The four most important components of random forest regression are stride time, maximum toe clearance, heel strike angle and stride length. This suggests a potential relationship between fatigue and these gait measurements. The results also showed that there was only a moderate correlation between fatigue and extended disability scale.

According to the research results published in the Journal of neural engineering and rehabilitation, the spatiotemporal gait measurement based on inertial measurement unit (IMU) system can predict the fatigue degree of patients with multiple sclerosis (MS).

Gait disorder is often the result of the effect of MS on systemic function. However, the most common symptom of MS is fatigue. Previous data have shown the relationship between the two, but wearable biosensors that provide gait data have not been used in early studies.

In the current study, researchers attempted to further determine the relationship between patient reported fatigue and spatiotemporal gait measurements collected from wearable foot piercing IMU sensors. They also investigated changes in gait measurements during the six minute walk test (6MWT) and the best way to predict fatigue in MS patients.

The study included 49 MS patients (female, 32; mean age, 41.6 years). The researchers collected data from the MS center of the University Hospital of Erlangen in Germany and Quellenhof, a neurological rehabilitation center. Each patient wore an IMU on the lateral ankle of each foot. The researchers then determined spatiotemporal gait parameters from the 6MWT and assessed fatigue using the Borg scale.

Then, the researchers used fatigue as the dependent variable and the combination of normalized gait parameters as the independent variable for regression analysis. In order to minimize the type 1 error, they performed principal component analysis on the data and converted the normalized gait parameters into components with significant changes.

The six principal components used explain more than 90% of the variance in the dataset. Researchers used random forest regression to predict fatigue. The model was verified by 10 fold cross validation (mean absolute error, 1.38 ± 1.07).

The four most important components of random forest regression are stride time, maximum toe clearance, heel strike angle and stride length. This suggests a potential relationship between fatigue and these gait measurements. The results also showed that there was only a moderate correlation between fatigue and extended disability scale.

Limitations of this study include relatively small data sets, not considering factors that affect fatigue (such as sleep disorders, depression and cognitive impairment), and the need to investigate other fatigue measurement scales and their relationship with Borg scale.

Finally, the researchers concluded: “wearable sensors can be used for gait analysis in unlimited, continuous and long-term ground walking pains. We observed a correlation between self perceived fatigue and spatiotemporal gait parameters at the end of a detailed task such as 6MWT. Therefore, “the system can be used by clinicians as a monitoring tool in their treatment and intervention programs to reduce the fatigue degree of MS patients. “

Editor: hfy

Leave a Reply

Your email address will not be published. Required fields are marked *