National Heart Centre Singapore (NHCS) has announced the recent establishment of the Cardiovascular Systems Imaging and Artificial Intelligence (CVS.AI) research laboratory. Studying various aspects of heart disease using different imaging methods, it is the first AI-driven research laboratory to be established in Singapore and Southeast Asia.

NHCS sees more than 120,000 outpatient visits annually and performed nearly 17,000 cardiovascular scans in 2021. These cardiac scans assess the structure and function of the heart, and are used to detect new disease, monitor progression, and help guide treatments for heart disease.

Recognizing the value of the vast information collected and available in NHCS from the cardiac imaging it has done, NHCS set up the CVS.AI research laboratory to leverage the power of AI to harness this information to enhance the prediction and detection of cardiovascular disease. 

The objectives of setting up the CVS.AI research laboratory are:

  • To provide AI techniques in capturing and interpreting cardiac images to detect and predict cardiovascular disease 
  • To conduct in-depth assessments of cardiovascular disease amongst at-risk populations in Singapore 
  • To enable the discovery new knowledge of complex patterns of cardiovascular disease

“We are excited to spearhead research on the use of AI in the clinical setting. AI can potentially capture and make sense of complex image information in a multidimensional manner that cannot be done by a human interpreter—improving predictability and reducing labor-and time-intensive processes in analyzing scan images,” says NHCS Consultant Lohendran Baskaran, MD, who is the clinical lead for the laboratory. For instance, what would usually take between two to four hours of interpretation work by a radiologist, can be done within minutes by AI. 

Following an earlier research study where AI technology was used in a group of patients’ CT scans to analyze their coronary arteries and understand patterns of progression of atheromatous plaque, the team is now working on integrating various types of information from CT scans to predict markers of future heart disease.

 “We witness how AI can help us make sense of small and large pieces of data, and even predict future progression of disease, and its great potential to address challenges relating to diagnostics, and eventually improve health outcomes,” says Associate Professor Zhong Liang, PhD, the technical lead of the laboratory, and principal investigator, National Heart Research Institute Singapore, NHCS.

The use of AI can report a great level of details, providing clinicians with in-depth insight of the patient’s condition. For instance, AI can detect sub-millimeter changes beyond the human capacity, such as empowering MRI to evaluate the curvature of the heart at 0.5 mm, and to detect and monitor subtle changes in cardiac muscle injury when blood vessels as small as 0.2 mm in caliber are not supplying cardiac tissue. These details enable us to track and monitor the progression of disease, as well as help us manage and institute treatment at a much earlier stage, potentially increasing its benefits to our patients.

The CVS.AI research laboratory will drive initiatives to improve the precision and efficiency in predicting and identifying the world’s largest killer—cardiovascular disease, such as coronary artery disease, congenital heart disease, and heart failure. The various types of cardiac imaging modalities used include CT, MRI, and nuclear imaging.

“With the integration of AI strategies into cardiovascular care provision such as the use of AI algorithms in assessing patients’ risk of heart disease, physicians will be able to identify at-risk individuals and institute personalised preventative therapeutic interventions effectively. Over time, this can potentially save costs, broaden accessibility of cardiac scans to patients, and improve overall healthcare for our population,” says Terrance Chua, MD, NHCS’s medical director.

“With the burgeoning AI industry, partnerships with local enterprises may also be possible to further catalyze AI tools and techniques to benefit our heart patients,” Chua adds.