In this work, we develop a device, called 'Walk-Even', that can provide real-time feedback to correct gait asymmetry commonly exhibited in post-stroke survivors and persons with certain neurological disorders. The device computes gait parameters, including gait time, swing time, and stance time of each leg, to detect gait asymmetry and provide corresponding real-time biofeedback by means of auditory and electrotactile stimulation to actively correct the user's gait. The system consists of customized forcesensor-embedded insoles adjustable to fit any shoe size, electrotactile and auditory feedback circuits, microcontroller, and wireless XBee transceivers. The device also offers data saving capability. To validate its accuracy and reliability, we compared the gait measurements from our device with a commercial gait and balance assessment device, Zeno Walkway. The results show good correlation and agreement in a validity study with six healthy subjects and reliability study with seventeen healthy subjects. In addition, preliminary testing on six post-stroke patients after an 8-week training shows that the Walk-Even device helps to improve gait symmetry, foot pressure and forefoot loading of the affected side. Thus, initial testing indicates that the device is accurate in measuring the gait parameters and effective in improving gait symmetry using real-time feedback. The device is portable and low cost and has the potential for use in a non-clinical setting for patients that can walk independently without assistance. A more extensive testing with stroke patients is still ongoing.
We performed an electronic database search of published works from 2012 to mid-2021 that focus on human gait studies and apply machine learning techniques. We identified six key applications of machine learning using gait data: 1) Gait analysis where analyzing techniques and certain biomechanical analysis factors are improved by utilizing artificial intelligence algorithms, 2) Health and Wellness, with applications in gait monitoring for abnormal gait detection, recognition of human activities, fall detection and sports performance, 3) Human Pose Tracking using one-person or multi-person tracking and localization systems such as OpenPose, Simultaneous Localization and Mapping (SLAM), etc., 4) Gait-based biometrics with applications in person identification, authentication, and re-identification as well as gender and age recognition 5) “Smart gait” applications ranging from smart socks, shoes, and other wearables to smart homes and smart retail stores that incorporate continuous monitoring and control systems and 6) Animation that reconstructs human motion utilizing gait data, simulation and machine learning techniques. Our goal is to provide a single broad-based survey of the applications of machine learning technology in gait analysis and identify future areas of potential study and growth. We discuss the machine learning techniques that have been used with a focus on the tasks they perform, the problems they attempt to solve, and the trade-offs they navigate.
Asymmetrical gait and a reduction in weight bearing on the affected side are a common finding in chronic stroke survivors. The purpose of this pilot study was to determine the effectiveness of a shoe insole device that we developed, called Walk-Even, in correcting asymmetric gait in chronic stroke survivors. Six individuals with chronic (>6 months) stroke underwent 8 weeks of intervention with 2 sessions/week, each consisting of 20 minutes of gait training and 20 minutes of lower-extremity strength training. The 2 control participants underwent conventional gait training, while 4 participants underwent gait training using the Walk-Even. Following intervention, all the participants improved on most of the gait measures: peak pressure of the foot, time of transfer of weight from heel-to-forefoot, center of pressure (COP) trajectory, COP velocity, asymmetry ratio of stance, mean-force-heel, mean-force-metatarsals, Timed “Up and Go,” and Activities-specific Balance Scale. The improvement was more pronounced in the 4 participants that underwent training with Walk-Even compared to the control participants. This pilot study suggests that a combination of strength and gait training with real-time feedback may reduce temporal asymmetry and enhance weight-bearing on the affected side in chronic stroke survivors. A large randomized controlled study is needed to confirm its efficacy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.