Search

New AI software can detect heart disease through smartphone  

Press Release 04:30 PM, 3 Aug, 2024
New AI software can detect heart disease through smartphone  

An interdisciplinary team from the Li Ka Shing Faculty of Medicine at the University of Hong Kong (HKUMed) has developed an innovative artificial intelligence (AI) software system, patented as Vitogram®, that transforms a regular smartphone into a stethoscope, enabling users to detect symptoms of heart disease, and allowing for prompt medical consultation without the need for specialized equipment.

The innovative AI system has demonstrated an 81pc accuracy rate in detecting valvular heart disease, matching the effectiveness of traditional stethoscope examinations by doctors. 

This breakthrough technology promises significant advancements in early disease detection, personalized health management, heart disease screening, telemedicine, and remote monitoring.

The project, led by Prof Joshua Ho Wing-kei, Associate Professor at the School of Biomedical Sciences, HKUMed, and Lead Scientist at the HKU Laboratory of Data Discovery for Health (D24H), addresses the gap in digital health devices capable of analyzing heart sounds.

Prof Ho says, "Analyzing heart sounds provides critical health data comparable to that obtained by a stethoscope. Our AI system, ‘Vitogram®,’ turns a smartphone into an AI-powered health tool, enabling convenient self-assessment of heart health."

The concept for this technology arose during the Covid-19 pandemic when many avoided medical visits for minor ailments, missing early heart disease detection opportunities.

 Prof Ho expressed pride over the innovative software winning the Gold Medal at the 49th International Exhibition of Inventions of Geneva.

Dr. Wong Chun-ka, Cardiologist and Clinical Assistant Professor in the Department of Medicine, School of Clinical Medicine, HKUMed, has been leading a clinical trial since December 2023 to validate the AI system’s efficacy.

Preliminary results from 363 participants show the system achieves 97pc accuracy in heart rate estimation and 81pc accuracy in detecting significant valvular heart disease, rivaling traditional medical-grade stethoscopes and cardiologist assessments.

Dr. Wong highlighted the technology's potential to uncover hidden cases of valvular heart disease, enabling early intervention and regular self-checks to prevent severe conditions and reduce recovery times.

The research team plans to expand the study to include a larger and more diverse population and additional heart disease symptoms.

The Laboratory of Data Discovery for Health (D²4H) focuses on compiling extensive data resources and developing advanced analytics to enhance global public health and precision medicine.

Led by the University of Hong Kong and supported by renowned academic institutions, D²4H aims to collaborate with organizations such as the World Health Organization and the China Centre for Disease Control and Prevention.

By integrating interdisciplinary efforts, D²4H seeks to advance healthcare technology in Hong Kong, the Greater Bay Area, and beyond, striving to produce transformative health solutions with global benefits.