Classification of Neurological Gait Disorders Using Multi-task Feature Learning

Abstract

As our population ages, neurological impairments and degeneration of the musculoskeletal system yield gait abnormalities, which can significantly reduce quality of life. Gait rehabilitative therapy has been widely adopted to help these patients maximize community participation and living independence. To further improve the precision and efficiency of rehabilitative therapy, more objective methods need to be developed based on sensory data. In this paper, an algorithmic framework is proposed to provide classification of gait disorders caused by two common neurological diseases, stroke and Parkinson's Disease (PD), from ground contact force (GCF) data. An advanced machine learning method, multi-task feature learning (MTFL), is used to jointly train classification models of a subject's gait in three classes, post-stroke, PD and healthy gait.Gait parameters related to mobility, balance, strength and rhythm are used as features for the classification. Out of all the features used, the MTFL models capture the more important ones per disease, which will help provide better objective assessment and therapy progress tracking. To evaluate the proposed 1 The first two authors contributed equally to this work. methodology we use data from a human participant study, which includes five PD patients, three post-stroke patients, and three healthy subjects. Despite the diversity of abnormalities, the evaluation shows that the proposed approach can successfully distinguish post-stroke and PD gait from healthy gait, as well as post-stroke from PD gait, with Area Under the Curve (AUC) score of at least 0.96. Moreover, the methodology helps select important gait features to better understand the key characteristics that distinguish abnormal gaits and design personalized treatment.

The Role of Movement Analysis in Diagnosing and Monitoring Neurodegenerative Conditions: Insights from Gait and Postural Control

“…Promising attempts to model and classify dementia and PD using measures of gait and postural control with a variety of classification tools (e.g., support vector machines, hidden Markov models, multilayer layer perception, neural networks, etc.) have been reported [163,164,165,167,168,169,170,171]. Even though the perfect classification accuracy is reported with various techniques, the optimal method or combination of approaches has not been identified, much less tested.…”

Internet of Things Technologies and Machine Learning Methods for Parkinson’s Disease Diagnosis, Monitoring and Management: A Systematic Review

“…Furthermore, GRF signals have fed ML models for differential diagnosis of several neurological disorders. Papavasileiou et al [107] discriminate PD patients from post-stroke patients and from healthy controls with a multiplicative multi-task feature learning model, achieving 0.88-0.994 AUROC. Additionally, in [108], kNN, SVM and RF classifiers manage to discriminate PD patients from Huntington's disease (HD) and amyotrophic lateral sclerosis (ALS) patients, as well as from healthy controls with 81.25-90.91% accuracy.…”

Abnormal Gait Recognition Algorithm Based on LSTM-CNN Fusion Network

“…The potential of the human gait in biometrics authentication is proved by the CNN gait recognition algorithm. Using multitask feature learning (MTFL) in [14], the author classifies gait parameters related to mobility, balance, strength, and rhythm into two types of common neurological disorders, stroke and Parkinson's disease (PD). The evaluation shows that the proposed method can successfully distinguish between a stroke gait, PD gait and healthy gait.…”

“…This sensor-based wearable method relies on inertial sensors placed in different parts of the human body, including the waist and the ankle, to record gait information [8]. Therefore, wearable inertial sensors have been used in various studies related to inertial analysis, such as abnormal and normal gait analysis [14], fall detection [10], patient rehabilitation and treatment [11], and neurological system disease analysis [1]. The author provides a framework for data acquisition and signal processing to authenticate users according to their gait signatures.…”