Foot type classification using sensor-enabled footwear and 1D-CNN. (1st December 2020)

• Record Type: Journal Article
• Authors: Mei, Zhanyong; Ivanov, Kamen; Zhao, Guoru; Wu, Yuanyuan; Liu, Mingzhe; Wang, Lei
• Is Part Of: Measurement. Volume 165(2020)
• Publication Date: 2020-12-01
• Subjects: Foot type classification -- Sensor insole -- 1D CNN -- Inertial sensor -- Force sensor; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.108184
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: Modern lifestyle and lack of regular foot checks pose risks for lower limb health. Technologies can help for automated foot type screening and early alarm. We use sensor insoles and artificial intelligence for foot type classification. We achieved an accuracy of 99.26% to classify normal, cavus, and planus feet. The suggested method can contribute to affordable foot mass screening. Abstract: Poor selection of footwear, underestimation of foot health, sedentary life, and lack of accessible foot screening can have significant long-term adverse effects on the health of lower limbs. Unobtrusive, pervasive methods for automated foot screening have the potential to allow for timely detection of foot abnormalities. In the present study, we describe a proof-of-concept where data collected through sensor-enabled insoles and processed through one-dimensional convolutional neural networks were used to distinguish normal, cavus, and planus feet. We explored several combinations of sensor modalities to find the one that reflects foot types optimally. The highest accuracy of classification of 99.26% was achieved when angular velocity and force sensing were combined. Based on results, we suggest that sensor insoles, combined with optimal classification techniques, could be used for foot screening.