Detecting Parkinsonian Tremor From IMU Data Collected in-the-Wild Using Deep Multiple-Instance Learning.

Parkinson's Disease (PD) is a slowly evolving neurological disease that affects about [Formula: see text] of the population above 60 years old, causing symptoms that are subtle at first, but whose intensity increases as the disease progresses. Automated detection of these symptoms could offer clues as to the early onset of the disease, thus improving the expected clinical outcomes of the patients via appropriately targeted interventions. This potential has led many researchers to develop methods that use widely available sensors to measure and quantify the presence of PD symptoms such as tremor, rigidity and braykinesia. However, most of these approaches operate under controlled settings, such as in lab or at home, thus limiting their applicability under free-living conditions. In this work, we present a method for automatically identifying tremorous episodes related to PD, based on IMU signals captured via a smartphone device. We propose a Multiple-Instance Learning approach, wherein a subject is represented as an unordered bag of accelerometer signal segments and a single, expert-provided, tremor annotation. Our method combines deep feature learning with a learnable pooling stage that is able to identify key instances within the subject bag, while still being trainable end-to-end. We validate our algorithm on a newly introduced dataset of 45 subjects, containing accelerometer signals collected entirely in-the-wild. The good classification performance obtained in the conducted experiments suggests that the proposed method can efficiently navigate the noisy environment of in-the-wild recordings.

Smartwatch-based Activity Analysis During Sleep for Early Parkinson's Disease Detection.

Parkinson's Disease (PD) is the second most common neurodegenerative disorder with the non-motor symptoms preceding the motor impairment that is needed for clinical diagnosis. In the current study, an angle-based analysis that processes activity data during sleep from a smartwatch for quantification of sleep quality, when applied on controls and PD patients, is proposed. Initially, changes in their arm angle due to activity are captured from the smartwatch triaxial accelerometry data and used for the estimation of the corresponding binary state (awake/sleep). Then, sleep metrics (i.e., sleep efficiency index, total sleep time, sleep fragmentation index, sleep onset latency, and wake after sleep onset) are computed and used for the discrimination between controls and PD patients. A process of validation of the proposed approach when compared with the PSG-based ground truth in an in-the-clinic setting, resulted in comparable state estimation. Moreover, data from 15 early PD patients and 11 healthy controls were used as a test set, including 1,376 valid sleep recordings in-the-wild setting. The univariate analysis of the extracted sleep metrics achieved up to 0.77 AUC in early PD patients vs. healthy controls classification and exhibited a statistically significant correlation (up to 0.46) with the clinical PD Sleep Scale 2 counterpart Items. The findings of the proposed method show the potentiality to capture non-motor behavior from users' nocturnal activity to detect PD in the early stage.

Multiple-Instance Learning for In-The-Wild Parkinsonian Tremor Detection.

Parkinson's Disease (PD) is a neurodegenerative disorder that manifests through slowly progressing symptoms, such as tremor, voice degradation and bradykinesia. Automated detection of such symptoms has recently received much attention by the research community, owing to the clinical benefits associated with the early diagnosis of the disease. Unfortunately, most of the approaches proposed so far, operate under a strictly laboratory setting, thus limiting their potential applicability in real world conditions. In this work, we present a method for automatically detecting tremorous episodes related to PD, based on acceleration signals. We propose to address the problem at hand, as a case of Multiple-Instance Learning, wherein a subject is represented as an unordered bag of signal segments and a single, expert-provided, ground-truth. We employ a deep learning approach that combines feature learning and a learnable pooling stage and is trainable end-to-end. Results on a newly introduced dataset of accelerometer signals collected in-the-wild confirm the validity of the proposed approach.

Early Parkinson's Disease Detection via Touchscreen Typing Analysis using Convolutional Neural Networks.

Parkinson's Disease (PD) is the second most common neurodegenerative disorder worldwide, causing both motor and non-motor symptoms. In the early stages, symptoms are mild and patients may ignore their existence. As a result, they do not undergo any related clinical examination; hence delaying their PD diagnosis. In an effort to remedy such delay, analysis of data passively captured from user's interaction with consumer technologies has been recently explored towards remote screening of early PD motor signs. In the current study, a smartphone-based method analyzing subjects' finger interaction with the smartphone screen is developed for the quantification of fine-motor skills decline in early PD using Convolutional Neural Networks. Experimental results from the analysis of keystroke typing in-the-clinic data from 18 early PD patients and 15 healthy controls have shown a classification performance of 0.89 Area Under the Curve (AUC) with 0.79/0.79 sensitivity/specificity, respectively. Evaluation of the generalization ability of the proposed approach was made by its application on typing data arising from a separate self-reported cohort of 27 PD patients' and 84 healthy controls' daily usage with their personal smartphones (data in-the-wild), achieving 0.79 AUC with 0.74/0.78 sensitivity/specificity, respectively. The results show the potentiality of the proposed approach to process keystroke dynamics arising from users' natural typing activity to detect PD, which contributes to the development of digital tools for remote pathological symptom screening.

The cortical excitability profile of patients with the G209A SNCA mutation versus patients with sporadic Parkinson's disease: A transcranial magnetic stimulation study.

Mutations in the α-synuclein gene are a rare cause of Parkinson's disease. We investigated, by single-pulse TMS, the cortical excitability profile of nine α-synuclein patients in comparison with 24 idiopathic PD patients, subdivided into "akinetic" (n=17) and "tremor-dominant" (n=7) subgroups. The comparative assessment of rest motor threshold, active MEP and Silent Period Input/Output curves indicated that the cortical excitability of α-Synuclein patients is similar to patients with the "akinetic" form of PD. Both groups of patients exhibited differences in excitatory and inhibitory brain circuits from "tremor-dominant" patients indicating that varying clinical phenotypes are associated with differential profiles of corticospinal excitability.

Cognitive event-related potentials in multiple sclerosis: Correlation with MRI and neuropsychological findings.

BACKGROUND: Cognitive event-related potentials (ERPs) have been previously correlated with T2 lesion load (Τ2LL) in patients with multiple sclerosis (MS). It is currently unknown, however, whether ERPs also correlate with brain atrophy or the presence of T1 hypointense lesions ("black holes") which reflect tissue destruction and axonal loss. The primary aim of the current study is to explore the effect of neuroradiological parameters such as brain atrophy, T1 and T2 lesion load on auditory ERPs in MS patients. In addition, we correlated cognitive impairment with neurophysiological (ERP) and neuroradiological (MRI) variables and investigated whether a combination of ERP and MRI parameters is capable of distinguishing patients suffering from secondary progressive (SP), primary progressive (PP) and relapsing-remitting (RR) MS. MATERIALS AND METHODS: The study sample consisted of fifty nine MS patients (mean age±SD: 37.82±1.38 years; average disease duration: 6.76±5.3 years) and twenty six age-matched controls (mean age±SD: 41.42±15.39 years). The patients' EDSS and NRS scores were 3.77±2.14 (range: 1-7.5) and 75.88±11.99 (range: 42-94) respectively. ERPs were recorded using the auditory "odd-ball" paradigm. T1 and T2 lesions were automatically segmented using an edge-finding tool and total lesion volumes were calculated. MRI measures of brain atrophy included third ventricle width (THIRDVW) and the ratio of mid-sagittal corpus callosum area to the mid-sagittal intracranial skull surface area (CC/MISS). Statistical analysis was performed using multiple regression, principal component (PCA) and discriminant analysis. RESULTS: T1 lesion load emerged as the most significant predictor of P300 and N200 latency. The rest of the endogenous ERPs parameters (P300 amplitude, N200 amplitude) were not significantly correlated with the MRI variables. PCA of pooled neuroradiological and neurophysiological markers suggested that four components accounted for 64.6% of the total variability. Discriminant analysis based on ERP & MRI markers classified correctly 79.63% of patients in RR, PP and SP subgroups. CONCLUSION: T1 lesion load is the most significant MRI correlate of auditory ERPs in MS patients. Importantly, ERPs in combination with MRI variables can be usefully employed for distinguishing patients with different subtypes of MS.

[Findings from molecular imaging with SPET camera and 123I-ioflupane in the differential diagnosis of Parkinsonism and essential tremor].

The aim of the present study was to evaluate the use of the radiopharmaceutical 123I-ioflupane in the diagnosis and differential diagnosis of Parkinsonism (P) and essential tremor (ET). Forty-three consecutive patients, aged 35-72 years, presenting symptoms and signs compatible with P, plus 11 normal volunteers, aged 40-60 years, were enrolled for the study. The radiopharmaceutical was injected iv in a dose of 185 MBq and tomographic acquisition in a single-headed Pegasys gamma-camera (ADAC, USA), 3-4 hours post injection was performed in order to evaluate the activity of the presynaptic nigro-striatal dopaminergic transporter. After reconstruction and reorientation, semiquantitative analysis was performed evaluating counts/pixel: a) in the striatum and its parts (caudate nucleus and putamen) of both hemispheres and b) in the visual cortex representing non specific binding. According to our results, all 21 individuals with ET were correctly evaluated with this method, whilst 21/22 patients were diagnosed as having P. No statistical difference concerning the binding of the radioligand to the striatum and its parts was found between normal volunteers and patients with ET. Based on the present results in 21 of our patients, the diagnosis and treatment procedure were changed, while in the remaining 22 patients diagnosis and treatment were confirmed. According to our data, as well as to the data from others, molecular imaging (SPET) with 123I-ioflupane can properly differentiate individuals with ET from those having P, in order to avoid an unnecessary use of drugs that may even cause side effects. All our patients were re-examined after eight months. At that time the above results and the treatment that was given to them meanwhile, were positively evaluated.