What question did this study set out to answer?

The aim is to develop and validate machine learning models that predict the risk of chemotherapy-induced cardiotoxicity in breast cancer patients.

January 10, 2026Open Access

Machine learning models using multimodal data accurately predict chemotherapy-induced cardiotoxicity in breast cancer

Key Result

The XGBoost model accurately predicted chemotherapy-induced cardiotoxicity with an AUC of 0.782 in breast cancer patients, identifying key risk factors like age and ECG abnormalities.

Key Points

The aim is to develop and validate machine learning models that predict the risk of chemotherapy-induced cardiotoxicity in breast cancer patients.
Retrospective analysis of data from 423 female breast cancer patients who received chemotherapy.
Data types included demographics, clinical variables, echocardiographic parameters, ECG findings, and cardiac biomarkers.
Data was split into training and validation sets at a 7:3 ratio.
Used seven feature selection methods and eight machine learning algorithms to create models.
Chemotherapy-induced cardiotoxicity occurred in 111 patients (26.24%).
Identified five robust predictors: age, baseline left ventricular ejection fraction <60%, anthracycline-trastuzumab treatment, chemotherapy cycles, and abnormal ECG findings.
The XGBoost algorithm achieved an area under the curve of 0.782, indicating strong predictive performance.

Structured PICO

Does a machine learning model using multimodal data accurately predict chemotherapy-induced cardiotoxicity in female breast cancer patients?

Population

423 female breast cancer patients (age 20-70 years) with no prior history of cardiovascular disease who received chemotherapy

Intervention

Extreme gradient boosting (XGBoost) machine learning model integrating multimodal data (demographics, clinical variables, echocardiography, ECG, and biomarkers)

Comparator

Other machine learning algorithms (SVM, LR, RF, KNN, NB, DT, LGBM)

Outcome

Chemotherapy-related cardiac dysfunction (CTRCD), defined as LVEF decline ≥10% from baseline to <53%, or new/worsening ECG abnormalities, or elevated cardiac biomarkers (hsTnT >14 ng/L or NT-proBNP >125 pg/mL)composite

An XGBoost machine learning model integrating clinical, imaging, and biomarker data can accurately predict the risk of chemotherapy-related cardiac dysfunction in breast cancer patients.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Background Despite significant advances in breast cancer therapy, chemotherapy-related cardiac dysfunction (CTRCD) remains a critical clinical challenge. This study aimed to develop and validate machine learning (ML) models that integrate multimodal data to predict the risk of CTRCD in female breast cancer patients. Methods We retrospectively analyzed data from 423 female breast cancer patients who received chemotherapy between January 2020 and January 2025. Multimodal data included demographic information, clinical variables, echocardiographic parameters, electrocardiographic (ECG) findings, and cardiac biomarkers. The dataset was randomly split into training and validation sets in a 7:3 ratio. Seven feature selection methods and eight ML algorithms were employed to construct and compare predictive models. Results Among the 423 patients, CTRCD occurred in 111 patients (26.24%). Five variables were identified as robust predictors: age, baseline left ventricular ejection fraction 60%, anthracycline–trastuzumab combination therapy, chemotherapy cycles, and abnormal ECG findings. Among all models evaluated, the extreme gradient boosting (XGBoost) algorithm demonstrated the best performance, achieving an area under the curve of 0.782 (95% CI: 0.681–0.883) in 10-fold cross-validation. Conclusion The XGBoost-based model showed strong predictive ability and may serve as a practical tool for early risk stratification and timely clinical management of CTRCD.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper