SmartHeart: Smarter Cardiac Sensing via Integrated Signal Processing
by IT on 09-04-2019
SmartHeart is an IT internal project that tackles the challenges of transforming how we address public health policies for cardiology. It builds upon original work developed by the team in the field of cardiac sensing and signal processing, having as main outputs: Novel machine learning methods for automated feature extraction and pre-screening of cardiac diseases from PCG & ECG data; An improved low-cost robust hardware and software for cloud-based cardiac signal acquisition and storage; Integration with a “Google Docs”-like online platform for collaborative multimodal cardiac data sharing and annotation; A prototype of a novel device to enable simultaneous PCG & ECG data acquisition in the form factor of a standard stethoscope with synchronized data acquisition.
Achieving these ambitious objectives was made possible by the unique opportunity created by SmartHeart to join three research groups of IT – the Computer graphics and interactive multimedia-Po, Network Architectures and Protocols-Cv, and Pattern and Image Analysis-Lx groups – which have independently created state-of-the-art technologies and accumulated knowledge in this field. Key to the success of the project was the technological maturity and success of previous projects such as DigiScope and BITalino, combined with the highly multi-disciplinary profile of the research team, ranging from signal processing (Miguel Coimbra, Nuno Garcia, Nuno Pombo, Hugo Silva, Mónica Martins), software engineering (Pedro Brandão, Hugo Silva, Nuno Garcia, Nuno Pombo), artificial intelligence (Nuno Pombo), hardware (Hugo Silva, Margarida Reis) and cardiopneumology (Cristina Oliveira). A collaboration with the Polytechnic Institute of Setúbal (Ricardo Baptista, Miguel Rocha) has further enriched the project, with industrial and materials engineering.
In addition to the main outputs, the project has resulted in a patent-pending suite of technologies, which will start to be showcased to industry stakeholders. The team believes to have significantly contributed to the development of smarter cardiac sensing tools that integrate innovative hardware and signal processing methods.