Editorial: Contribution of artificial intelligence-based tools to the study of Parkinson's disease and other movement disorders

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social communication and the presence of restricted interests and repetitive behaviors. There have been recent concerns about increased prevalence, and this article seeks to elaborate on factors that may influence prevalence rates, including recent changes to the diagnostic criteria. The authors review evidence that ASD is a neurobiological disorder influenced by both genetic and environmental factors affecting the developing brain, and enumerate factors that correlate with ASD risk. Finally, the article describes how clinical evaluation begins with developmental screening, followed by referral for a definitive diagnosis, and provides guidance on screening for comorbid conditions. Show more

Parkinsonism, as a gradually progressive disorder, has a prodromal interval during which neurodegeneration has begun but cardinal manifestations have not fully developed. A systematic direct assessment of this interval has never been performed. Since patients with idiopathic REM sleep behaviour disorder are at very high risk of parkinsonism, they provide a unique opportunity to observe directly the development of parkinsonism. Patients with idiopathic REM sleep behaviour disorder in an ongoing cohort study were evaluated annually with several quantitative motor measures, including the Unified Parkinson's Disease Rating Scale, Purdue Pegboard, alternate-tap test and timed up-and-go. Patients who developed parkinsonism were identified from this cohort and matched according to age to normal controls. Their results on motor testing from the preceding years were plotted, and then assessed with regression analysis, to determine when markers first deviated from normal values. Sensitivity and specificity of quantitative motor markers for diagnosing prodromal parkinsonism were assessed. Of 78 patients, 20 developed parkinsonism. On regression analysis, the Unified Parkinson's Disease Rating Scale first intersected normal values at an estimated 4.5 years before diagnosis. Voice and face akinesia intersected earliest (estimated prodromal interval = 9.8 years), followed by rigidity (4.4 years), gait abnormalities (4.4 years) and limb bradykinesia (4.2 years). Quantitative motor tests intersected normal values at longer prodromal intervals than subjective examination (Purdue Pegboard = 8.6 years, alternate-tap = 8.2, timed up-and-go = 6.3). Using Purdue Pegboard and the alternate-tap test, parkinsonism could be detected with 71-82% sensitivity and specificity 3 years before diagnosis, whereas a Unified Parkinson's Disease Rating Scale score >4 identified prodromal parkinsonism with 88% sensitivity and 94% specificity 2 years before diagnosis. Removal of action tremor scores improved sensitivity to 94% and specificity to 97% at 2 years before diagnosis (cut-off >3). Although distinction between conditions was often difficult, prodromal dementia with Lewy bodies appeared to have a slower progression than Parkinson's disease (prodromal interval = 6.0 versus 3.8 years). Using a cut-off of Unified Parkinson's Disease Rating Scale >3 (excluding action tremor), 25% of patients with 'still-idiopathic' REM sleep behaviour disorder demonstrated evidence of possible prodromal parkinsonism. Therefore, using direct assessment of motor examination before parkinsonism in a REM sleep behaviour disorder, we have estimated a prodromal interval of ∼4.5 years on the Unified Parkinson's Disease Rating Scale; other quantitative markers may detect parkinsonism earlier. Simple quantitative motor measures may be capable of reliably detecting parkinsonism, even before a clinical diagnosis can be made by experienced movement disorders neurologists. Show more

Freezing of gait (FOG) is a serious gait disturbance, common in mid- and late-stage Parkinson’s disease, that affects mobility and increases fall risk. Wearable sensors have been used to detect and predict FOG with the ultimate aim of preventing freezes or reducing their effect using gait monitoring and assistive devices. This review presents and assesses the state of the art of FOG detection and prediction using wearable sensors, with the intention of providing guidance on current knowledge, and identifying knowledge gaps that need to be filled and challenges to be considered in future studies. This review searched the Scopus, PubMed, and Web of Science databases to identify studies that used wearable sensors to detect or predict FOG episodes in Parkinson’s disease. Following screening, 74 publications were included, comprising 68 publications detecting FOG, seven predicting FOG, and one in both categories. Details were extracted regarding participants, walking task, sensor type and body location, detection or prediction approach, feature extraction and selection, classification method, and detection and prediction performance. The results showed that increasingly complex machine-learning algorithms combined with diverse feature sets improved FOG detection. The lack of large FOG datasets and highly person-specific FOG manifestation were common challenges. Transfer learning and semi-supervised learning were promising for FOG detection and prediction since they provided person-specific tuning while preserving model generalization. Show more