Category
Clinical Biomechanics
Document Type
Paper
Abstract
This study aimed to determine the accuracy of distinguishing patients with early Parkinson’s disease (PD) (n=27) from healthy controls (n=50) using a convolutional neural network (CNN) technique with an artificial intelligence deep learning algorithm based on a 6-minute walk test (6MWT) using wearable sensors. After wearing the six sensors, the participants performed the 6MWT, and the time-series data were converted into new images. The main results demonstrated the highest discrimination accuracy of 72% on the left arm gyroscope data. The results confirmed the possibility of using CNN models to distinguish between individuals with early PD and controls. Moreover, the 6MWT using sensors may contribute to early diagnosis as an objective indicator in clinical settings.
Recommended Citation
Choi, Hyejin; Youm, Changhong; Park, Hwayoung; Kim, Bohyun; Hwang, Juseon; and Kim, Minsoo
(2024)
"POSSIBILITY OF EARLY DETECTION OF PARKINSON’S DISEASE USING CONVOLUTIONAL NEURAL NETWORK DURING SIX-MINUTE WALK TEST,"
ISBS Proceedings Archive: Vol. 42:
Iss.
1, Article 42.
Available at:
https://commons.nmu.edu/isbs/vol42/iss1/42