This study aims to develop automated diagnostic tools to aid in the identification of rapid-eye-movement (REM) sleep behaviour disorder (RBD). Those diagnosed with RBD enact their dreams and therefore present an abnormal characteristic of movement during REM sleep. Several methods have been proposed for RBD detection that use electromyogram (EMG) recordings and manually annotated sleep stages to objectively quantify abnormal REM movement. In this work we further develop these proven techniques with additional features that incorporate the relationship of muscle movement between sleep stages and general sleep architecture. Performance is evaluated using polysomnography (PSG) recordings from 43 aged-matched healthy controls and subjects diagnosed with RBD obtained from multiple institutions and publicly available resources. Using a random forest classifier with established and additional features, the performance of RBD detection was shown to improve upon established metrics (achieving 88% accuracy, 91% sensitivity, and 86% specificity).
\n \n\n \n \nCurrent sleep medicine relies on the supervised analysis of polysomnographic measurements, comprising amongst others electroencephalogram (EEG), electromyogram (EMG), and electrooculogram (EOG) signals. Convolutional neural networks (CNN) provide an interesting framework to automated classification of sleep based on these raw waveforms. In this study, we compare existing CNN approaches to four databases of pathological and physiological subjects. The best performing model resulted in Cohen's Kappa of $\\kappa = 0 .75$ on healthy subjects and $\\kappa = 0 .64$ on patients suffering from a variety of sleep disorders. Further, we show the advantages of additional sensor data (i.e., EOG and EMG). Deep learning approaches require a lot of data which is scarce for less prevalent diseases. For this, we propose a transfer learning procedure by pretraining a model on large public data and fine-tune this on each subject from a smaller dataset. This procedure is demonstrated using a private REM Behaviour Disorder database, improving sleep classification by 24.4%.
\n \n\n \n \nThe parietal avoiding-rejection behaviour syndrome, first described by Denny-Brown in the rhesus monkey, has been reported only rarely in humans. Here, we describe a patient with rejection behaviour in the setting of progressive cognitive decline accompanied by cortical myoclonus.
\n \n\n \n \nOBJECTIVES: To investigate whether proton magnetic resonance spectroscopy (1H MRS) can detect cortical dysfunction in non-demented patients with Parkinson's disease, and to correlate changes with cognitive function on formal neuropsychological testing. METHODS: Multivoxel 1H MRS was performed in 17 patients with levodopa treated idiopathic Parkinson's disease with out clinical dementia, and 10 age match ed control subjects. Measurements of N-acetylaspartate (NAA)/choline (Cho), NAA/creatine+phosphocreatine (Cr), and Cho/Cr were obtained from right and left temporoparietal cortex and occipital cortex. Fourteen patients with Parkinson's disease underwent a full battery of neuropsychological testing including performance and verbal subtests of the WAIS-R, Boston naming test, FAS test, and California verbal learning test. RESULTS: There were significant temporoparietal cortex reductions in NAA/Cr ratios in right and left averaged spectra of the patients with Parkinson's disease (p=0.012 after Bonferroni correction) and in spectra contralateral to the worst clinically affected limbs of the patients with Parkinson's disease compared with controls (p = 0.003 after Bonferroni correction). There was a significant correlation between reduction in NAA/Cr ratios and measures of global cognitive decline, occurring independently of motor impairment (p=0.019). CONCLUSIONS: This study suggests that 1H MRS can detect temporoparietal cortical dysfunction in non-demented patients with Parkinson's disease. Further longitudinal studies are needed to investigate whether these 1H MRS changes are predictive of future cognitive impairment in the subset of patients with Parkinson's disease who go on to develop dementia, or occur as part of the normal Parkinson's disease process.
\n \n\n \n \nParkinson's disease (PD) is the second most common neurodegenerative disease affecting millions worldwide. Bespoke subject-specific treatment (medication or deep brain stimulation (DBS)) is critical for management, yet depends on precise assessment cardinal PD symptoms - bradykinesia, rigidity and tremor. Clinician diagnosis is the basis of treatment, yet it allows only a cross-sectional assessment of symptoms which can vary on an hourly basis and is liable to inter- and intra-rater subjectivity across human examiners. Automated symptomatic assessment has attracted significant interest to optimise treatment regimens between clinician visits, however, no wearable has the capacity to simultaneously assess all three cardinal symptoms. Challenges in the measurement of rigidity, mapping muscle activity out-of-clinic and sensor fusion have inhibited translation. In this study, we address all through a novel wearable sensor system and machine learning algorithms. The sensor system is composed of a force-sensor, three inertial measurement units (IMUs) and four custom mechanomyography (MMG) sensors. The system was tested in its capacity to predict Unified Parkinson's Disease Rating Scale (UPDRS) scores based on quantitative assessment of bradykinesia, rigidity and tremor in PD patients. 23 PD patients were tested with the sensor system in parallel with exams conducted by treating clinicians and 10 healthy subjects were recruited as a comparison control group. Results prove the system accurately predicts UPDRS scores for all symptoms (85.4% match on average with physician assessment) and discriminates between healthy subjects and PD patients (96.6% on average). MMG features can also be used for remote monitoring of severity and fluctuations in PD symptoms out-of-clinic. This closed-loop feedback system enables individually tailored and regularly updated treatment, facilitating better outcomes for a very large patient population.
\n \n\n \n \nAttention-deficit/hyperactivity disorder (ADHD) and Parkinson's disease (PD) involve pathological changes in brain structures such as the basal ganglia, which are essential for the control of motor and cognitive behavior and impulsivity. The cause of ADHD and PD remains unknown, but there is increasing evidence that both seem to result from a complicated interplay of genetic and environmental factors affecting numerous cellular processes and brain regions. To explore the possibility of common genetic pathways within the respective pathophysiologies, nine ADHD candidate single nucleotide polymorphisms (SNPs) in seven genes were tested for association with PD in 5333 cases and 12,019 healthy controls: one variant, respectively, in the genes coding for synaptosomal-associated protein 25\u00a0k (SNAP25), the dopamine (DA) transporter (SLC6A3; DAT1), DA receptor D4 (DRD4), serotonin receptor 1B (HTR1B), tryptophan hydroxylase 2 (TPH2), the norepinephrine transporter SLC6A2 and three SNPs in cadherin 13 (CDH13). Information was extracted from a recent meta-analysis of five genome-wide association studies, in which 7,689,524 SNPs in European samples were successfully imputed. No significant association was observed after correction for multiple testing. Therefore, it is reasonable to conclude that candidate variants implicated in the pathogenesis of ADHD do not play a substantial role in PD.
\n \n\n \n \nThe catechol-O-methyltranferase (COMT) is one of the main enzymes that metabolise dopamine in the brain. The Val158Met polymorphism in the COMT gene (rs4680) causes a trimodal distribution of high (Val/Val), intermediate (Val/Met) and low (Met/Met) enzyme activity. We tested whether the Val158Met polymorphism is a modifier of the age at onset (AAO) in Parkinson's disease (PD). The rs4680 was genotyped in a total of 16 609 subjects from five independent cohorts of European and North American origin (5886 patients with PD and 10 723 healthy controls). The multivariate analysis for comparing PD and control groups was based on a stepwise logistic regression, with gender, age and cohort origin included in the initial model. The multivariate analysis of the AAO was a mixed linear model, with COMT genotype and gender considered as fixed effects and cohort and cohort-gender interaction as random effects. COMT genotype was coded as a quantitative variable, assuming a codominant genetic effect. The distribution of the COMT polymorphism was not significantly different in patients and controls (p=0.22). The Val allele had a significant effect on the AAO with a younger AAO in patients with the Val/Val (57.1\u00b113.9, p=0.03) than the Val/Met (57.4\u00b113.9) and the Met/Met genotypes (58.3\u00b113.5). The difference was greater in men (1.9 years between Val/Val and Met/Met, p=0.007) than in women (0.2 years, p=0.81). Thus, the Val158Met COMT polymorphism is not associated with PD in the Caucasian population but acts as a modifier of the AAO in PD with a sexual dimorphism: the Val allele is associated with a younger AAO in men with idiopathic PD.
\n \n\n \n \n