A solution developed by INESC TEC will make access to cardiac diagnoses easier in regions where reaching specialised care is more difficult. The project, which began this year, relies on Artificial Intelligence (AI) solutions.
In certain regions of the world, where the nearest doctor may be kilometres away, cardiovascular diseases claim the most lives; the statistics speak for themselves: 80% of deaths caused by heart problems occur in developing countries.
And these are the regions that INESC TEC seeks to reach – to detect and to act. Within the scope of the PULSE – Reliable Pervasive mULti-Sensor Cardiac Examination project, the Institute’s researchers are developing innovative AI solutions for the non-invasive assessment of the electromechanical activity of the heart.
These AI and deep learning methods hold great potential: not only can they identify disease patterns in heart sound and electrocardiogram signals with high accuracy, but they also eliminate the need for an expert to be present – a major advantage in remote and rural areas.
“From a tech innovation perspective, the project aims to create new multimodal deep learning algorithms capable of incorporating and exploring physiological knowledge to improve performance, and to develop trust and reliability metrics for AI systems applied to healthcare,” explained Francesco Renna, researcher at INESC TEC.
The PULSE solution will provide combined analysis of three physiological signals: heart sounds (PCG), electrocardiograms (ECG) and photoplethysmograms (PPG). But how will these signals be collected? The answer lies in “low-cost” devices – e.g., multimodal digital stethoscopes and wearables – which enable early screening of cardiovascular diseases and remote patient monitoring. “Especially in contexts with limited access to specialised care,” added the researcher.
The INESC TEC-led project was officially launched in July, and involves the Faculty of Medicine of the University of Porto, the Gaia/Espinho Local Health Unit, the Centre for Informatics and Systems of the University of Coimbra, as well as international partners in Brazil (Real Hospital Português, Recife) and in Africa (Carnegie Mellon University – Rwanda).
PULSE also represents the latest chapter in the Institute’s Multiscope research line. Funded under INESC TEC’s Internal Seed Projects programme (designed to support internal initiatives), the research “focuses on cardiovascular disease screening through the analysis of multimodal signals collected using non-invasive, low-cost devices, powered by innovative AI solutions.”
“The technological innovation of Multiscope lies in the design and development of AI algorithms that automate cardiovascular disease screening based on data collected by the multimodal stethoscope,” concluded Francesco Renna.
Two years ago, in collaboration with Carnegie Mellon University Africa (CMU Africa) and in partnership with clinical organisations, INESC TEC collected cardiac data in Rwanda to detect cardiovascular diseases caused by rheumatic fever.
The researcher mentioned in this news piece is associated with INESC TEC and UP-FCUP.
News, current topics, curiosities and so much more about INESC TEC and its community!
