Prescriptive Machine Learning in Pediatric Cardiology

Recent technological advances in data collection, transmission, visualization, and machine learning make it possible to develop healthcare systems with the aim of monitoring the health status of patients. Pediatric Cardiologists rely on referrals from family physicians for abnormal heart conditions. Family physicians are not trained well enough to recognize abnormal murmurs and err on the side of caution, which results in a large number of unnecessary referrals causing a strain on the healthcare system. This proposal proposes the development of software designed to receive the ECG signals, process them in real-time on a computing device such as a laptop in a family physician’s office, and create meaningful data summarization/visualization graphics, followed by prescriptive analysis and recommendations using machine learning (ML). The proposed system is designed to help physicians and medical assistants who have problems distinguishing ECGs and work in deprived regions. Kardio Diagnostix is founded by two Pediatric Cardiologists. They will be advising the interns on the collection, cleaning, and processing of ECG signals as well as model evaluation and software testing.

Faculty Supervisor:

Pawan Lingras

Student:

Partner:

Kardio Diagnostix

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

Saint Mary's University

Program: