Machine Learning Applications in Critically Ill Brain Injured Patients

The utilization of Artificial Intelligence (AI) and Machine Learning (ML) is paving the way for significant strides in patient diagnosis, treatment, and prognostication in neurocritical care. These technologies offer the potential to unravel complex patterns within vast datasets ranging from vast clinical data and EEG (electroencephalogram) readings to advanced cerebral imaging facilitating a more nuanced understanding of patient conditions. Despite their promise, the implementation of AI and ML faces substantial hurdles. Historical biases within training data, the challenge of interpreting multifaceted data streams, and the "black box" nature of ML algorithms present barriers to widespread clinical adoption. Moreover, ethical considerations around data privacy and the need for transparent, explainable models remain paramount to ensure trust and efficacy in clinical decision-making.This article reflects on the emergence of AI and ML as integral tools in neurocritical care, discussing their roles from the perspective of both their scientific promise and the associated challenges. We underscore the importance of extensive validation in diverse clinical settings to ensure the generalizability of ML models, particularly considering their potential to inform critical medical decisions such as withdrawal of life-sustaining therapies. Advancement in computational capabilities is essential for implementing ML in clinical settings, allowing for real-time analysis and decision support at the point of care. As AI and ML are poised to become commonplace in clinical practice, it is incumbent upon health care professionals to understand and oversee these technologies, ensuring they adhere to the highest safety standards and contribute to the realization of personalized medicine. This engagement will be pivotal in integrating AI and ML into patient care, optimizing outcomes in neurocritical care through informed and data-driven decision-making.

Prediction of early neurologic deterioration in patients with perforating artery territory infarction using machine learning: a retrospective study

Background

Early neurological deterioration (END) is a frequent complication in patients with perforating artery territory infarction (PAI), leading to poorer outcomes. Therefore, we aimed to apply machine learning (ML) algorithms to predict the occurrence of END in PAI and investigate related risk factors.

Methods

This retrospective study analyzed a cohort of PAI patients, excluding those with severe stenosis of the parent artery. We included demographic characteristics, clinical features, laboratory data, and imaging variables. Recursive feature elimination with cross-validation (RFECV) was performed to identify critical features. Seven ML algorithms, namely logistic regression, random forest, adaptive boosting, gradient boosting decision tree, histogram-based gradient boosting, extreme gradient boosting, and category boosting, were developed to predict END in PAI patients using these critical features. We compared the accuracy of these models in predicting outcomes. Additionally, SHapley Additive exPlanations (SHAP) values were introduced to interpret the optimal model and assess the significance of input features.

Results

The study enrolled 1,020 PAI patients with a mean age of 60.46 (range 49.11-71.81) years. Of these, 30.39% were women, and 129 (12.65%) experienced END. RFECV selected 13 critical features, including blood urea nitrogen (BUN), total cholesterol (TC), low-density-lipoprotein cholesterol (LDL-C), apolipoprotein B (apoB), atrial fibrillation, loading dual antiplatelet therapy (DAPT), single antiplatelet therapy (SAPT), argatroban, the basal ganglia, the thalamus, the posterior choroidal arteries, maximal axial infarct diameter (measured at < 15 mm), and stroke subtype. The gradient-boosting decision tree had the highest area under the curve (0.914) among the seven ML algorithms. The SHAP analysis identified apoB as the most significant variable for END.

Conclusion

Our results suggest that ML algorithms, especially the gradient-boosting decision tree, are effective in predicting the occurrence of END in PAI patients.

Predicting Outcomes Following Lower Extremity Endovascular Revascularization Using Machine Learning

BACKGROUND

Lower extremity endovascular revascularization for peripheral artery disease carries nonnegligible perioperative risks; however, outcome prediction tools remain limited. Using machine learning, we developed automated algorithms that predict 30-day outcomes following lower extremity endovascular revascularization.

METHODS AND RESULTS

The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent lower extremity endovascular revascularization (angioplasty, stent, or atherectomy) for peripheral artery disease between 2011 and 2021. Input features included 38 preoperative demographic/clinical variables. The primary outcome was 30-day postprocedural major adverse limb event (composite of major reintervention, untreated loss of patency, or major amputation) or death. Data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 machine learning models were trained using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve. Overall, 21 886 patients were included, and 30-day major adverse limb event/death occurred in 1964 (9.0%) individuals. The best performing model for predicting 30-day major adverse limb event/death was extreme gradient boosting, achieving an area under the receiver operating characteristic curve of 0.93 (95% CI, 0.92-0.94). In comparison, logistic regression had an area under the receiver operating characteristic curve of 0.72 (95% CI, 0.70-0.74). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.09. The top 3 predictive features in our algorithm were (1) chronic limb-threatening ischemia, (2) tibial intervention, and (3) congestive heart failure.

CONCLUSIONS

Our machine learning models accurately predict 30-day outcomes following lower extremity endovascular revascularization using preoperative data with good discrimination and calibration. Prospective validation is warranted to assess for generalizability and external validity.

Using Machine Learning (XGBoost) to Predict Outcomes After Infrainguinal Bypass for Peripheral Artery Disease

OBJECTIVE

To develop machine learning (ML) algorithms that predict outcomes after infrainguinal bypass.

BACKGROUND

Infrainguinal bypass for peripheral artery disease carries significant surgical risks; however, outcome prediction tools remain limited.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent infrainguinal bypass for peripheral artery disease between 2003 and 2023. We identified 97 potential predictor variables from the index hospitalization [68 preoperative (demographic/clinical), 13 intraoperative (procedural), and 16 postoperative (in-hospital course/complications)]. The primary outcome was 1-year major adverse limb event (composite of surgical revision, thrombectomy/thrombolysis, or major amputation) or death. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained 6 ML models using preoperative features. The primary model evaluation metric was the area under the receiver operating characteristic curve (AUROC). The top-performing algorithm was further trained using intraoperative and postoperative features. Model robustness was evaluated using calibration plots and Brier scores.

RESULTS

Overall, 59,784 patients underwent infrainguinal bypass, and 15,942 (26.7%) developed 1-year major adverse limb event/death. The best preoperative prediction model was XGBoost, achieving an AUROC (95% CI) of 0.94 (0.93-0.95). In comparison, logistic regression had an AUROC (95% CI) of 0.61 (0.59-0.63). Our XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs (95% CI's) of 0.94 (0.93-0.95) and 0.96 (0.95-0.97), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.08 (preoperative), 0.07 (intraoperative), and 0.05 (postoperative).

CONCLUSIONS

ML models can accurately predict outcomes after infrainguinal bypass, outperforming logistic regression.

Machine Learning to Predict Outcomes of Endovascular Intervention for Patients With PAD

Importance

Endovascular intervention for peripheral artery disease (PAD) carries nonnegligible perioperative risks; however, outcome prediction tools are limited.

Objective

To develop machine learning (ML) algorithms that can predict outcomes following endovascular intervention for PAD.

Design, Setting, and Participants

This prognostic study included patients who underwent endovascular intervention for PAD between January 1, 2004, and July 5, 2023, with 1 year of follow-up. Data were obtained from the Vascular Quality Initiative (VQI), a multicenter registry containing data from vascular surgeons and interventionalists at more than 1000 academic and community hospitals. From an initial cohort of 262 242 patients, 26 565 were excluded due to treatment for acute limb ischemia (n = 14 642) or aneurysmal disease (n = 3456), unreported symptom status (n = 4401) or procedure type (n = 2319), or concurrent bypass (n = 1747). Data were split into training (70%) and test (30%) sets.

Exposures

A total of 112 predictive features (75 preoperative [demographic and clinical], 24 intraoperative [procedural], and 13 postoperative [in-hospital course and complications]) from the index hospitalization were identified.

Main Outcomes and Measures

Using 10-fold cross-validation, 6 ML models were trained using preoperative features to predict 1-year major adverse limb event (MALE; composite of thrombectomy or thrombolysis, surgical reintervention, or major amputation) or death. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). After selecting the best performing algorithm, additional models were built using intraoperative and postoperative data.

Results

Overall, 235 677 patients who underwent endovascular intervention for PAD were included (mean [SD] age, 68.4 [11.1] years; 94 979 [40.3%] female) and 71 683 (30.4%) developed 1-year MALE or death. The best preoperative prediction model was extreme gradient boosting (XGBoost), achieving the following performance metrics: AUROC, 0.94 (95% CI, 0.93-0.95); accuracy, 0.86 (95% CI, 0.85-0.87); sensitivity, 0.87; specificity, 0.85; positive predictive value, 0.85; and negative predictive value, 0.87. In comparison, logistic regression had an AUROC of 0.67 (95% CI, 0.65-0.69). The XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs of 0.94 (95% CI, 0.93-0.95) and 0.98 (95% CI, 0.97-0.99), respectively.

Conclusions and Relevance

In this prognostic study, ML models were developed that accurately predicted outcomes following endovascular intervention for PAD, which performed better than logistic regression. These algorithms have potential for important utility in guiding perioperative risk-mitigation strategies to prevent adverse outcomes following endovascular intervention for PAD.

Using machine learning to predict outcomes following suprainguinal bypass

OBJECTIVE

Suprainguinal bypass for peripheral artery disease (PAD) carries important surgical risks; however, outcome prediction tools remain limited. We developed machine learning (ML) algorithms that predict outcomes following suprainguinal bypass.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent suprainguinal bypass for PAD between 2003 and 2023. We identified 100 potential predictor variables from the index hospitalization (68 preoperative [demographic/clinical], 13 intraoperative [procedural], and 19 postoperative [in-hospital course/complications]). The primary outcomes were major adverse limb events (MALE; composite of untreated loss of patency, thrombectomy/thrombolysis, surgical revision, or major amputation) or death at 1 year following suprainguinal bypass. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). The best performing algorithm was further trained using intra- and postoperative data. Model robustness was evaluated using calibration plots and Brier scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, symptom status, procedure type, prior intervention for PAD, concurrent interventions, and urgency.

RESULTS

Overall, 16,832 patients underwent suprainguinal bypass, and 3136 (18.6%) developed 1-year MALE or death. Patients with 1-year MALE or death were older (mean age, 64.9 vs 63.5 years; P < .001) with more comorbidities, had poorer functional status (65.7% vs 80.9% independent at baseline; P < .001), and were more likely to have chronic limb-threatening ischemia (67.4% vs 47.6%; P < .001) than those without an outcome. Despite being at higher cardiovascular risk, they were less likely to receive acetylsalicylic acid or statins preoperatively and at discharge. Our best performing prediction model at the preoperative stage was XGBoost, achieving an AUROC of 0.92 (95% confidence interval [CI], 0.91-0.93). In comparison, logistic regression had an AUROC of 0.67 (95% CI, 0.65-0.69). Our XGBoost model maintained excellent performance at the intra- and postoperative stages, with AUROCs of 0.93 (95% CI, 0.92-0.94) and 0.98 (95% CI, 0.97-0.99), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.12 (preoperative), 0.11 (intraoperative), and 0.10 (postoperative). Of the top 10 predictors, nine were preoperative features including chronic limb-threatening ischemia, previous procedures, comorbidities, and functional status. Model performance remained robust on all subgroup analyses.

CONCLUSIONS

We developed ML models that accurately predict outcomes following suprainguinal bypass, performing better than logistic regression. Our algorithms have potential for important utility in guiding perioperative risk mitigation strategies to prevent adverse outcomes following suprainguinal bypass.

Predicting Outcomes Following Endovascular Abdominal Aortic Aneurysm Repair Using Machine Learning

OBJECTIVE

To develop machine learning (ML) models that predict outcomes following endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA).

BACKGROUND

EVAR carries non-negligible perioperative risks; however, there are no widely used outcome prediction tools.

METHODS

The National Surgical Quality Improvement Program targeted database was used to identify patients who underwent EVAR for infrarenal AAA between 2011 and 2021. Input features included 36 preoperative variables. The primary outcome was 30-day major adverse cardiovascular event (composite of myocardial infarction, stroke, or death). Data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 ML models were trained using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve. Model robustness was evaluated with calibration plot and Brier score. Subgroup analysis was performed to assess model performance based on age, sex, race, ethnicity, and prior AAA repair.

RESULTS

Overall, 16,282 patients were included. The primary outcome of 30-day major adverse cardiovascular event occurred in 390 (2.4%) patients. Our best-performing prediction model was XGBoost, achieving an area under the receiver operating characteristic curve (95% CI) of 0.95 (0.94-0.96) compared with logistic regression [0.72 [0.70-0.74)]. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.06. Model performance remained robust on all subgroup analyses.

CONCLUSIONS

Our newer ML models accurately predict 30-day outcomes following EVAR using preoperative data and perform better than logistic regression. Our automated algorithms can guide risk mitigation strategies for patients being considered for EVAR.

Predicting outcomes following lower extremity open revascularization using machine learning

Lower extremity open revascularization is a treatment option for peripheral artery disease that carries significant peri-operative risks; however, outcome prediction tools remain limited. Using machine learning (ML), we developed automated algorithms that predict 30-day outcomes following lower extremity open revascularization. The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent lower extremity open revascularization for chronic atherosclerotic disease between 2011 and 2021. Input features included 37 pre-operative demographic/clinical variables. The primary outcome was 30-day major adverse limb event (MALE; composite of untreated loss of patency, major reintervention, or major amputation) or death. Our data were split into training (70%) and test (30%) sets. Using tenfold cross-validation, we trained 6 ML models. Overall, 24,309 patients were included. The primary outcome of 30-day MALE or death occurred in 2349 (9.3%) patients. Our best performing prediction model was XGBoost, achieving an area under the receiver operating characteristic curve (95% CI) of 0.93 (0.92-0.94). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.08. Our ML algorithm has potential for important utility in guiding risk mitigation strategies for patients being considered for lower extremity open revascularization to improve outcomes.

Identification of CXC Chemokine Receptor 2 (CXCR2) as a Novel Eosinophils-Independent Diagnostic Biomarker of Pediatric Eosinophilic Esophagitis by Integrated Bioinformatic and Machine-Learning Analysis

Background

Eosinophilic esophagitis (EoE) is a complex allergic condition frequently accompanied by various atopic comorbidities in children, which significantly affects their life qualities. Therefore, this study aimed to evaluate pivotal molecular markers that may facilitate the diagnosis of EoE in pediatric patients.

Methods

Three available EoE-associated gene expression datasets in children: GSE184182, GSE 197702, GSE55794, along with GSE173895 were downloaded from the GEO database. Differentially expressed genes (DEGs) identified by "limma" were intersected with key module genes identified by weighted gene co-expression network analysis (WGCNA), and the shared genes went through functional enrichment analysis. The protein-protein interaction (PPI) network and the machine learning algorithms: least absolute shrinkage and selection operator (LASSO), random forest (RF), and XGBoost were used to reveal candidate diagnostic markers for EoE. The receiver operating characteristic (ROC) curve showed the efficacy of differential diagnosis of this marker, along with online databases predicting its molecular regulatory network. Finally, we performed gene set enrichment analysis (GSEA) and assessed immune cell infiltration of EoE/control samples by using the CIBERSORT algorithm. The correlations between the key diagnostic biomarker and immune cells were also investigated.

Results

The intersection of 936 DEGs and 1446 key module genes in EoE generated 567 genes, which were primarily enriched in immune regulation. Following the construction of the PPI network and filtration by machine learning, CXCR2 served as a potential diagnostic biomarker of pediatric EoE with a perfect diagnostic efficacy (AUC = ~1.00) in regional tissue/peripheral whole blood samples. Multiple infiltrated immune cells were observed to participate in disrupting the homeostasis of esophageal epithelium to varying degrees.

Conclusion

The immune-correlated CXCR2 gene was proved to be a promising diagnostic indicator for EoE, and dysregulated regulatory T cells (Tregs)/neutrophils might play a crucial role in the pathogenesis of EoE in children.

Using machine learning to predict outcomes following open abdominal aortic aneurysm repair

OBJECTIVE

Prediction of outcomes following open abdominal aortic aneurysm (AAA) repair remains challenging with a lack of widely used tools to guide perioperative management. We developed machine learning (ML) algorithms that predict outcomes following open AAA repair.

METHODS

The Vascular Quality Initiative (VQI) database was used to identify patients who underwent elective open AAA repair between 2003 and 2023. Input features included 52 preoperative demographic/clinical variables. All available preoperative variables from VQI were used to maximize predictive performance. The primary outcome was in-hospital major adverse cardiovascular event (MACE; composite of myocardial infarction, stroke, or death). Secondary outcomes were individual components of the primary outcome, other in-hospital complications, and 1-year mortality and any reintervention. We split our data into training (70%) and test (30%) sets. Using 10-fold cross-validation, six ML models were trained using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. The top 10 predictive features in our final model were determined based on variable importance scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, rurality, median area deprivation index, proximal clamp site, prior aortic surgery, and concomitant procedures.

RESULTS

Overall, 12,027 patients were included. The primary outcome of in-hospital MACE occurred in 630 patients (5.2%). Compared with patients without a primary outcome, those who developed in-hospital MACE were older with more comorbidities, demonstrated poorer functional status, had more complex aneurysms, and were more likely to require concomitant procedures. Our best performing prediction model for in-hospital MACE was XGBoost, achieving an AUROC of 0.93 (95% confidence interval, 0.92-0.94). Comparatively, logistic regression had an AUROC of 0.71 (95% confidence interval, 0.70-0.73). For secondary outcomes, XGBoost achieved AUROCs between 0.84 and 0.94. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.05. These findings highlight the excellent predictive performance of the XGBoost model. The top three predictive features in our algorithm for in-hospital MACE following open AAA repair were: (1) coronary artery disease; (2) American Society of Anesthesiologists classification; and (3) proximal clamp site. Model performance remained robust on all subgroup analyses.

CONCLUSIONS

Open AAA repair outcomes can be accurately predicted using preoperative data with our ML models, which perform better than logistic regression. Our automated algorithms can help guide risk-mitigation strategies for patients being considered for open AAA repair to improve outcomes.

Machine learning to predict mortality for aneurysmal subarachnoid hemorrhage (aSAH) using a large nationwide EHR database

Aneurysmal subarachnoid hemorrhage (aSAH) develops quickly once it occurs and threatens the life of patients. We aimed to use machine learning to predict mortality for SAH patients at an early stage which can help doctors make clinical decisions. In our study, we applied different machine learning methods to an aSAH cohort extracted from a national EHR database, the Cerner Health Facts EHR database (2000-2018). The outcome of interest was in-hospital mortality, as either passing away while still in the hospital or being discharged to hospice care. Machine learning-based models were primarily evaluated by the area under the receiver operating characteristic curve (AUC). The population size of the SAH cohort was 6728. The machine learning methods achieved an average of AUCs of 0.805 for predicting mortality with only the initial 24 hours' EHR data. Without losing the prediction power, we used the logistic regression to identify 42 risk factors, -examples include age and serum glucose-that exhibit a significant correlation with the mortality of aSAH patients. Our study illustrates the potential of utilizing machine learning techniques as a practical prognostic tool for predicting aSAH mortality at the bedside.

Predicting outcomes following open revascularization for aortoiliac occlusive disease using machine learning

OBJECTIVE

Open surgical treatment options for aortoiliac occlusive disease carry significant perioperative risks; however, outcome prediction tools remain limited. Using machine learning (ML), we developed automated algorithms that predict 30-day outcomes following open aortoiliac revascularization.

METHODS

The National Surgical Quality Improvement Program (NSQIP) targeted vascular database was used to identify patients who underwent open aortoiliac revascularization for atherosclerotic disease between 2011 and 2021. Input features included 38 preoperative demographic/clinical variables. The primary outcome was 30-day major adverse limb event (MALE; composite of untreated loss of patency, major reintervention, or major amputation) or death. The 30-day secondary outcomes were individual components of the primary outcome, major adverse cardiovascular event (MACE; composite of myocardial infarction, stroke, or death), individual components of MACE, wound complication, bleeding, other morbidity, non-home discharge, and unplanned readmission. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. Variable importance scores were calculated to determine the top 10 predictive features. Performance was assessed on subgroups based on age, sex, race, ethnicity, symptom status, procedure type, and urgency.

RESULTS

Overall, 9649 patients were included. The primary outcome of 30-day MALE or death occurred in 1021 patients (10.6%). Our best performing prediction model for 30-day MALE or death was XGBoost, achieving an AUROC of 0.95 (95% confidence interval [CI], 0.94-0.96). In comparison, logistic regression had an AUROC of 0.79 (95% CI, 0.77-0.81). For 30-day secondary outcomes, XGBoost achieved AUROCs between 0.87 and 0.97 (untreated loss of patency [0.95], major reintervention [0.88], major amputation [0.96], death [0.97], MACE [0.95], myocardial infarction [0.88], stroke [0.93], wound complication [0.94], bleeding [0.87], other morbidity [0.96], non-home discharge [0.90], and unplanned readmission [0.91]). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.05. The strongest predictive feature in our algorithm was chronic limb-threatening ischemia. Model performance remained robust on all subgroup analyses of specific demographic/clinical populations.

CONCLUSIONS

Our ML models accurately predict 30-day outcomes following open aortoiliac revascularization using preoperative data, performing better than logistic regression. They have potential for important utility in guiding risk-mitigation strategies for patients being considered for open aortoiliac revascularization to improve outcomes.

Comparison of Prognostic Models for Functional Outcome in Aneurysmal Subarachnoid Hemorrhage Based on Machine Learning

BACKGROUND

Controversy exists regarding the superiority of the performance of prognostic tools based on advanced machine learning (ML) algorithms for patients with aneurysmal subarachnoid hemorrhage (aSAH). However, it is unclear whether ML prognostic models will benefit patients due to the lack of a comprehensive assessment. We aimed to develop and evaluate ML models for predicting unfavorable functional outcomes for aSAH patients and identify the model with the greatest performance.

METHODS

In this retrospective study, a dataset of 955 patients with aSAH was used to construct and validate prognostic models for functional outcomes assessed using the modified Rankin scale during a follow-up period of 3-6 months. Clinical scores and clinical and radiological features on admission and secondary complications were used to construct models based on 5 ML algorithms (i.e., logistic regression [LR], k-nearest neighbor, extreme gradient boosting, random forest, and artificial neural network). For evaluation among the models, the area under the receiver operating characteristic curve, area under the precision-recall curve, calibration curve, and decision curve analysis were used.

RESULTS

Composite models had significantly higher area under the receiver operating characteristic curves than did simple models in predicting unfavorable functional outcomes. Compared with other composite models (random forest and extreme gradient boosting) with good calibration, LR had the highest area under the precision-recall score and showed the greatest benefit in decision curve analysis.

CONCLUSIONS

Of the 5 studied ML models, the conventional LR model outperformed the advanced algorithms in predicting the prognosis and could be a useful tool for health care professionals.

Predicting the Outcome of Patients with Aneurysmal Subarachnoid Hemorrhage: A Machine-Learning-Guided Scorecard

Aneurysmal subarachnoid hemorrhage (aSAH) frequently causes long-term disability, but predicting outcomes remains challenging. Routine parameters such as demographics, admission status, CT findings, and blood tests can be used to predict aSAH outcomes. The aim of this study was to compare the performance of traditional logistic regression with several machine learning algorithms using readily available indicators and to generate a practical prognostic scorecard based on machine learning. Eighteen routinely available indicators were collected as outcome predictors for individuals with aSAH. Logistic regression (LR), random forest (RF), support vector machines (SVMs), and fully connected neural networks (FCNNs) were compared. A scorecard system was established based on predictor weights. The results show that machine learning models and a scorecard achieved 0.75~0.8 area under the curve (AUC) predicting aSAH outcomes (LR 0.739, RF 0.749, SVM 0.762~0.793, scorecard 0.794). FCNNs performed best (~0.95) but lacked interpretability. The scorecard model used only five factors, generating a clinically useful tool with a total cutoff score of ≥5, indicating poor prognosis. We developed and validated machine learning models proven to predict outcomes more accurately in individuals with aSAH. The parameters found to be the most strongly predictive of outcomes were NLR, lymphocyte count, monocyte count, hypertension status, and SEBES. The scorecard system provides a simplified means of applying predictive analytics at the bedside using a few key indicators.

Machine learning to predict outcomes following endovascular abdominal aortic aneurysm repair

BACKGROUND

Endovascular aneurysm repair (EVAR) for abdominal aortic aneurysm (AAA) carries important perioperative risks; however, there are no widely used outcome prediction tools. The aim of this study was to apply machine learning (ML) to develop automated algorithms that predict 1-year mortality following EVAR.

METHODS

The Vascular Quality Initiative database was used to identify patients who underwent elective EVAR for infrarenal AAA between 2003 and 2023. Input features included 47 preoperative demographic/clinical variables. The primary outcome was 1-year all-cause mortality. Data were split into training (70 per cent) and test (30 per cent) sets. Using 10-fold cross-validation, 6 ML models were trained using preoperative features with logistic regression as the baseline comparator. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score.

RESULTS

Some 63 655 patients were included. One-year mortality occurred in 3122 (4.9 per cent) patients. The best performing prediction model for 1-year mortality was XGBoost, achieving an AUROC (95 per cent c.i.) of 0.96 (0.95-0.97). Comparatively, logistic regression had an AUROC (95 per cent c.i.) of 0.69 (0.68-0.71). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.04. The top 3 predictive features in the algorithm were 1) unfit for open AAA repair, 2) functional status, and 3) preoperative dialysis.

CONCLUSIONS

In this data set, machine learning was able to predict 1-year mortality following EVAR using preoperative data and outperformed standard logistic regression models.

Unveiling Osteoporosis Through Radiomics Analysis of Hip CT Imaging

RATIONALE AND OBJECTIVES

This study aims to investigate the use of radiomics analysis of hip CT imaging to unveil osteoporosis.

MATERIALS AND METHODS

The researchers analyzed hip CT scans from a cohort of patients, including both osteoporotic and healthy individuals. Radiomics technique are employed to extract a comprehensive array of features from these images, encompassing texture, shape, and intensity alterations. Radiomics analysis using the 10 most commonly used machine learning models was employed to identify screened radiomics features for the detection of osteoporosis in patients. In addition to radiomics features, the basic information of patients is also utilized as training data for these machine learning models to accurately identify the presence of osteoporosis. A comparison would be made between the efficiency of recognizing radiomics features and the efficiency of recognizing patient basic information. The machine learning model that achieves the highest performance would be chosen to integrate patient basic information and radiomics features for the development of clinical nomograms.

RESULT

After a thorough screening process, 16 radiomics features were selected as input parameters for the machine learning model. In the test group, the highest accuracy achieved using radiomics features was 0.849, with an area under the curve (AUC) of 0.919. Evaluation of clinical features identified age and gender as closely associated with osteoporosis. Among these features, the KNN model exhibited the highest accuracy of 0.731 and an AUC of 0.658 in the test group. Comparing the performance of radiomics and clinical features, radiomics features demonstrated superior AUC values in the machine learning models. Ultimately, the XGBoost model, utilizing both radiomics and clinical features, was selected as the final Nomogram prediction model. In the test group, this model achieved an accuracy of 0.882 and an AUC of 0.886 in screening for osteoporosis.

CONCLUSION

Radiomics features derived from hip CT scans exhibit strong screening capabilities for osteoporosis. Furthermore, when combined with easily obtainable clinical features like patient age and gender, an effective screening efficacy for osteoporosis can be achieved.

Machine learning applications for the prediction of extended length of stay in geriatric hip fracture patients

BACKGROUND

Geriatric hip fractures are one of the most common fractures in elderly individuals, and prolonged hospital stays increase the risk of death and complications. Machine learning (ML) has become prevalent in clinical data processing and predictive models. This study aims to develop ML models for predicting extended length of stay (eLOS) among geriatric patients with hip fractures and to identify the associated risk factors.

AIM

To develop ML models for predicting the eLOS among geriatric patients with hip fractures, identify associated risk factors, and compare the performance of each model.

METHODS

A retrospective study was conducted at a single orthopaedic trauma centre, enrolling all patients who underwent hip fracture surgery between January 2018 and December 2022. The study collected various patient characteristics, encompassing demographic data, general health status, injury-related data, laboratory examinations, surgery-related data, and length of stay. Features that exhibited significant differences in univariate analysis were integrated into the ML model establishment and subsequently cross-verified. The study compared the performance of the ML models and determined the risk factors for eLOS.

RESULTS

The study included 763 patients, with 380 experiencing eLOS. Among the models, the decision tree, random forest, and extreme Gradient Boosting models demonstrated the most robust performance. Notably, the artificial neural network model also exhibited impressive results. After cross-validation, the support vector machine and logistic regression models demonstrated superior performance. Predictors for eLOS included delayed surgery, D-dimer level, American Society of Anaesthesiologists (ASA) classification, type of surgery, and sex.

CONCLUSION

ML proved to be highly accurate in predicting the eLOS for geriatric patients with hip fractures. The identified key risk factors were delayed surgery, D-dimer level, ASA classification, type of surgery, and sex. This valuable information can aid clinicians in allocating resources more efficiently to meet patient demand effectively.

Predicting Major Adverse Cardiovascular Events Following Carotid Endarterectomy Using Machine Learning

Background Carotid endarterectomy (CEA) is a major vascular operation for stroke prevention that carries significant perioperative risks; however, outcome prediction tools remain limited. The authors developed machine learning algorithms to predict outcomes following CEA. Methods and Results The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent CEA between 2011 and 2021. Input features included 36 preoperative demographic/clinical variables. The primary outcome was 30-day major adverse cardiovascular events (composite of stroke, myocardial infarction, or death). The data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 machine learning models were trained using preoperative features. The primary metric for evaluating model performance was area under the receiver operating characteristic curve. Model robustness was evaluated with calibration plot and Brier score. Overall, 38 853 patients underwent CEA during the study period. Thirty-day major adverse cardiovascular events occurred in 1683 (4.3%) patients. The best performing prediction model was XGBoost, achieving an area under the receiver operating characteristic curve of 0.91 (95% CI, 0.90-0.92). In comparison, logistic regression had an area under the receiver operating characteristic curve of 0.62 (95% CI, 0.60-0.64), and existing tools in the literature demonstrate area under the receiver operating characteristic curve values ranging from 0.58 to 0.74. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.02. The strongest predictive feature in our algorithm was carotid symptom status. Conclusions The machine learning models accurately predicted 30-day outcomes following CEA using preoperative data and performed better than existing tools. They have potential for important utility in guiding risk-mitigation strategies to improve outcomes for patients being considered for CEA.

Using machine learning to predict outcomes following carotid endarterectomy

OBJECTIVE

Prediction of outcomes following carotid endarterectomy (CEA) remains challenging, with a lack of standardized tools to guide perioperative management. We used machine learning (ML) to develop automated algorithms that predict outcomes following CEA.

METHODS

The Vascular Quality Initiative (VQI) database was used to identify patients who underwent CEA between 2003 and 2022. We identified 71 potential predictor variables (features) from the index hospitalization (43 preoperative [demographic/clinical], 21 intraoperative [procedural], and 7 postoperative [in-hospital complications]). The primary outcome was stroke or death at 1 year following CEA. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). After selecting the best performing algorithm, additional models were built using intra- and postoperative data. Model robustness was evaluated using calibration plots and Brier scores. Performance was assessed on subgroups based on age, sex, race, ethnicity, insurance status, symptom status, and urgency of surgery.

RESULTS

Overall, 166,369 patients underwent CEA during the study period. In total, 7749 patients (4.7%) had the primary outcome of stroke or death at 1 year. Patients with an outcome were older with more comorbidities, had poorer functional status, and demonstrated higher risk anatomic features. They were also more likely to undergo intraoperative surgical re-exploration and have in-hospital complications. Our best performing prediction model at the preoperative stage was XGBoost, achieving an AUROC of 0.90 (95% confidence interval [CI], 0.89-0.91). In comparison, logistic regression had an AUROC of 0.65 (95% CI, 0.63-0.67), and existing tools in the literature demonstrate AUROCs ranging from 0.58 to 0.74. Our XGBoost models maintained excellent performance at the intra- and postoperative stages, with AUROCs of 0.90 (95% CI, 0.89-0.91) and 0.94 (95% CI, 0.93-0.95), respectively. Calibration plots showed good agreement between predicted and observed event probabilities with Brier scores of 0.15 (preoperative), 0.14 (intraoperative), and 0.11 (postoperative). Of the top 10 predictors, eight were preoperative features, including comorbidities, functional status, and previous procedures. Model performance remained robust on all subgroup analyses.

CONCLUSIONS

We developed ML models that accurately predict outcomes following CEA. Our algorithms perform better than logistic regression and existing tools, and therefore, have potential for important utility in guiding perioperative risk mitigation strategies to prevent adverse outcomes.

Field Measures Are All You Need: Predicting Need for Surgery in Elderly Ground-Level Fall Patients via Machine Learning

BACKGROUND

As ground-level falls (GLFs) are a significant cause of mortality in elderly patients, field triage plays an essential role in patient outcomes. This research investigates how machine learning algorithms can supplement traditional t-tests to recognize statistically significant patterns in medical data and to aid clinical guidelines.

METHODS

This is a retrospective study using data from 715 GLF patients over 75 years old. We first calculated P-values for each recorded factor to determine the factor's significance in contributing to a need for surgery (P < .05 is significant). We then utilized the XGBoost machine learning method to rank contributing factors. We applied SHapley Additive exPlanations (SHAP) values to interpret the feature importance and provide clinical guidance via decision trees.

RESULTS

The three most significant P-values when comparing patients with and without surgery are as follows: Glasgow Coma Scale (GCS) (P < .001), no comorbidities (P < .001), and transfer-in (P = .019). The XGBoost algorithm determined that GCS and systolic blood pressure contribute most strongly. The prediction accuracy of these XGBoost results based on the test/train split was 90.3%.

DISCUSSION

When compared to P-values, XGBoost provides more robust, detailed results regarding the factors that suggest a need for surgery. This demonstrates the clinical applicability of machine learning algorithms. Paramedics can use resulting decision trees to inform medical decision-making in real time. XGBoost's generalizability power increases with more data and can be tuned to prospectively assist individual hospitals.

Predicting sepsis onset in ICU using machine learning models: a systematic review and meta-analysis

BACKGROUND

Sepsis is a life-threatening condition caused by an abnormal response of the body to infection and imposes a significant health and economic burden worldwide due to its high mortality rate. Early recognition of sepsis is crucial for effective treatment. This study aimed to systematically evaluate the performance of various machine learning models in predicting the onset of sepsis.

METHODS

We conducted a comprehensive search of the Cochrane Library, PubMed, Embase, and Web of Science databases, covering studies from database inception to November 14, 2022. We used the PROBAST tool to assess the risk of bias. We calculated the predictive performance for sepsis onset using the C-index and accuracy. We followed the PRISMA guidelines for this study.

RESULTS

We included 23 eligible studies with a total of 4,314,145 patients and 26 different machine learning models. The most frequently used models in the studies were random forest (n = 9), extreme gradient boost (n = 7), and logistic regression (n = 6) models. The random forest (test set n = 9, acc = 0.911) and extreme gradient boost (test set n = 7, acc = 0.957) models were the most accurate based on our analysis of the predictive performance. In terms of the C-index outcome, the random forest (n = 6, acc = 0.79) and extreme gradient boost (n = 7, acc = 0.83) models showed the highest performance.

CONCLUSION

Machine learning has proven to be an effective tool for predicting sepsis at an early stage. However, to obtain more accurate results, additional machine learning methods are needed. In our research, we discovered that the XGBoost and random forest models exhibited the best predictive performance and were most frequently utilized for predicting the onset of sepsis.

TRIAL REGISTRATION

CRD42022384015.

Artificial Intelligence and Neurosurgery: Tracking Antiplatelet Response Patterns for Endovascular Intervention

Platelets play a critical role in blood clotting and the development of arterial blockages. Antiplatelet therapy is vital for preventing recurring events in conditions like coronary artery disease and strokes. However, there is a lack of comprehensive guidelines for using antiplatelet agents in elective neurosurgery. Continuing therapy during surgery poses a bleeding risk, while discontinuing it before surgery increases the risk of thrombosis. Discontinuation is recommended in neurosurgical settings but carries an elevated risk of ischemic events. Conversely, maintaining antithrombotic therapy may increase bleeding and the need for transfusions, leading to a poor prognosis. Artificial intelligence (AI) holds promise in making difficult decisions regarding antiplatelet therapy. This paper discusses current clinical guidelines and supported regimens for antiplatelet therapy in neurosurgery. It also explores methodologies like P2Y12 reaction units (PRU) monitoring and thromboelastography (TEG) mapping for monitoring the use of antiplatelet regimens as well as their limitations. The paper explores the potential of AI to overcome such limitations associated with PRU monitoring and TEG mapping. It highlights various studies in the field of cardiovascular and neuroendovascular surgery which use AI prediction models to forecast adverse outcomes such as ischemia and bleeding, offering assistance in decision-making for antiplatelet therapy. In addition, the use of AI to improve patient adherence to antiplatelet regimens is also considered. Overall, this research aims to provide insights into the use of antiplatelet therapy and the role of AI in optimizing treatment plans in neurosurgical settings.

Comparison of intracranial pressure prediction in hydrocephalus patients among linear, non-linear, and machine learning regression models in Thailand

BACKGROUND

Hydrocephalus (HCP) is one of the most significant concerns in neurosurgical patients because it can cause increased intracranial pressure (ICP), resulting in mortality and morbidity. To date, machine learning (ML) has been helpful in predicting continuous outcomes. The primary objective of the present study was to identify the factors correlated with ICP, while the secondary objective was to compare the predictive performances among linear, non-linear, and ML regression models for ICP prediction.

METHODS

A total of 412 patients with various types of HCP who had undergone ventriculostomy was retrospectively included in the present study, and intraoperative ICP was recorded following ventricular catheter insertion. Several clinical factors and imaging parameters were analyzed for the relationship with ICP by linear correlation. The predictive performance of ICP was compared among linear, non-linear, and ML regression models.

RESULTS

Optic nerve sheath diameter (ONSD) had a moderately positive correlation with ICP (r=0.530, P<0.001), while several ventricular indexes were not statistically significant in correlation with ICP. For prediction of ICP, random forest (RF) and extreme gradient boosting (XGBoost) algorithms had low mean absolute error and root mean square error values and high R2 values compared to linear and non-linear regression when the predictive model included ONSD and ventricular indexes.

CONCLUSIONS

The XGBoost and RF algorithms are advantageous for predicting preoperative ICP and establishing prognoses for HCP patients. Furthermore, ML-based prediction could be used as a non-invasive method.

Assessing Metabolic Markers in Glioblastoma Using Machine Learning: A Systematic Review

Glioblastoma (GBM) is a common and deadly brain tumor with late diagnoses and poor prognoses. Machine learning (ML) is an emerging tool that can create highly accurate diagnostic and prognostic prediction models. This paper aimed to systematically search the literature on ML for GBM metabolism and assess recent advancements. A literature search was performed using predetermined search terms. Articles describing the use of an ML algorithm for GBM metabolism were included. Ten studies met the inclusion criteria for analysis: diagnostic (n = 3, 30%), prognostic (n = 6, 60%), or both (n = 1, 10%). Most studies analyzed data from multiple databases, while 50% (n = 5) included additional original samples. At least 2536 data samples were run through an ML algorithm. Twenty-seven ML algorithms were recorded with a mean of 2.8 algorithms per study. Algorithms were supervised (n = 24, 89%), unsupervised (n = 3, 11%), continuous (n = 19, 70%), or categorical (n = 8, 30%). The mean reported accuracy and AUC of ROC were 95.63% and 0.779, respectively. One hundred six metabolic markers were identified, but only EMP3 was reported in multiple studies. Many studies have identified potential biomarkers for GBM diagnosis and prognostication. These algorithms show promise; however, a consensus on even a handful of biomarkers has not yet been made.