Neurodegenerative diseases detection and grading using gait dynamics

Abstract

Detection of neurodegenerative diseases such as Parkinson’s disease, Huntington’s disease, Amyotrophic Lateral Sclerosis, and grading of these diseases’ severity have high clinical significance. These tasks based on walking analysis stand out compared to other methods due to their simplicity and non-invasiveness. This study has emerged to realize an artificial intelligence-based disease detection and severity prediction system for neurodegenerative diseases using gait features obtained from gait signals. For the detection of the disease, the problem is divided into parts which are subgroups of 4 classes consisting of Parkinson’s, Huntington’s, Amyotrophic Lateral Sclerosis diseases, and the control group. In addition, the disease vs. control subgroup where all diseases are collected under a single label, the subgroups where each disease is separately against the control group. For disease severity grading, each disease was divided into subgroups and a solution was sought for the prediction problem mentioned by various machine and deep learning methods separately for each group. In this context, the resulting detection performance was measured by the metrics of Accuracy, F 1 Score, Precision, and Recall while the resulting prediction performance was measured by the metrics such as R, R 2 , MAE, MedAE, MSE, and RMSE.

A Unified Framework for Neurological Disease Detection and Gait Classification Using Deep Graph Learning

“…Slemensek et al [9] introduced a wearable system adept at capturing gait motion data with precision, facilitating not only the classification of gait activities but also the identification of potential risk factors, thereby aiming to improve individuals' overall quality of life. Berke et al [10] proposed an advanced artificial intelligence-driven system tailored for detecting neurodegenerative diseases and predicting their severity by leveraging gait features extracted from gait signals. Their segmentation approach enables a targeted analysis of disease-specific gait patterns and characteristics, enhancing diagnostic precision.…”

“…Slemensek et al [9] introduced a wearable system for gait motion data capture, albeit with limitations in accurately classifying gait activities and identifying risk factors. Similarly, Berke et al [10] devised an AI-driven system for neurodegenerative disease detection, yet their approach may lack the versatility International Journal of Intelligent Engineering and Systems, Vol.17, No.3, 2024 DOI: 10.22266/ijies2024.0630.41 needed for comprehensive gait analysis. Shanmuga Sundari et al [11] focused on predicting the Age of Gait using specific walking tasks, which may overlook broader gait abnormalities.…”

Data-driven Gait based Severity Classification for Parkinson's Disease using Duo Spatiotemporal Convoluted Kernel Boosted ResNet model

“…However, neuroimaging techniques are cost-effective in capturing motion illuminations and other force estimations [5]. Hence, the gait analysis via placing sensors on the human foot becomes an integral role in analyzing leg movements which is a highly accurate and affordable technique [6].…”

Using Machine Learning Algorithms for Neurodegenerative Disease Gait Classification

“…The highestaccuracy classification rate between healthy and neu-rodegenerative diseases groups was 96.83%, achieved by using SVM. Similarly, in [14] the extra trees method and convolutional neural networks were additionally included, however the best accuracy was achieved with the K Nearest Neighbor method. Another example is presented in [15], which shows that uses of radial function neural networks have been able to classify 93.75%.…”