Respirascan

RespiraScan is a portable, non invasive wearable that uses IMU sensors and machine learning to screen respiratory and musculoskeletal lung conditions. It captures three dimensional chest motion, applies Kalman filtering for noise reduction, and uses LSTM models to classify Ankylosing Spondylitis, interstitial lung diseases, and occupational lung diseases. The system achieves about 90 percent classification accuracy, offering a low cost and accessible alternative to traditional imaging based diagnostics, especially for early detection and resource limited settings.

This project was an ISEF ’25 Finalist and won the Gold Crest Award.