Researchers at Stevens Institute of Technology in the United States have advanced artificial intelligence aimed at diagnosing heart disease. The system listens to heart rhythm recordings and analyzes faint acoustic cues that may signal pathology. Findings have been published in IEEE Xplore (IEEE Xplore), highlighting how AI can interpret the sounds produced by the heart to identify potential problems.
Traditionally, clinicians rely on a stethoscope to hear heart sounds, yet some murmurs or subtle signals may be missed, especially in busy clinical environments. The new technology reduces dependence on human hearing during examinations. By processing audio recordings that capture every sound produced by the heart in action, the AI framework provides a structured analysis of cardiac acoustics.
Initial evaluations show the tool can identify heart valve disease within seconds, offering rapid preliminary insights for clinicians. In broader tests, the approach analyzed the full audio signal of cardiac activity to recognize five distinct diseases with a reported 93% accuracy, even when multiple conditions coexisted. This level of performance suggests strong potential for assisting frontline screening and triage in primary care, emergency departments, and remote health settings.
Researchers emphasize that many heart defects go undetected due to human error or limitations in traditional auscultation. The AI-based method, with its high accuracy, could complement existing diagnostic workflows and help improve early detection rates. As the technology continues to evolve, teams are focusing on refining the signal processing, expanding the catalog of detectable conditions, and validating performance across diverse patient groups.
Earlier work by other scientists described a new category of cardiovascular conditions, underscoring how evolving computational tools can reshape understanding and detection in cardiology. The current line of investigation builds on that foundation, aiming to integrate AI-driven auscultation into routine clinical practice with rigorous validation and thoughtful deployment.