Computer-based assessment of movement difficulties in Parkinson's disease

A Dual-Modal Attention-Enhanced Deep Learning Network for Quantification of Parkinson's Disease Characteristics

It is well known that most patients with Parkinson's disease (PD) have different degree of movement disorders, such as shuffling, festination and akinetic episodes, which could degenerate the life quality of PD patients. Therefore, it is very useful to develop a computerized tool to provide an objective evaluation of PD patients' gait. In this study, we implemented a novel gait evaluating approach to provide not only a binary classification of PD gaits and normal walking, but also a quantification of the PD gaits to relate them to the PD severity level. The proposed system is a dual-modal deep-learning-based model, where left and right gait is modeled separately by a convolutional neural network (CNN) followed by an attention-enhanced long short-term memory (LSTM) network. The left and right samples for model training and testing were segmented sequentially from multiple 1D vertical ground reaction force (VGRF) signals according to the detected gait cycle. Experimental results indicate that our model can provide state-of-the-art performance in terms of classification accuracy. It is expected that the proposed model can be a useful gait assistance to provide a quantitative evaluation of PD gaits with high confidence and accuracy if trained suitably.

Technologies Assessing Limb Bradykinesia in Parkinson's Disease

Background: The MDS-UPDRS (Movement Disorders Society – Unified Parkinson’s Disease Rating Scale) is the most widely used scale for rating impairment in PD. Subscores measuring bradykinesia have low reliability that can be subject to rater variability. Novel technological tools can be used to overcome such issues.

Objective: To systematically explore and describe the available technologies for measuring limb bradykinesia in PD that were published between 2006 and 2016.

Methods: A systematic literature search using PubMed (MEDLINE), IEEE Xplore, Web of Science, Scopus and Engineering Village (Compendex and Inspec) databases was performed to identify relevant technologies published until 18 October 2016.

Results: 47 technologies assessing bradykinesia in PD were identified, 17 of which offered home and clinic-based assessment whilst 30 provided clinic-based assessment only. Of the eligible studies, 7 were validated in a PD patient population only, whilst 40 were tested in both PD and healthy control groups. 19 of the 47 technologies assessed bradykinesia only, whereas 28 assessed other parkinsonian features as well. 33 technologies have been described in additional PD-related studies, whereas 14 are not known to have been tested beyond the pilot phase.

Conclusion: Technology based tools offer advantages including objective motor assessment and home monitoring of symptoms, and can be used to assess response to intervention in clinical trials or routine care. This review provides an up-to-date repository and synthesis of the current literature regarding technology used for assessing limb bradykinesia in PD. The review also discusses the current trends with regards to technology and discusses future directions in development.

Remotely Assessing Symptoms of Parkinson's Disease Using Videoconferencing: A Feasibility Study

Purpose. To evaluate the feasibility of assessing a person's symptoms of Parkinson's disease (PD) in their home using the videoconferencing technology they already possess, without a home visit. Method. Eleven participants with PD completed the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS) face-to-face and then via videoconferencing within a two-week period. Participants used free software and the computers and webcams available at their home to complete the videoconference assessment with a clinical rater scoring remotely. Clinical raters and participants provided feedback on the experience. Results. Excluding rigidity and postural stability, between zero and seven items could not be completed in the assessment of each participant (median 2.0, IQR 1.0-4.0). Between face-to-face and videoconference assessments, the median difference in scores was 3.0 (IQR 1.5-9.0). Content analysis of feedback identified the clinical raters' reasons why some scoring could not be completed and the participants' hope for future clinical application. Conclusions. In using free everyday technology available in participants' homes, MDS-UPDRS ratings could be obtained without an initial home visit; however some items were unable to be scored for some participants. Use of a protocol or technological advances are likely to reduce missing items.

Valproic Acid Neuroprotection in the 6-OHDA Model of Parkinson's Disease Is Possibly Related to Its Anti-Inflammatory and HDAC Inhibitory Properties

Parkinson's disease is a neurodegenerative disorder where the main hallmark is the dopaminergic neuronal loss. Besides motor symptoms, PD also causes cognitive decline. Although current therapies focus on the restoration of dopamine levels in the striatum, prevention or disease-modifying therapies are urgently needed. Valproic acid (VA) is a wide spectrum antiepileptic drug, exerting many biochemical and physiological effects. It has been shown to inhibit histone deacetylase which seems to be associated with the drug neuroprotective action. The objectives were to study the neuroprotective properties of VA in a model of Parkinson's disease, consisting in the unilateral striatal injection of the neurotoxin 6-OHDA. For that, male Wistar rats (250 g) were divided into the groups: sham-operated (SO), untreated 6-OHDA-lesioned, and 6-OHDA-lesioned treated with VA (25 or 50 mg/kg). Oral treatments started 24 h after the stereotaxic surgery and continued daily for 2 weeks, when the animals were subjected to behavioral evaluations (apomorphine-induced rotations and open-field tests). Then, they were sacrificed and had their mesencephalon, striatum, and hippocampus dissected for neurochemical (DA and DOPAC determinations), histological (Fluoro-Jade staining), and immunohistochemistry evaluations (TH, OX-42, GFAP, TNF-alpha, and HDAC). The results showed that VA partly reversed behavioral and neurochemical alterations observed in the untreated 6-OHDA-lesioned rats. Besides, VA also decreased neuron degeneration in the striatum and reversed the TH depletion observed in the mesencephalon of the untreated 6-OHDA groups. This neurotoxin increased the OX-42 and GFAP immunoreactivities in the mesencephalon, indicating increased microglia and astrocyte reactivities, respectively, which were reversed by VA. In addition, the immunostainings for TNF-alpha and HDAC demonstrated in the untreated 6-OHDA-lesioned rats were also decreased after VA treatments. These results were observed not only in the CA1 and CA3 subfields of the hippocampus, but also in the temporal cortex. In conclusion, we showed that VA partly reversed the behavioral, neurochemical, histological, and immunohistochemical alterations observed in the untreated 6-OHDA-lesioned animals. These effects are probably related to the drug anti-inflammatory activity and strongly suggest that VA is a potential candidate to be included in translational studies for the treatment of neurodegenerative diseases as PD.