A Machine-learning-based Method to Detect Degradation of Motor Control Stability with Applications to Diagnosis of Presymptomatic Parkinson's Disease

Abstract

Parkinsons disease (PD), a neuro-degenerative disorder, is often detected by onset of its motor symptoms such as rest tremor. Unfortunately, motor symptoms appear only when approximately 40%-60% of the dopaminergic neurons in the substantia nigra are lost. In most cases, by the time PD is clinically diagnosed, the disease may already have started 4 to 6 years beforehand. So there is a need for developing a test for detecting PD before the onset of the motor symptoms. This phase of PD is referred to as Presymptomatic PD (PPD). The motor symptoms of Parkinsons Disease are manifestations of instability in the sensorimotor system that develops gradually due to the neuro-degenerative process. In this paper, based on the above insight, we propose a new method that can potentially be used to detect degradation of motor control stability which can be employed for the detection of PPD. The proposed method tracks the tendency of a feedback control system to transition to an unstable state, and uses machine learning algorithm for its robust detection. This method is explored using simulations of a simple pendulum with PID controller as a conceptual representation for both healthy and PPD individuals. We also propose an example task with physiological measurements that can be used with this method and potentially be employed in a clinical setting. We present representative data collected through such a task, thereby demonstrating the feasibility to generate data for the proposed method. Author summaryParkinsons disease (PD) is a neuro-degenerative disorder that develops and progresses over several years. Currently, one is able to diagnose PD only after the appearance of motor symptoms (symptoms in movements of body parts), which unfortunately may be 4 to 6 years after the neuro-degeneration may have started. It has been shown that there are benefits to diagnosing PD at early stages, motivating the need to explore tools for diagnosing PD in the pre-symptomatic stage referred to as Presymptomatic Parkinsons disease (PPD). In this paper, a novel approach is explored that utilises the insight that the motor symptoms in PD may be seen as an instability in the feedback-control system that controls movements of body parts (sensory-motor loop). The proposed method uses a series of simple movement tasks performed by an individual in a clinic as the input to detect any gradual degradation of movement control that is leading to an instability, but before the instability and consequently the symptoms are manifested. This method is tested through extensive simulations and a potential experimental realisation with preliminary data. While a full-fledged validation will be undertaken as part of future work, initial results show promise and feasibility of further data collection.