Gloria Vergara-Diaz

Recent machine learning advancements in sensor-based mobility analysis: Deep learning for Parkinson's disease assessment

The development of wearable sensors has opened the door for long-term assessment of movement disorders. However, there is still a need for developing methods suitable to monitor motor symptoms in and outside the clinic. The purpose of this paper was to investigate deep learning as a method for this monitoring. Deep learning recently broke records in speech and image classification, but it has not been fully investigated as a potential approach to analyze wearable sensor data. We collected data from ten patients with idiopathic Parkinsons disease using inertial measurement units. Several motor tasks were expert-labeled and used for classification. We specifically focused on the detection of bradykinesia. For this, we compared standard machine learning pipelines with deep learning based on convolutional neural networks. Our results showed that deep learning outperformed other state-of-the-art machine learning algorithms by at least 4.6 % in terms of classification rate. We contribute a discussion of the advantages and disadvantages of deep learning for sensor-based movement assessment and conclude that deep learning is a promising method for this field.

Complex Upper-Limb Movements Are Generated by Combining Motor Primitives that Scale with the Movement Size

The hand trajectory of motion during the performance of one-dimensional point-to-point movements has been shown to be marked by motor primitives with a bell-shaped velocity profile. Researchers have investigated if motor primitives with the same shape mark also complex upper-limb movements. They have done so by analyzing the magnitude of the hand trajectory velocity vector. This approach has failed to identify motor primitives with a bell-shaped velocity profile as the basic elements underlying the generation of complex upper-limb movements. In this study, we examined upper-limb movements by analyzing instead the movement components defined according to a Cartesian coordinate system with axes oriented in the medio-lateral, antero-posterior, and vertical directions. To our surprise, we found out that a broad set of complex upper-limb movements can be modeled as a combination of motor primitives with a bell-shaped velocity profile defined according to the axes of the above-defined coordinate system. Most notably, we discovered that these motor primitives scale with the size of movement according to a power law. These results provide a novel key to the interpretation of brain and muscle synergy studies suggesting that human subjects use a scale-invariant encoding of movement patterns when performing upper-limb movements.

Tai Chi for Reducing Dual-task Gait Variability, a Potential Mediator of Fall Risk in Parkinson’s Disease: A Pilot Randomized Controlled Trial:

Objectives To assess the feasibility and inform design features of a fully powered randomized controlled trial (RCT) evaluating the effects of Tai Chi (TC) in Parkinson’s disease (PD) and to select outcomes most responsive to TC assessed during off-medication states. Design Two-arm, wait-list controlled RCT. Settings Tertiary care hospital. Subjects Thirty-two subjects aged 40–75 diagnosed with idiopathic PD within 10 years. Interventions Six-month TC intervention added to usual care (UC) versus UC alone. Outcome Measures Primary outcomes were feasibility-related (recruitment rate, adherence, and compliance). Change in dual-task (DT) gait stride-time variability (STV) from baseline to 6 months was defined, a priori, as the clinical outcome measure of primary interest. Other outcomes included: PD motor symptom progression (Unified Parkinson’s Disease Rating Scale [UPDRS]), PD-related quality of life (PDQ-39), executive function (Trail Making Test), balance confidence (Activity-Specific Balance Confidence Scale, ABC), and Timed Up and Go test (TUG). All clinical assessments were made in the off-state for PD medications. Results Thirty-two subjects were enrolled into 3 sequential cohorts over 417 days at an average rate of 0.08 subjects per day. Seventy-five percent (12/16) in the TC group vs 94% (15/16) in the UC group completed the primary 6-month follow-up assessment. Mean TC exposure hours overall: 52. No AEs occurred during or as a direct result of TC exercise. Statistically nonsignificant improvements were observed in the TC group at 6 months in DT gait STV (TC [20.1%] vs UC [−0.1%] group [effect size 0.49; P = .47]), ABC, TUG, and PDQ-39. UPDRS progression was modest and very similar in TC and UC groups. Conclusions Conducting an RCT of TC for PD is feasible, though measures to improve recruitment and adherence rates are needed. DT gait STV is a sensitive and logical outcome for evaluating the combined cognitive-motor effects of TC in PD.

Poster 272 Risk Factors of Reamputation or Amputation of the Contralateral Lower Limb in Amputees with Dysvascular Disease

Disclosures: G. Vergara-Diaz: I Have No Relevant Financial Relationships To Disclose. Objective: To identify risk factors leading to reamputation or amputation of the contralateral lower limb in patients with a major lowerextremity amputation (MLEA) due to dysvascular disease. Design: Retrospective study Setting: We examined the medical records and ancillary files of adult patients who underwent a lower-limb amputation and were admitted at Spaulding Rehabilitation Hospital (SRH) for inpatient care from August 2013 to August 2014. Participants: We selected medical records of adults with a dysvascular MLEA, and excluded those of subjects who had an amputation due to complications following unrelated surgical procedures, joint disease, trauma, or cancer. Hence, the medical records of 59 subjects were included in the study. Interventions: None Main Outcome Measures: Ipsilateral reamputation or major amputation of the contralateral lower limb. Results or Clinical Course: Patients were 61.0 13.4 years old, mostly male (68%), white Caucasian (45%), with high prevalence of hypertension (85%), diabetes (81%), dyslipidemia (63%), smoking habit or history (69%), depression (69%) and evidence of coronary artery disease (40%), who underwent a below-knee amputation (83%). 27% of the patients were reamputated, 81% contra-lateral, 60% within a year from the first MLEA. Acute vascular event (P1⁄4.01), perioperative infection of the stump (P1⁄4.05), anxiety (P1⁄4.05), and evidence of peripheral disease in the contralateral limb (P1⁄4.05), were associated with a new amputation in univariate analysis. Conclusion: About 25% of the patients who are admitted to the Amputees Clinic are there due to a second reamputation, 80% of which after undergoing bilateral amputation. In about one third of the cases, the reamputation or amputation of the contralateral limb occurred within a year. Acute vascular event at the moment of the first MLEA was identified as the main risk factor associated with a new amputation.