Related Experiment Video
Updated: Jun 12, 2025
Murine Surgical Model of Topical Elastase Induced Descending Thoracic Aortic Aneurysm
Published on: August 24, 2019
Enhanced machine learning models for predicting one-year mortality in individuals suffering from type A aortic
Jing Zhang1, Wuyu Xiong2, Jiajuan Yang3
1Department of Cardiology, The First College of Clinical Medical Science, China Three Gorges University & Yichang Central People's Hospital, Yichang, China; Central Laboratory, The First College of Clinical Medical Science, China Three Gorges University & Yichang Central People's Hospital, Yichang, China; Hubei Key Laboratory of Ischemic Cardiovascular Disease, Yichang, China; Hubei Provincial Clinical Research Center for Ischemic Cardiovascular Disease, Yichang, China.
A machine learning model predicts 1-year mortality in patients with type A aortic dissection. Key factors like age and white blood cell count influence outcomes, while interventions improve survival.
Area of Science:
- Cardiovascular Medicine
- Medical Informatics
- Machine Learning in Healthcare
Background:
- Type A aortic dissection (TAAD) is a life-threatening condition with significant mortality.
- Accurate risk stratification is crucial for guiding clinical decisions and improving patient outcomes in TAAD.
- Existing prognostic models may not fully capture the complexity of TAAD mortality predictors.
Purpose of the Study:
- To develop and validate an interpretable machine learning model for predicting 1-year mortality in TAAD patients.
- To identify key clinical and laboratory factors associated with 1-year mortality in TAAD.
- To create a user-friendly tool for clinical decision support in TAAD management.
Main Methods:
- A cohort of 289 TAAD patients was divided into training (202) and validation (87) sets.
- The Least Absolute Shrinkage and Selection Operator (LASSO) method identified 8 key predictors.
- A Treebag model was developed and validated using accuracy, F1-Score, Brier score, AUC, and AUPRC; Shapley Additive Explanations (SHAP) assessed predictor importance.
Main Results:
- The Treebag model demonstrated high predictive performance with a Brier score of 0.128 and AUC of 0.91.
- Older age and elevated white blood cell count were risk factors; higher systolic blood pressure, lymphocyte count, carbon dioxide combining power, eosinophil count, beta-blocker use, and surgical intervention were protective.
- A web-based application, TAAD One-Year Prognostic Risk Assessment Web, was developed for clinical use.
Conclusions:
- The developed Treebag model accurately predicts 1-year mortality in TAAD patients.
- Key factors influencing survival include surgical intervention, beta-blocker use, and management of specific laboratory and physiological parameters.
- This interpretable model and associated application offer a valuable tool for enhancing patient management and outcomes in TAAD.
