Citation: Jeong, K.; Lee, K.-C. Artificial Neural Network-Based Abnormal Gait Pattern Classification Using Smart Shoes with a Gyro Sensor. Electronics 2022, 11, 3614. https://doi.org/10.3390/ electronics11213614 Academic Editors: Teen-Hang Meen, Chun-Yen Chang, Charles Tijus and Po-Lei Lee Received: 7 October 2022 Accepted: 2 November 2022 Published: 5 November 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). electronics Article Artificial Neural Network-Based Abnormal Gait Pattern Classification Using Smart Shoes with a Gyro Sensor Kimin Jeong 1 and Kyung-Chang Lee 2, * 1 Medical Device Development Center, Daegu-Gyeongbuk Medical Innovation Foundation, Daegu 41061, Korea 2 Department of Intelligent Robot Engineering, Pukyong National University, Busan 48513, Korea * Correspondence: gclee@pknu.ac.kr Abstract: Recently, as a wearable-sensor-based approach, a smart insole device has been used to analyze gait patterns. By adding a small low-power sensor and an IoT device to the smart insole, it is possible to monitor human activity, gait pattern, and plantar pressure in real time and evaluate exercise function in an uncontrolled environment. The sensor-embedded smart soles prevent any feeling of heterogeneity, and WiFi technology allows acquisition of data even when the user is not in a laboratory environment. In this study, we designed a sensor data-collection module that uses a miniaturized low-power accelerometer and gyro sensor, and then embedded it in a shoe to collect gait data. The gait data are sent to the gait-pattern classification module via a Wi-Fi network, and the ANN model classifies the gait into gait patterns such as in-toeing gait, normal gait, or out-toeing gait. Finally, the feasibility of our model was confirmed through several experiments. Keywords: gait analysis; artificial neural network; abnormal gait pattern; smart shoes 1. Introduction Changes in gait pattern such as fluctuations in walking speed or swaying of the body are often used as early indicators of cognitive impairment. This is because gait pattern is determined by individual gait characteristics such as lifestyle and differences in skeletal muscles. In addition, walking is an inherent human behavior that constantly changes from infancy to old age, but individual gait patterns are often fixed in childhood if poor habits develop. Therefore, it is necessary to identify an abnormal gait pattern and treat it early. Such findings have increased interest in the design of gait-pattern monitoring and assessment methods [1–6]. Gait-pattern monitoring and assessment systems can be classified into two different types: marker-based and wearable-sensor-based. The marker-based approach obtains walking-motion data using body-attached sensor systems using accelerometer [7,8] or EMG [9], video-based systems [10], active magnetic trackers [11], or optical-marker sys- tems [12]. However, these cannot be used outside the laboratory environment, invade privacy, and are expensive. The wearable-sensor-based approaches [13,14] use accelerom- eter sensors, pressure sensors, or biosensors worn on clothing or shoes equipped with low-power devices and portable memory for long-term ambulatory monitoring. It is possi- ble to capture and analyze gait information in real time over relatively long distances and outside the laboratory environment [12]. Recently, as a wearable-sensor-based approach, a smart-insole device that has a sensor embedded in the insole has been used to analyze gait patterns [15–18]. Such insoles can be put into any shoe, have the advantage of being compact and inexpensive, and can easily be integrated into small electronic devices. Therefore, smart insoles are currently among the best wearable devices to obtain gait information. In particular, by adding a small low-power sensor and an IoT device to the smart insole, it is possible to monitor the wearer’s activity, gait pattern, and plantar pressure in real time, and to evaluate exercise Electronics 2022, 11, 3614. https://doi.org/10.3390/electronics11213614 https://www.mdpi.com/journal/electronics