⚡ Quick Summary
This systematic review evaluated the use of machine learning (ML) models based on coronary computed tomography angiography (CCTA) to predict major adverse cardiovascular events (MACEs). The findings revealed that ML models, particularly those using logistic regression, demonstrated significant potential in enhancing diagnostic accuracy for high-risk patients.
🔍 Key Details
- 📊 Dataset: Ten studies included in the analysis
- 🧩 Features used: Radiomic features extracted from CCTA
- ⚙️ Technology: Various ML algorithms, including logistic regression and random forest
- 🏆 Performance: Pooled AUROC of 0.7879 (training) and 0.7981 (testing)
🔑 Key Takeaways
- 📊 CCTA is a first-line noninvasive imaging test for coronary artery disease.
- 💡 Machine learning enhances the diagnostic performance of CCTA in predicting MACEs.
- 👩🔬 Logistic regression achieved an AUROC of 0.8229 in the testing group.
- 🏆 Random forest models reached an AUROC of 0.8444 in the training group.
- 🔍 High heterogeneity was observed among the studies included in the review.
- 🌍 The study followed TRIPOD guidelines for transparent reporting of prediction models.
- 🆔 Clinical trials are needed for further validation of these ML models.
📚 Background
Coronary artery disease (CAD) is a leading cause of morbidity and mortality worldwide. The advent of coronary computed tomography angiography (CCTA) has revolutionized the noninvasive assessment of CAD, allowing for early detection and intervention. However, the integration of machine learning (ML) into this diagnostic process has the potential to further enhance predictive accuracy, particularly for major adverse cardiovascular events (MACEs).
🗒️ Study
This systematic review aimed to evaluate the diagnostic value of ML models constructed using radiomic features from CCTA in predicting MACEs. The authors conducted a comprehensive search across five online databases, assessing the methodological quality of the included studies and ensuring adherence to the TRIPOD guidelines for transparent reporting.
📈 Results
The analysis included ten studies, with a notable distinction between training and testing groups. The pooled area under the receiver operating characteristic (AUROC) curve for ML models predicting MACEs was found to be 0.7879 in the training set and 0.7981 in the testing set. Logistic regression models outperformed others, achieving an AUROC of 0.8229 in the testing group.
🌍 Impact and Implications
The findings of this review underscore the significant potential of ML models in improving the diagnostic accuracy of CCTA for predicting MACEs. By leveraging radiomic features, healthcare professionals can better identify high-risk patients, leading to timely interventions and improved patient outcomes. This integration of technology into clinical practice could pave the way for more personalized and effective cardiovascular care.
🔮 Conclusion
This systematic review highlights the promising role of machine learning in enhancing the predictive capabilities of CCTA for MACEs. The superior performance of logistic regression models, particularly when combined with clinical features, suggests a need for further clinical trials to validate these findings. The future of cardiovascular diagnostics looks bright with the continued integration of advanced technologies!
💬 Your comments
What are your thoughts on the integration of machine learning in cardiovascular diagnostics? We would love to hear your insights! 💬 Leave your comments below or connect with us on social media:
The Machine Learning Models in Major Cardiovascular Adverse Events Prediction Based on Coronary Computed Tomography Angiography: Systematic Review.
Abstract
BACKGROUND: Coronary computed tomography angiography (CCTA) has emerged as the first-line noninvasive imaging test for patients at high risk of coronary artery disease (CAD). When combined with machine learning (ML), it provides more valid evidence in diagnosing major adverse cardiovascular events (MACEs). Radiomics provides informative multidimensional features that can help identify high-risk populations and can improve the diagnostic performance of CCTA. However, its role in predicting MACEs remains highly debated.
OBJECTIVE: We evaluated the diagnostic value of ML models constructed using radiomic features extracted from CCTA in predicting MACEs, and compared the performance of different learning algorithms and models, thereby providing clinical recommendations for the diagnosis, treatment, and prognosis of MACEs.
METHODS: We comprehensively searched 5 online databases, Cochrane Library, Web of Science, Elsevier, CNKI, and PubMed, up to September 10, 2024, for original studies that used ML models among patients who underwent CCTA to predict MACEs and reported clinical outcomes and endpoints related to it. Risk of bias in the ML models was assessed by the Prediction Model Risk of Bias Assessment Tool, while the radiomics quality score (RQS) was used to evaluate the methodological quality of the radiomics prediction model development and validation. We also followed the TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) guidelines to ensure transparency of ML models included. Meta-analysis was performed using Meta-DiSc software (version 1.4), which included the I² score and Cochran Q test, along with StataMP 17 (StataCorp) to assess heterogeneity and publication bias. Due to the high heterogeneity observed, subgroup analysis was conducted based on different model groups.
RESULTS: Ten studies were included in the analysis, 5 (50%) of which differentiated between training and testing groups, where the training set collected 17 kinds of models and the testing set gathered 26 models. The pooled area under the receiver operating characteristic (AUROC) curve for ML models predicting MACEs was 0.7879 in the training set and 0.7981 in the testing set. Logistic regression (LR), the most commonly used algorithm, achieved an AUROC of 0.8229 in the testing group and 0.7983 in the training group. Non-LR models yielded AUROCs of 0.7390 in the testing set and 0.7648 in the training set, while the random forest (RF) models reached an AUROC of 0.8444 in the training group.
CONCLUSIONS: Study limitations included a limited number of studies, high heterogeneity, and the types of included studies. The performance of ML models for predicting MACEs was found to be superior to that of general models based on basic feature extraction and integration from CCTA. Specifically, LR-based ML diagnostic models demonstrated significant clinical potential, particularly when combined with clinical features, and are worth further validation through more clinical trials.
TRIAL REGISTRATION: PROSPERO CRD42024596364; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024596364.
Author: [‘Ma Y’, ‘Li M’, ‘Wu H’]
Journal: J Med Internet Res
Citation: Ma Y, et al. The Machine Learning Models in Major Cardiovascular Adverse Events Prediction Based on Coronary Computed Tomography Angiography: Systematic Review. The Machine Learning Models in Major Cardiovascular Adverse Events Prediction Based on Coronary Computed Tomography Angiography: Systematic Review. 2025; 27:e68872. doi: 10.2196/68872