Fuzzy Logic-Based Gait Phase Detection Using Passive Markers

Abstract

With the advancement in technology, gait analysis plays a vital role in sports, science, rehabilitation, geriatric care, and medical diagnostics. Identification of accurate gait phase is of paramount importance. The objective of this paper is to put forward a novel approach via passive marker-based optical approach that automatically recognizes gait subphases using fuzzy logic approach from hip and knee angle parameters extracted at RAMAN lab at MNIT, Jaipur. In addition to stance phase and swing phase, the approach is capable of detecting all the subphases such as initial swing, mid swing, and terminal swing, loading response, mid stance, terminal stance and preswing. The prototype of the system provides an effective and accurate gait phase that could be used for understanding patients' gait pathology and in control strategies for active lower extremity prosthetics and orthotics. It is an automated, easy to use, and very cost-efficient yet reliable model.

A Standalone Real-Time Gait Phase Detection Using Fuzzy-Logic Implementation in Arduino Nano

“…Analytical study of human walk and gait cycle started with Galileo, Da Vinci, and Newton's description in the fifteenth and sixteenth Centuries. Weber brothers in 1836 Germany gave the earliest apt description of the gait cycle [17]. In recent years, clinical analysis of human gait has gained traction, with the seminal work done by Perry et al [20], and Whittle [2].…”

“…According to Perry's Classical gait model [20], the average contribution of sub-phases of a gait cycle as a whole on a time scale is mentioned in Table 1. However, a few authors have reported loading response as 0-10% by considering 7 sub-phases in a gait cycle [17,21].…”

“…Researchers have used fuzzy-logic techniques to control upper limbs [15] and lower limb prostheses [16]. Among these studies, a few researchers worked on image-based control [17], whereas others have used wearable sensors such as Electromyography (EMG), IMU sensors, FSR sensors on foot insole, and knee and ankle angle sensors [18]. During the literature survey, all authors used the Fuzzy Logic toolbox in MATLAB software to implement the fuzzy-logic rule for their applications; a few researchers also used the Neuro-Fuzzy Designer toolbox to design, train, and test adaptive neuro-fuzzy inference systems (ANFIS) using input/output training data.…”

Gait Phase Identification and Damping Control for Knee Orthosis Using Time Series Forest Classifier

“…C. Prakash et al described a unique method for identifying gait phases by utilizing an optical approach with passive markers. It uses fuzzy logic to automatically recognize different subphases of gait, resulting in a precise and reliable measurement useful for analyzing gait abnormalities in patients and for the use in the development of control strategies for active lower-extremity prosthetics and orthotics [9]. Taborri et al proposed a distributed classifier for gait phase detection in an active knee orthosis for pediatric subjects with neurological diseases.…”