Comparative Motor Pre-clinical Assessment in Parkinson’s Disease Using Supervised Machine Learning Approaches

This paper presents the results of a pilot study to assess the feasibility of using accelerometer data to estimate the severity of symptoms and motor complications in patients with Parkinson's disease. A support vector machine (SVM) classifier was implemented to estimate the severity of tremor, bradykinesia and dyskinesia from accelerometer data features. SVM-based estimates were compared with clinical scores derived via visual inspection of video recordings taken while patients performed a series of standardized motor tasks. The analysis of the video recordings was performed by clinicians trained in the use of scales for the assessment of the severity of Parkinsonian symptoms and motor complications. Results derived from the accelerometer time series were analyzed to assess the effect on the estimation of clinical scores of the duration of the window utilized to derive segments (to eventually compute data features) from the accelerometer data, the use of different SVM kernels and misclassification cost values, and the use of data features derived from different motor tasks. Results were also analyzed to assess which combinations of data features carried enough information to reliably assess the severity of symptoms and motor complications. Combinations of data features were compared taking into consideration the computational cost associated with estimating each data feature on the nodes of a body sensor network and the effect of using such data features on the reliability of SVM-based estimates of the severity of Parkinsonian symptoms and motor complications.

Background Neuroimaging studies help us better understand the pathophysiology and symptoms of Parkinson's disease (PD). In several of these studies, diffusion tensor imaging (DTI) was used to investigate structural changes in cerebral tissue. Although data have been provided as regards to specific brain areas, a whole brain meta-analysis is still missing. Methods We compiled 39 studies in this meta-analysis: 14 used fractional anisotropy (FA), 1 used mean diffusivity (MD), and 24 used both indicators. These studies comprised 1855 individuals, 1087 with PD and 768 healthy controls. Regions of interest were classified anatomically (subcortical structures; white matter; cortical areas; cerebellum). Our statistical analysis considered the disease effect size (DES) as the main variable; the heterogeneity index (I2) and Pearson's correlations between the DES and co-variables (demographic, clinical and MRI parameters) were also calculated. Results Our results showed that FA-DES and MD-DES were able to distinguish between patients and healthy controls. Significant differences, indicating degenerations, were observed within the substantia nigra, the corpus callosum, and the cingulate and temporal cortices. Moreover, some findings (particularly in the corticospinal tract) suggested opposite brain changes associated with PD. In addition, our results demonstrated that MD-DES was particularly sensitive to clinical and MRI parameters, such as the number of DTI directions and the echo time within white matter. Conclusions Despite some limitations, DTI appears as a sensitive method to study PD pathophysiology and severity. The association of DTI with other MRI methods should also be considered and could benefit the study of brain degenerations in PD.

Parkinsonism is a progressive motor disease that affects 1.5 million Americans and is the second most common neurodegenerative disease after Alzheimer’s. Typical neuropathological features of Parkinson’s disease (PD) include degeneration of dopaminergic neurons located in the pars compacta of the substantia nigra that project to the striatum (nigro-striatal pathway) and depositions of cytoplasmic fibrillary inclusions (Lewy bodies) which contain ubiquitin and α-synuclein. The cardinal motor signs of PD are tremors, rigidity, slow movement (bradykinesia), poor balance, and difficulty in walking (Parkinsonian gait). In addition to motor symptoms, non-motor symptoms that include autonomic and psychiatric as well as cognitive impairments are pressing issues that need to be addressed. Several different mechanisms play an important role in generation of Lewy bodies; endoplasmic reticulum (ER) stress induced unfolded proteins, neuroinflammation and eventual loss of dopaminergic neurons in the substantia nigra of mid brain in PD. Moreover, these diverse processes that result in PD make modeling of the disease and evaluation of therapeutics against this devastating disease difficult. Here, we will discuss diverse mechanisms that are involved in PD, neuroprotective and therapeutic strategies currently in clinical trial or in preclinical stages, and impart views about strategies that are promising to mitigate PD pathology.