Machine Learning Based Automatic Rating for Cardinal Symptoms of Parkinson

Objective: We developed and investigated the feasibility of a machine learning-based automated rating for the two cardinal symptoms of Parkinson’s disease (PD); resting tremor and bradykinesia.

Methods: Using Openpose, a deep-learning-based human pose estimation program, we analyzed video-clips for resting tremor and finger tapping of the bilateral upper limbs of 55 patients with PD (110 arms). Key motion parameters, including resting tremor amplitude and finger tapping speed, amplitude, and fatigue, were extracted to develop a machine learning-based automatic Unified Parkinson’s Disease Rating Scale (UPDRS) rating using support vector machine (SVM) method. To evaluate the performance of this model, we calculated weighted kappa and intra-class correlation coefficients (ICC) between the model and the gold standard rating by a movement disorder specialist who is trained and certified by the movement disorder society (MDS) for UPDRS rating. These values were compared to weighted kappa and ICC between a non-trained human rater and the gold standard rating.

Results: For resting tremors, the SVM model showed a very good to excellent reliability range with the gold standard rating (kappa 0.791; ICC 0.927), with both values higher than that of non-trained human rater (kappa 0.662; ICC 0.861). For finger tapping, the SVM model showed a very good reliability range with the gold standard rating (kappa 0.700 and ICC 0.793), which was comparable to non-trained human raters (kappa 0.627; ICC 0.797).

Conclusion: Machine learning-based algorithms that automatically rate PD cardinal symptoms are feasible, with more accurate results than non-trained human rating results.

Classification of Evidence: This study provides Class II evidence that machine learning-based automated rating of resting tremor and bradykinesia in people with PD has very good reliability compared to a rating by a movement disorder specialist.