Cross-Validated Machine Learning Stacking Ensemble Regression for CCTA-Based Coronary Artery Blockage Prediction

DOI

10.2991/978-94-6239-791-0_7How to use a DOI?

Keywords

Coronary CT Angiography; Stacking Ensemble Learning; Feature Engineering; Coronary Artery Disease; Cardiovascular Risk Prediction; Machine Learning; Clinical Decision Support; Non-invasive Diagnosis

Abstract

Coronary artery disease (CAD) remains one of the top mortality causes around the world. Coronary CT angiography (CCTA) is an efficient non-invasive technique used for evaluation of coronary artery stenosis. It is nowadays used in various clinical applications, including screening and preprocedural assessment. There have been new strides in machine learning technology that sparked interest of researchers into the application of automated techniques for the analysis of cardiovascular risk factors and subsequent prediction of coronary events based on the use of CCTA data related [3, 4, 5]. However, many current solutions are based on deep learning approaches that typically require large amounts of annotated imaging data as well as high computational power which may hinder their application in practice [6, 7]. In the paper, we present a machine learning technique aimed at calculating coronary artery blockage in percentage based on the structured information about patients comprising their demographic, metabolic, and lifestyle data. A stacking ensemble regression model was established by merging a number of base learners along with a Ridge regression meta-learner, with the aim to enhance predictive stability. By utilizing the developed clinical features, the model was able to reach an R2 score of around 0.98, thus showing high predictive quality. Then to enhance the interpretability of the developed system, the LLM (large language model) was also introduced to produce the patient-level explanations and clinical conclusions that are based on the prediction outcomes made.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ayush Vaze
AU  - Swarangi Gurav
AU  - Purushottam Mahajan
AU  - Kalyani Batwal
AU  - Omkaresh Kulkarni
PY  - 2026
DA  - 2026/10/09
TI  - Cross-Validated Machine Learning Stacking Ensemble Regression for CCTA-Based Coronary Artery Blockage Prediction
BT  - Proceedings of the International Conference on Artificial Intelligence and Machine Learning for Agro-Ecosystems (ICAIML-AE 2026)
PB  - Atlantis Press
SP  - 64
EP  - 74
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-791-0_7
DO  - 10.2991/978-94-6239-791-0_7
ID  - Vaze2026
ER  -