Radar heart sounds detect heart dysfunction without physical contact
Clinical radar heart sounds with echocardiographic phenotypes of ventricular dysfunction and valvular disease
Computer Vision and Pattern Recognition
Summary
Heart sounds can reveal how well your heart is working. The authors collected radar-recorded heart sounds from patients and matched them with ultrasound images that show heart function. This helps connect radar signals to heart problems like poor pumping or leaking valves. Their data and analysis show radar can pick up important mechanical signals related to heart disease. This work makes it easier for others to study heart health using radar heart sound technology.
What this means in practice
- •For cardiology device developers: Develop non-contact heart monitoring devices using radar signals linked to echocardiographic heart dysfunction markers.$Commercial implications: It enables creating radar-based diagnostic tools for ventricular and valvular heart diseases sellable to hospitals and clinics.
- •For biomedical signal analysts: Analyze paired radar and echocardiography data to extract cardiac mechanical features associated with heart dysfunction across patients.
Authors
Jinye Li, Qingchao Chen
Abstract
Radar heart sounds capture cardiac mechanical activity without contact, while echocardiography characterizes ventricular and valvular function. Paired measurements connect these signals to structural heart disease, yet such data remain sparsely represented in public resources. Here we present 537 radar heart-sound recordings paired with one echocardiographic report for each of 162 patients. The cohort covers three overlapping phenotypes: left ventricular ejection fraction (LVEF) below 50%, regional wall-motion abnormality and at least moderate valvular regurgitation, present in 38, 43 and 39 patients, respectively. Original complex signals, processed radar heart sounds and temporal quality segmentation support signal-to-phenotype analysis. Technical validation retains 95.93% of signal duration and confirms reproducible waveform reconstruction. Patient-level five-fold evaluation achieves a mean outer-test AUROC of $0.770\pm0.138$ for reduced LVEF, identifying mechanical information associated with ventricular systolic dysfunction. This resource supports research linking non-contact cardiac mechanical signals to echocardiographic structural and functional phenotypes.