Related Experiment Video
Updated: Jan 13, 2026
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
An interpretable machine learning tool for predicting perioperative cardiac events in patients scheduled for hip
Danila Azzolina1, Gianmaria Cammarota2,3, Enrico Boero4
1Biostatistics and Clinical Trial Methodology Unit, Clinical Research Center DEMeTra, Department of Translational Medicine, University of Naples Federico II, Naples, Italy.
Insights
This study developed a machine learning tool to predict Major Adverse Cardiac Events (MACE) in elderly hip fracture patients. Integrating lung ultrasound improved risk prediction, aiding targeted preventive strategies.
Area of Science:
- Cardiology
- Machine Learning
- Geriatric Surgery
Background:
- Elderly patients undergoing hip fracture surgery face high risks of perioperative Major Adverse Cardiac Events (MACE).
- MACE significantly impacts postoperative outcomes in this vulnerable population.
- Existing risk assessment tools may not fully capture MACE risk in hip fracture patients.
Purpose of the Study:
- To develop an interpretable machine learning (ML) model for predicting MACE in elderly hip fracture patients.
- To integrate clinical and novel ultrasound-based variables for enhanced risk prediction.
- To create a tool for personalized, real-time risk estimation.
Main Methods:
- Analysis of 877 patients from the multicenter LUSHIP study.
- Inclusion of demographics, Revised Cardiac Risk Index (RCRI), functional status, and preoperative lung ultrasound (LUS) scores.
- Development of an ensemble meta-model combining Gradient Boosting Machine (GBM) and Elastic-Net Regularized Generalized Linear Models (GLMNET), validated via bootstrap resampling.
Main Results:
- The ensemble ML model achieved an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.86, with 72% sensitivity and 83% specificity.
- Integration of LUS scores significantly improved risk prediction (AUC=0.78) compared to traditional tools like RCRI alone.
- Key predictors identified were LUS score, RCRI score, and patient age; a web-based application was developed for risk estimation.
Conclusions:
- The interpretable ML model offers improved perioperative cardiac risk stratification for elderly hip fracture patients.
- The model's ability to integrate LUS data provides a non-invasive, bedside biomarker for risk assessment.
- This tool can potentially guide targeted preventive strategies and optimize resource allocation in surgical care.
Background:
Elderly patients undergoing surgery for hip fractures are at high risk for perioperative Major Adverse Cardiac Events (MACE), which can markedly compromise postoperative outcomes. This study aims to develop a machine learning (ML) based, interpretable tool to predict MACE using clinical and ultrasound-based variables in this population.
Methods:
We analyzed data from 877 patients in the multicenter LUSHIP study, incorporating demographics, Revised Cardiac Risk Index (RCRI), functional status, and preoperative lung ultrasound (LUS) scores. Multiple ML models were trained and validated using bootstrap resampling. The final ensemble meta-model combined GBM (Gradient Boosting Machine) and GLMNET (Elastic-Net Regularized Generalized Linear Models).
Results:
The ensemble model achieved an AUROC of 0.86, with sensitivity and specificity of 0.72 and 0.83, respectively. These results significantly improve over traditional tools such as the Revised Cardiac Risk Index (RCRI), particularly when used alone. A significant contribution of this work is the integration of lung ultrasound (LUS) as a non-invasive, bedside biomarker, which notably improved risk prediction compared to the performance of the individual LUS marker alone (AUC = 0.78). Relevant predictors for the ML model are LUS score, RCRI score, and patient age. A web-based Shiny application was developed to enable real-time personalized risk estimation.
Conclusion:
This interpretable ML model improves perioperative cardiac risk stratification and profiling in elderly hip fracture patients and may guide targeted preventive strategies and resource allocation.
Trial Registration:
CT04074876.
