Singapore researchers invent new AI tools to speed up heart disease diagnosis

Singapore-A new tool has been invented by Singapore researchers that may lead to a faster diagnosis of heart disease.

It is equipped with artificial intelligence (AI) and uses an electrocardiogram (ECG) with an accuracy of 98.5%. ECG measures the electrical activity of the heartbeat to detect abnormalities in the heart.

The Singapore Heart Foundation figures show that nearly one of the three people who died in Singapore in 2019 was due to heart disease or stroke.

Deaths from cardiovascular disease in Singapore also increased from 2017 to 2019.

Early detection can also prevent complications such as kidney disease.

Invented by a team from Nanyang Technological University (NTU), Ngee Ann Polytechnique, and National Heart Center of Singapore (NHCS), this new diagnostic tool is an AI machine learning algorithm that allows computers to learn from past experiences like humans. using.

Using the algorithm, researchers trained tools to recognize patient ECG patterns by inputting examples of ECG signals that reflect cardiovascular disease.

In a pilot study, researchers used this tool to analyze ECG signals from 92 healthy individuals and 7 patients with coronary artery disease; 148 had a heart attack and 15 had congestive heart failure. I was suffering from.

The data was taken from four public databases.

The team found that AI tools could identify ECG signals associated with healthy people and patients with three different cardiovascular diseases with an accuracy of over 98.5%.

The study was published in a peer-reviewed scientific journal in Computers in Biology and Medicine in May.

“Although additional tests are needed to identify a particular disease, our diagnostic tools allow doctors to triage patients more efficiently and streamline the number and types of downstream confirmatory tests,” he said. Clinical Associate Professor Tan Ru San, Senior Consultant in Cardiology, said at NHCS, co-author of the study.

Associate Professor Eddie N in Quie of the Faculty of Mechanical and Aerospace Engineering, Nanyang Technological University, who co-led the research, said diagnostic tools could lead to advances in the fusion of AI and medical solutions.

This tool is still in the preliminary stages of research and needs to be validated with more data.

The team plans to begin further trials using a large database from local hospitals within 18-24 months to validate its clinical use.

Professor Tan said: “It’s very exciting, but it also requires tampering with our enthusiasm for reality … this is just the beginning.”

“Technology can be improved at any time and there may be some great tricks the team can actually do, but I think this is a really promising start,” he added.

Dr. Daniel Yeo, a cardiologist at Apex Heart Clinic at Gleneagles Hospital, welcomed this development, and new AI tools help doctors, nurses, and other related healthcare professionals identify anomalous ECGs for a more reliable assessment. States to help you do.

“The number of cardiologists is small and we cannot read every electrocardiogram performed, so this AI tool is useful in common practice environments such as large and busy polyclinics,” says Dr. Yeo. Stated. The tool is not too expensive.

Dr. Paul Ong, a cardiologist at Heart Specialist International at the Mount Elizabeth Novena Specialist Center, said AI tools help unskilled people quickly identify patients with cardiac emergencies in ambulances and emergency departments. Said it would be useful. It relies on AI to make diagnoses and miss the patient’s medical history and test results.

The researchers hope that this tool can complement current heart disease diagnostic techniques such as magnetic resonance imaging.

He also states that this tool can be extended to detect other heart conditions such as atrial fibrillation, which is a form of irregular rhythm of the human heart.

The tool can also be used on wearable devices such as smart watches to monitor the patient’s heart condition in the future.