Induced Acoustic Resonance for Noninvasive Bone Fracture Detection Using Digital Signal Processing and Machine Learning

Abstract

Bone fractures require radiographic imaging services for complete diagnosis; however, cost, time, access, and availability of medical personnel are barriers for people living in remote areas and developing countries. Inspired by the automated detection of fractures in industry, this paper presents a technique and proof-of-concept device using modern digital signal processing and machine learning techniques to detect bone fracture. Using animal bones with synthetic soft tissues, fractures were identified by inducing vibrations and analyzing the resulting harmonics to detect structural defects, achieving 93.6% accuracy and offering an avenue for further research.

Publication
2020 IEEE Global Humanitarian Technology Conference (GHTC 2020)