Smart Insoles With Textile Capacitive Sensors for Efficient Gait Phase Recognition
Year: 2025
Type: article
Abstract: Gait analysis is important in understanding human movement and diagnosing various mobility-related disorders. A crucial step in this procedure is the recognition of the different phases of human gait. An innovative, cost-effective prototype of shoe insoles with sensors is presented, designed to capture data which are used for gait phase recognition with the aid of artificial intelligence. Each insole is embedded with capacitive sensors to record measurements and is accompanied by an inertial measurement unit (IMU) that monitors foot motion during walking. Using these prototypes, we collected data from multiple volunteers while they performed simple walking exercises. The collected data, having undergone preprocessing, were used to train a Long Short-Term Memory (LSTM)-based neural network that recognizes the different gait phases during walking. The resulting model can segment a gait session into distinct phases with high accuracy, effectively demonstrating the capability of the prototype pair of insoles to be used as an inexpensive way to record data that can be used for gait analysis. The presented method has potential applications in rehabilitation and the early detection of mobility impairments. The most important benefit of the developed system is that it constitutes an affordable alternative to commercial solutions for patients who require medical care through telemedicine.
Source: IEEE Sensors Journal
Institutions: University of Piraeus, University of Novi Sad
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,645
Award: G8316719087
OpenAlex ID: W4416749334