Online interpretable dynamic prediction models for clinically significant posthepatectomy liver failure based on machine learning algorithms: A retrospective cohort study.

BACKGROUND: Posthepatectomy liver failure (PHLF) is the leading cause of mortality in patients undergoing hepatectomy. However, practical models for accurately predicting the risk of PHLF are lacking. This study aimed to develop precise prediction models for clinically significant PHLF. METHODS: A total of 226 patients undergoing hepatectomy at a single center were recruited. The study outcome was clinically significant PHLF. Five pre- and postoperative machine learning (ML) models were developed and compared with four clinical scores, namely, the MELD, FIB-4, ALBI, and APRI scores. The robustness of the developed ML models was internally validated using 5-fold cross-validation by calculating the average of the evaluation metrics and was externally validated on an independent temporal dataset, including the area under the curve (AUC) and the area under the precision‒recall curve (AUPRC). SHapley Additive exPlanations analysis was performed to interpret the best performance model. RESULTS: Clinically significant PHLF was observed in 23 of 226 patients (10.2%). The variables in the preoperative model included creatinine, total bilirubin, and Child‒Pugh grade. In addition to the above factors, the extent of resection was also a key variable for the postoperative model. The pre- and postoperative artificial neural network (ANN) models exhibited excellent performance, with mean AUCs of 0.766 and 0.851, respectively, and mean AUPRC values of 0.441 and 0.645, whereas the MELD, FIB-4, ALBI, and APRI scores reached AUCs of 0.714, 0.498, 0.536 and 0.551, respectively, and AUPRC values of 0.204, 0.111, 0.128 and 0.163, respectively. In addition, the AUCs of the pre- and postoperative ANN models were 0.720 and 0.731, respectively, and the AUPRC values were 0.380 and 0.408, respectively, on the temporal dataset. CONCLUSION: Our online interpretable dynamic ML models outperformed common clinical scores and could function as a clinical decision support tool to identify patients at high risk of PHLF pre- and postoperatively.

A machine learning stacking model accurately estimating gastric fluid volume in patients undergoing elective sedated gastrointestinal endoscopy.

BACKGROUND: The current point-of-care ultrasound (POCUS) assessment of gastric fluid volume primarily relies on the traditional linear approach, which often suffers from moderate accuracy. This study aimed to develop an advanced machine learning (ML) model to estimate gastric fluid volume more accurately. METHODS: We retrospectively analyzed the clinical data and POCUS data (D1: craniocaudal diameter, D2: anteroposterior diameter) of 1386 patients undergoing elective sedated gastrointestinal endoscopy (GIE) at Nanjing First Hospital to predict gastric fluid volume using ML techniques, including six different ML models and a stacking model. We evaluated the models using the adjusted Coefficient of Determination (R2), mean absolute error (MAE) and root mean square error (RMSE). The SHapley Additive exPlanations (SHAP) method was used to interpret the importance of the variables. Finally, a web calculator was constructed to facilitate its clinical application. RESULTS: The stacking model (Linear regression + Multilayer perceptron) performed best, with the highest adjusted R2 of 0.718 (0.632 to 0.804). The mean prediction bias was 4 ml (MAE: 4.008 (3.68 to 4.336)), which is better than that of the linear model. D1 and D2 ranked high in the SHAP plot and performed better in the right lateral decubitus (RLD) than in the supine position. The web calculator can be accessed at https://cheason.shinyapps.io/Stacking_regressor/. CONCLUSION: The stacking model and its web calculator can serve as practical tools for accurately estimating gastric fluid volume in patients undergoing elective sedated GIE. It is recommended that anesthesiologists measure D1 and D2 in the patient's RLD position.

Accurately predicting the risk of unfavorable outcomes after endovascular coil therapy in patients with aneurysmal subarachnoid hemorrhage: an interpretable machine learning model.

BACKGROUND: Despite endovascular coiling as a valid modality in treatment of aneurysmal subarachnoid hemorrhage (aSAH), there is a risk of poor prognosis. However, the clinical utility of previously proposed early prediction tools remains limited. We aimed to develop a clinically generalizable machine learning (ML) models for accurately predicting unfavorable outcomes in aSAH patients after endovascular coiling. METHODS: Functional outcomes at 6 months after endovascular coiling were assessed via the modified Rankin Scale (mRS) and unfavorable outcomes were defined as mRS 3-6. Five ML algorithms (logistic regression, random forest, support vector machine, deep neural network, and extreme gradient boosting) were used for model development. The area under precision-recall curve (AUPRC) and receiver operating characteristic curve (AUROC) was used as main indices of model evaluation. SHapley Additive exPlanations (SHAP) method was applied to interpret the best-performing ML model. RESULTS: A total of 371 patients were eventually included into this study, and 85.4% of them had favorable outcomes. Among the five models, the DNN model had a better performance with AUPRC of 0.645 (AUROC of 0.905). Postoperative GCS score, size of aneurysm, and age were the top three powerful predictors. The further analysis of five random cases presented the good interpretability of the DNN model. CONCLUSION: Interpretable clinical prediction models based on different ML algorithms have been successfully constructed and validated, which would serve as reliable tools in optimizing the treatment decision-making of aSAH. Our DNN model had better performance to predict the unfavorable outcomes at 6 months in aSAH patients compared with Yan's nomogram model.

Development and validation of a novel nomogram model to assess the risk of gastric contents in outpatients undergoing elective sedative gastrointestinal endoscopy procedures.

BACKGROUND: Gastric contents may contribute to patients' aspiration during anesthesia. Ultrasound can accurately assess the risk of gastric contents in patients undergoing sedative gastrointestinal endoscopy (GIE) procedures, but its efficiency is limited. Therefore, developing an accurate and efficient model to predict gastric contents in outpatients undergoing elective sedative GIE procedures is greatly desirable. METHODS: This study retrospectively analyzed 1501 patients undergoing sedative GIE procedures. Gastric contents were observed under direct gastroscopic vision and suctioned through the endoscope. High-risk gastric contents were defined as having solid content or liquid volume > 25 ml and pH < 2.5; otherwise, they were considered low-risk gastric contents. Univariate analysis and multivariate analysis were used to select the independent risk factors to predict high-risk gastric contents. Based on the selected independent risk factors, we assigned values to each independent risk factor and established a novel nomogram. The performance of the nomogram was verified in the testing cohort by the metrics of discrimination, calibration, and clinical usefulness. In addition, an online accessible web calculator was constructed. RESULTS: We found BMI, cerebral infarction, cirrhosis, male, age, diabetes, and gastroesophageal reflux disease were risk factors for gastric contents. The AUROCs were 0.911 and 0.864 in the development and testing cohort, respectively. Moreover, the nomogram showed good calibration ability. Decision curve analysis and Clinical impact curve demonstrated that the predictive nomogram was clinically useful. The website of the nomogram was https://medication.shinyapps.io/dynnomapp/. CONCLUSIONS: This study demonstrates that clinical variables can be combined with algorithmic techniques to predict gastric contents in outpatients. Nomogram was constructed from routine variables, and the web calculator had excellent clinical applicability to assess the risk of gastric contents accurately and efficiently in outpatients, assist anesthesiologists in assessment and identify the most appropriate patients for ultrasound.

A convenient machine learning model to predict full stomach and evaluate the safety and comfort improvements of preoperative oral carbohydrate in patients undergoing elective painless gastrointestinal endoscopy.

BACKGROUND AND AIMS: Assessment of the patient's gastric contents is the key to avoiding aspiration incidents, however, there is no effective method to determine whether elective painless gastrointestinal endoscopy (GIE) patients have a full stomach or an empty stomach. And previous studies have shown that preoperative oral carbohydrates (POCs) can improve the discomfort induced by fasting, but there are different perspectives on their safety. This study aimed to develop a convenient, accurate machine learning (ML) model to predict full stomach. And based on the model outcomes, evaluate the safety and comfort improvements of POCs in empty- and full stomach groups. METHODS: We enrolled 1386 painless GIE patients between October 2022 and January 2023 in Nanjing First Hospital, and 1090 patients without POCs were used to construct five different ML models to identify full stomach. The metrics of discrimination and calibration validated the robustness of the models. For the best-performance model, we further interpreted it through SHapley Additive exPlanations (SHAP) and constructed a web calculator to facilitate clinical use. We evaluated the safety and comfort improvements of POCs by propensity score matching (PSM) in the two groups, respectively. RESULTS: Random Forest (RF) model showed the greatest discrimination with the area under the receiver operating characteristic curve (AUROC) 0.837 [95% confidence interval (CI): 79.1-88.2], F1 71.5%, and best calibration with a Brier score of 15.2%. The web calculator can be visited at https://medication.shinyapps.io/RF_model/. PSM results demonstrated that POCs significantly reduced the full stomach incident in empty stomach group (p < 0.05), but no differences in full stomach group (p > 0.05). Comfort improved in both groups and was more significant in empty stomach group. CONCLUSIONS: The developed convenient RF model predicted full stomach with high accuracy and interpretability. POCs were safe and comfortably improved in both groups, with more benefit in empty stomach group. These findings may guide the patients' gastrointestinal preparation.

This study is the first model utilizing advanced ML techniques based on multiple clinical variables to identify full stomach. The model is suitable for patient-rich outpatient clinics, primary hospitals, remote regions, and specific clinical settings where POCUS is not available.The developed convenient RF model predicted full stomach with high accuracy and interpretability. The test cohort AUROC was 0.837. We further established an online accessible individualized risk calculator and provided waterfall plots to increase the interpretability of each prediction.The propensity score matching (PSM) showed that preoperative oral carbohydrate (POCs) were safe and comfortably improved in both groups, with more benefit in empty stomach group. These findings may provide information for anesthesiologists to guide patients on POCs.