The National Heart Centre Singapore (NHCS) is initiating a pilot program for a new AI-based solution aimed at the rapid prediction of coronary artery disease (CAD).

SENSE (Singapore Heart Lesion Analyser) is a machine learning system designed to automatically interpret heart imaging scans and assess CAD risk. It utilizes AI algorithms and proprietary post-processing technologies to evaluate four key CAD indicators from CT scans:

• Coronary calcium scores
• Epicardial adipose tissue
• Coronary stenosis
• Plaque

These parameters were identified through the NHCS APOLLO cardiac database.

According to NHCS, SENSE can produce reports for clinicians in 10 minutes, a significant reduction from the usual processing time of two to four hours.

Importance of the Initiative

CAD is responsible for nearly one-third of cardiovascular-related fatalities in Singapore. The NHCS aims to mitigate these deaths by employing AI for early detection and monitoring of CAD.

The SENSE project is being led by the CardioVascular Systems Imaging and Artificial Intelligence Research Laboratory (CVS.AI) in collaboration with the A*STAR Institute for Infocomm Research. Established in 2021, the laboratory has recently expanded its facilities and enhanced its GPU capabilities, enabling real-time processing of extensive datasets.

SENSE builds on the CVS.AI’s flagship APOLLO database, which contains approximately three million heart images and data from around 5,000 patients across major heart centers in Singapore.

“SENSE offers high precision in predicting CAD, enabling clinicians to implement timely and effective treatment strategies,” stated Lohendran Baskaran, NHCS assistant professor and director of CVS.AI.

In addition to NHCS, the CAD prediction tool will also be tested at two other prominent cardiac hospitals: the National University Hospital and Tan Tock Seng Hospital (TTSH) in the third quarter of this year.

Broader Context

In a related development, Health BETA, a Singaporean startup, is also developing an AI-driven tool to assess CAD risk using Asian genomics data to create an improved polygenic risk score for CAD. The founders, Tan Kae Yuan and Heng Chew Kiat, anticipate launching this solution later this year.