Search

Abstract BackgroundWith the increasing availability of data, computing resources, and easier-to-use software libraries, machine learning (ML) is increasingly used in disease detection and prediction, including for Parkinson disease (PD). Despite the large number of studies published every year, very few ML systems have been adopted for real-world use. In particular, a lack of external validity may result in poor performance of these systems in clinical practice. Additional methodological issues in ML design and reporting can also hinder clinical adoption, even for applications that would benefit from such data-driven systems. ObjectiveTo sample the current ML practices in PD applications, we conducted a systematic review of studies published in 2020 and 2021 that used ML models to diagnose PD or track PD progression. MethodsWe conducted a systematic literature review in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines in PubMed between January 2020 and April 2021, using the following exact string: “Parkinson’s” AND (“ML” OR “prediction” OR “classification” OR “detection” or “artificial intelligence” OR “AI”). The search resulted in 1085 publications. After a search query and review, we found 113 publications that used ML for the classification or regression-based prediction of PD or PD-related symptoms. ResultsOnly 65.5% (74/113) of studies used a holdout test set to avoid potentially inflated accuracies, and approximately half (25/46, 54%) of the studies without a holdout test set did not state this as a potential concern. Surprisingly, 38.9% (44/113) of studies did not report on how or if models were tuned, and an additional 27.4% (31/113) used ad hoc model tuning, which is generally frowned upon in ML model optimization. Only 15% (17/113) of studies performed direct comparisons of results with other models, severely limiting the interpretation of results. ConclusionsThis review highlights the notable limitations of current ML systems and techniques that may contribute to a gap between reported performance in research and the real-life applicability of ML models aiming to detect and predict diseases such as PD.

We created an overall assessment metric using a deep learning autoencoder to directly compare clinical outcomes in a comparison of lower limb amputees using two different prosthetic devices—a mechanical knee and a microprocessor-controlled knee. Eight clinical outcomes were distilled into a single metric using a seven-layer deep autoencoder, with the developed metric compared to similar results from principal component analysis (PCA). The proposed methods were used on data collected from ten participants with a dysvascular transfemoral amputation recruited for a prosthetics research study. This single summary metric permitted a cross-validated reconstruction of all eight scores, accounting for 83.29% of the variance. The derived score is also linked to the overall functional ability in this limited trial population, as improvements in each base clinical score led to increases in this developed metric. There was a highly significant increase in this autoencoder-based metric when the subjects used the microprocessor-controlled knee (p < 0.001, repeated measures ANOVA). A traditional PCA metric led to a similar interpretation but captured only 67.3% of the variance. The autoencoder composite score represents a single-valued, succinct summary that can be useful for the holistic assessment of highly variable, individual scores in limited clinical datasets.

Parkinson's disease (PD) is one of the most common neurodegenerative disorders, but it is often diagnosed after the majority of dopaminergic cells are already damaged. It is critical to develop biomarkers to identify the disease as early as possible for early intervention. PD patients appear to have an altered pupillary response consistent with an abnormality in photoreceptive retinal ganglion cells. Tracking the pupil size manually is a tedious process and offline automated systems can be prone to errors that may require intervention; for this reason in this work we describe a system for pupil size estimation with a user interface to allow rapid adjustment of parameters and extraction of pupil parameters of interest for the present study. We implemented a user-friendly system designed for clinicians to automate the process of tracking the pupil diameter to measure the post-illumination pupillary response (PIPR), permit manual corrections when needed, and continue automation after correction. Tracking was automated using a Kalman filter estimating the pupil center and diameter over time. The resulting system was tested on a PD classification task in which PD subjects are known to have similar responses for two wavelengths of light. The pupillary response is measured in the contralateral eye to two different light stimuli (470 and 610 nm) for 19 PD and 10 control subjects. The measured Net PIPR indicating different responsiveness to the wavelengths was 0.13 mm for PD subjects and 0.61 mm for control subjects, demonstrating a highly significant difference (p < 0.001). Net PIPR has the potential to be a biomarker for PD, suggesting further study to determine clinical validity.