Predicting Multiple Sclerosis from Gait Dynamics Using an Instrumented Treadmill – A Machine Learning Approach

Multiple Sclerosis (MS) is one of the most common neurological conditions worldwide whose prevalence is now greatest among people 50-60 years of age. While clinical presentations of MS are highly heterogeneous, mobility limitations is one of the most frequent symptoms. The aims of this study were to examine MS and disability related changes in spatiotemporal and kinetic gait features after normalization; and evaluate the effectiveness of a gait data-based machine learning (ML) framework for MS prediction (GML4MS). Methods: In this study, gait data during self-paced walking on an instrumented treadmill from 20 persons with MS and 20 age, weight, height and gender-matched healthy older adults (HOA) were obtained. We explored two normalization strategies, namely size-N (standard body size-based normalization) and regress-N (regression-based normalization using scaling factors derived by regressing gait features on multiple subject demographics) to minimize the dependency of derived gait features on the subject demographics; and proposed GML4MS, a ML based methodology to classify individual strides of older persons with MS (PwMS) from healthy controls, so as to generalize across different walking tasks and subjects after gait normalization. Results: We observed that regress-N improved the accuracy of identifying pathological gait using ML when compared to size-N. When generalizing from comfortable walking to walking while talking, gradient boosting machine achieved the optimal subject classification accuracy and AUC of 94.3% and 1.0, respectively and for subject generalization, a multilayer perceptron resulted in the best accuracy and AUC of 80% and 0.86, respectively, both with regress-N normalized data. Conclusion: The integration of gait data and ML to predict MS may provide a viable patient-centric approach to aid clinicians in disease monitoring and relapse treatment. This work is the first attempt to employ and demonstrate the potential of ML for this domain. Significance: The results of this study have future implications for the way regression normalized gait features may be clinically used to design ML-based disease prediction strategies and monitor disease progression in PwMS.