New Study: AI Helps Smartphones Detect Abnormal Lung Sounds in Children

What if a regular smartphone could help screen children for respiratory problems, without requiring an electronic stethoscope? Researchers at Seoul National University of Science and Technology in South Korea have developed a domain-adaptive deep learning framework that helps AI detect abnormal lung sounds in smartphone recordings, addressing a key challenge: the differences between smartphone microphones and medical-grade stethoscopes.

Why This Matters

• Promising results on independent data: The model achieved an AUROC of 0.862 on a separate dataset of smartphone-only recordings from 48 children, outperforming the other smartphone-domain approaches tested.
• A technical advance: Rather than simply combining recordings from different devices, the framework adapts the model to account for differences in how smartphones and electronic stethoscopes capture sound.
• Potential for more accessible screening: The approach could support home-based respiratory screening, remote consultations, and community healthcare, but further real-world validation is needed before clinical deployment.

Associate Professor Seong-Eun Kim explains the challenge and the team's approach: "To a neural network, the smartphone and electronic stethoscope effectively speak different languages. Instead of simply pooling the data from the two devices, our approach offers reliable domain-adaption, where the model learns from reliable labelled-electronic stethoscope recordings while adapting to the smartphone domain."

The study was published in the IEEE Journal of Biomedical and Health Informatics on July 30, 2026.