: A prospective, Cross-Sectional Study to Support Machine Learning Model Development for Screening Cardiovascular Disease Conditions in Populations with Diabetes and Hypertension: Mapping Mobile ECG Device Signals to Gold-Standard Tests

Given the high burden of cardiovascular disease, and difficulties in accessing and providing diagnostic services for CVDs among the rising numbers of patients with diabetes and hypertension in India, an Artificial Intelligence point of care tool that can detect high risk cardiovascular conditions will be useful for primary care. The goal of the study is to develop machine learning model to detect common cardiovascular disease conditions such as coronary artery disease in populations with diabetes and/or hypertension. This would improve early detection and enable timely treatment, especially in populations with limited healthcare access. 

The study proposes to collect data from patients seeking care for diabetes and/or hypertension at various levels of healthcare facilities at Christian Medical College Vellore. Patients will undergo point-of-care tests like 1 to 12 lead mobile ECGs, that can be used in primary care setting, as well as gold-standard diagnostic measurements such as echocardiograms and 12 lead ECGs, all of which will be reviewed by cardiologists for diagnosis of abnormalities. Standard assessments like vitals and POC blood sugar will also be collected. Additionally, we will gather information on medical history, treatment history, and health related behaviours to provide a comprehensive representation of the patient, which will help with model development. Christian Medical College Vellore will manage, clean and anonymize the dataset, following which it will be shared with Indian Institute of Technology Madras for model development.