AI-Assisted Non-Invasive Diagnosis of Coronary Artery Disease and Heart Failure via Treadmill Effort Test

BAP — Higher Education 2025 – Present · Principal Investigator

The project develops a non-invasive cardiovascular diagnostic pipeline that combines treadmill-based effort testing with deep-learning analysis of ECG and physiological signals.

The ambition is to deliver a clinically deployable system capable of detecting coronary artery disease and heart-failure risk earlier and at lower cost than traditional imaging-based assessment. The pipeline ingests multi-channel ECG, respiratory, and treadmill performance data, and applies modern deep-learning architectures for both classification of disease state and estimation of risk.

The work continues ASIL’s sustained track record in cardiology AI, including peer-reviewed studies on ECG-based cardiovascular screening, machine-learning analysis of treadmill exercise testing, and convolutional architectures for arrhythmia detection. Outputs target both clinical decision support and population-level cardiovascular screening.

Funder: Yıldız Technical University BAP — Higher Education Institutions Support. Project period: 2025–Present.