Researchers at the University of Illinois Propose An Artificial Intelligence (AI) Based Framework That Uses Video and Deep Learning Algorithm To Diagnose Parkinson’s Disease Gait Dysfunctions

Gait dysfunctions are modifications to your regular walking pattern that are frequently linked to a sickness or an anomaly in various body parts. An estimated 17% of falls in older persons are caused by gait dysfunction, making it one of the most frequent causes of falls.

Doctors frequently evaluate patients’ ability to walk when they suspect they may have a particular neurological ailment, such as multiple sclerosis or Parkinson’s disease. Someone’s walk alone may reveal information about an underlying neurological disease.

In a recent study, researchers investigated a method for assessing a person’s gait and identifying those who might have Parkinson’s disease or MS using conventional video cameras and AI. The findings demonstrate that the method can achieve accuracies of up to 79%.

Even in the early-to-mid stages of an illness, neurological problems can frequently result in minor changes in a person’s gait. Medical practitioners frequently employ specialized tools to look for neurological problems in a person’s gait, such as force plates, electromyography sensors, or lab-based motion capture systems. These tools can be expensive and require trained individuals to interpret the results.

According to the researcher,  “the integration of video of people walking and AI may allow for a wider range of health care providers in rural or underserved communities to identify early gait changes from neurological conditions and more efficiently provide a potential diagnosis.”

33 volunteers—10 with MS, 9 with Parkinson’s disease, and 14 without any neurological disorders—were enlisted by the researcher. Two common RGB cameras were used to record the movements of all of the volunteers while they walked on a treadmill while doing so.

“Researchers looked for differences between persons with and without MS or Parkinson’s disease by analyzing how these coordinates changed over time.”

The researchers created and evaluated 16 alternative AI systems to evaluate these gait motions. The top-performing algorithm—a convolutional deep-learning model—achieved 79 % accuracy in identifying a person’s neurological state, and several algorithms were more than 75% successful in doing so.

They were pleasantly surprised with the validation results of using somewhat inexpensive video equipment and open-source image processing software to get the performance they saw.” “If built properly, this might revolutionize the game.”

Neurological disorders’ progressive manifestation and because aging is varied, identifying sudden PwMS and PwPD are particularly challenging to modify. They showed a brand-new vision and deep learning pipeline for classifying Gait dynamics used with PwMS and PwPD. In this research, 3D multi-view fused body keypoint coordinates were derived from the captured gait videos, illustrating the advantages of neurological gait differentiation using DL structures. Further, they assessed how well this framework generalized to various walking tasks and people. Studying digital, a potential in-home gait is provided by a camera-based infrastructure monitoring instrument to support the diagnosis. This might also be advantageous to reduce MS and PD, professionals must work with patients as well.

This Article is written as a research summary article by Marktechpost Staff based on the research paper 'A Vision-Based Framework for Predicting Multiple Sclerosis and Parkinson’s Disease Gait Dysfunctions - A Deep Learning Approach'. All Credit For This Research Goes To Researchers on This Project. Check out the paper and reference article.
Please Don't Forget To Join Our ML Subreddit

Ashish kumar is a consulting intern at MarktechPost. He is currently pursuing his Btech from the Indian Institute of technology(IIT),kanpur. He is passionate about exploring the new advancements in technologies and their real life application.

🐝 Join the Fastest Growing AI Research Newsletter Read by Researchers from Google + NVIDIA + Meta + Stanford + MIT + Microsoft and many others...