Abstract
The global aging phenomenon has increased the number of individuals with age-related neurological movement disorders including Parkinson’s Disease (PD) and Essential Tremor (ET). Pathological Hand Tremor (PHT), which is considered among the most common motor symptoms of such disorders, can severely affect patients’ independence and quality of life. To develop advanced rehabilitation and assistive technologies, accurate estimation/prediction of nonstationary PHT is critical, however, the required level of accuracy has not yet been achieved. The lack of sizable datasets and generalizable modeling techniques that can fully represent the spectrotemporal characteristics of PHT have been a critical bottleneck in attaining this goal. This paper addresses this unmet need through establishing a deep recurrent model to predict and eliminate the PHT component of hand motion. More specifically, we propose a machine learning-based, assumption-free, and real-time PHT elimination framework, the PHTNet, by incorporating deep bidirectional recurrent neural networks. The PHTNet is developed over a hand motion dataset of 81 ET and PD patients collected systematically in a movement disorders clinic over 3 years. The PHTNet is the first intelligent systems model developed on this scale for PHT elimination that maximizes the resolution of estimation and allows for prediction of future and upcoming sub-movements.
Highlights
Unlike physiological tremor, which is identified with low amplitude vibrations occurring within the spectral range of 6 to 14 Hz8 and affects the performance of individuals in high precision tasks such as robotic surgery[9], Pathological Hand Tremor (PHT) represents higher amplitude motion occurring in the broader frequency range of 3–14 Hz10
From the results of the experiments over the validation and test sets presented in Figs. 5 and 9, and Table 2, we can clearly observe the superior performance of PHTNet in accurate estimation of the voluntary hand motion from pseudo-synthesized and real action tremor signals
The results clearly show the accuracy of PHTNet in estimating the voluntary component, when enough information is fed to the network
Summary
Unlike physiological tremor, which is identified with low amplitude vibrations occurring within the spectral range of 6 to 14 Hz8 and affects the performance of individuals in high precision tasks such as robotic surgery[9], PHT represents higher amplitude motion occurring in the broader frequency range of 3–14 Hz10. Existing methods including EBMFLC and WAKE, share a similar characteristic by assuming that the spectral contents of voluntary and involuntary components are distinct and one could be derived by removing the other from the measurement signal This assumption is not always realistic and has resulted in limited performance of the designed techniques and hindered their clinical translation. Besides concerns regarding the computational power and capacity of existing frameworks for an ultimate predictive model, there is a need for characterizing tremor based on a sizable inclusive dataset, that covers possible pathological variations causing diverse types of tremor signals in terms of spectrotemporal behavior, dynamic nature, temporal dependencies, and sub-movements Without such a data atlas, conservative and impractical assumptions would be considered to define a ground truth reference for designing and validating the techniques. The proposed PHTNet, which removes the lag and enhances the time resolution, provides a significant phase benefit, which is imperative for the control algorithm of rehabilitation and assistive technologies
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
