Wearable Antenna System for Osteoporosis Detection and Monitoring Using Machine Learning

This article presents a groundbreaking approach to osteoporosis detection and monitoring by integrating a new wearable monopole antenna design with advanced machine learning algorithm (neural network). Inspired by the intricate pattern of a Christmas snowflake, the system utilizes UWB electromagnetic waves and bone attenuation analysis for compact, noninvasive, and highly accurate bone health assessment. Fabricated entirely from textile materials, the antenna features remarkable performance metrics, including an impedance bandwidth of 4.9 to 12.6 GHz and a reflection coefficient consistently below −10 dB, within a compact form factor of 41.9 mm × 29.2 mm. Experimental validation and comparative studies demonstrate the effectiveness of this approach in precisely classifying osteoporosis levels, achieving an outstanding accuracy rate of 87%. This study signifies a significant advancement in osteoporosis detection and diagnosis, combining state-of-the-art antenna technology with advanced machine learning techniques. The developed system holds promise for early detection and personalized monitoring of osteoporosis, contributing to improved healthcare outcomes and enhanced quality of life for individuals at risk of bone-related diseases. © 2024, Electromagnetics Academy. All rights reserved.