Predicting motor, cognitive & functional impairment in Parkinson's

Abstract

Objective We recently demonstrated that 998 features derived from a simple 7‐minute smartphone test could distinguish between controls, people with Parkinson's and people with idiopathic Rapid Eye Movement sleep behavior disorder, with mean sensitivity/specificity values of 84.6‐91.9%. Here, we investigate whether the same smartphone features can be used to predict future clinically relevant outcomes in early Parkinson's. Methods A total of 237 participants with Parkinson's (mean (SD) disease duration 3.5 (2.2) years) in the Oxford Discovery cohort performed smartphone tests in clinic and at home. Each test assessed voice, balance, gait, reaction time, dexterity, rest, and postural tremor. In addition, standard motor, cognitive and functional assessments and questionnaires were administered in clinic. Machine learning algorithms were trained to predict the onset of clinical outcomes provided at the next 18‐month follow‐up visit using baseline smartphone recordings alone. The accuracy of model predictions was assessed using 10‐fold and subject‐wise cross validation schemes. Results Baseline smartphone tests predicted the new onset of falls, freezing, postural instability, cognitive impairment, and functional impairment at 18 months. For all outcome predictions AUC values were greater than 0.90 for 10‐fold cross validation using all smartphone features. Using only the 30 most salient features, AUC values greater than 0.75 were obtained. Interpretation We demonstrate the ability to predict key future clinical outcomes using a simple smartphone test. This work has the potential to introduce individualized predictions to routine care, helping to target interventions to those most likely to benefit, with the aim of improving their outcome.

A genome-wide association study coupled with machine learning approaches to identify influential demographic and genomic factors underlying Parkinson’s disease

“…Other top variables included constipation, loss of smell, dribbling of saliva, work in last 7 days, urgency to pass urine, engage in household activity, exercise in past 7 days, self-care, difficulty swallowing food or drink, talking or moving about in sleep, and unpleasant sensations in legs (Figure 1A). Multiple previous studies together identified most of these or very similar variables as significant variables associated with PD (Nielsen et al, 2017;Prashanth and Roy, 2018;Lo et al, 2019;Shah et al, 2020;Yu et al, 2022); this comprehensive list of top demographic/clinical variables identified in our study, based on their capability in predicting PD, adds further support to the influence of these factors in PD prediction. One of the top variables that has not been studied much is "unpleasant sensations in legs," which is ranked 14th among all the 139 demographic variables by both RF FI and ANN EG.…”

Disease Prediction Using Machine Learning on Smartphone-Based Eye, Skin, and Voice Data: Scoping Review

“…In another study, 237 participants diagnosed with Parkinson disease performed 7 smartphone-based tests, such as pronouncing “aaah” on the smartphone for as long as possible, pressing a button on the screen if it appears, pressing 2 alternate buttons on the screen, and holding the phone with their hand at rest or outstretched. Their balance and gait were also analyzed from the position of the smartphone [ 37 ]. The data obtained from the smartphone were used to train the machine learning algorithm using RF.…”

Digital Outcomes as Biomarkers of Disease Progression in Early Parkinson's Disease: A Systematic Review

“…Most (11/15) had an observational design, and only three studies were clinical trials for new investigational products. Sample sizes ranged from 7 to 388, except for three studies (Pagano et al 24 [n = 264], Lo et al 32 [n = 237] and Burq et al 20 [n = 388]), most had a sample size in the low double‐digit range. See Table 1 for details on study design and extracted data.…”