Exploring the Feasibility of Computer Vision for Detecting Post-Stroke Compensatory Movements.

Compensatory movements are commonly observed post-stroke and can negatively affect long-term motor recovery. In this context, a system that monitors movement quality and provides feedback would be beneficial. In this study, we aimed to detect compensatory movements during seated reaching using a conventional tablet camera and an open-source markerless body pose tracking algorithm called MediaPipe [1]. We annotated compensatory movements of stroke patients per frame based on the comparison between the paretic and non-paretic arms. We trained a binary classification model using the XGBoost algorithm to detect compensatory movements, which showed an average accuracy of 0.92 (SD 0.07) in leave-one-trial-out cross-validation across four participants. Although we observed good model performance, we also encountered challenges such as missing landmarks and misalignment, when using MediaPipe Pose. This study highlights the feasibility of using near real-time compensatory movement detection with a simple camera system in stroke rehabilitation. More work is necessary to assess the generalizability of our approach across diverse groups of stroke survivors and fully implement near real-time compensatory movement detection on a mobile device.

3D-printed external cranial protection following decompressive craniectomy after brain injury: A pilot feasibility cohort study.

OBJECTIVES: 3D-printed (3DP) customized temporary cranial protection solutions following decompressive craniectomy (DC) are currently not widely practiced. A pilot trial of a 3DP customized head protection prototype device (HPPD) on 10 subjects was conducted during the subacute rehabilitation phase. MATERIALS AND METHODS: Subjects > 30 days post-DC with stable cranial flaps and healed wounds were enrolled. HPPD were uniquely designed based on individuals' CT scan, where the base conformed to the surface of the individual's skin covering the cranial defect, and the lateral surface three-dimensionally mirrored, the contralateral healthy head. Each HPPD was fabricated using the fused deposition modeling method. These HPPD were then fitted on subjects using a progressive wearing schedule and monitored over 1, 2, 4, 6 and 8 follow-up (FU) weeks. Outcomes during FU included; reported wearing time/day (hours), subjective pain, discomfort, pruritus, dislodgment, cosmesis ratings; and observed wound changes. The primary outcome was safety and tolerability without pain or wound changes within 30 minutes of HPPD fitting. RESULTS: In all, 10 enrolled subjects received 12 HPPDs [5/10 male, mean (SD) age 46 (14) years, mean (SD) duration post-DC 110 days (76)] and all subjects tolerated 30 minutes of initial HPPD fitting without wound changes. The mean (SD) HPPD mass was 61.2 g (SD 19.88). During 8 weeks of FU, no HPPD-related skin dehiscence was observed, while 20% (2/10) had transient skin imprints, and 80% (8/10) reported self-limiting pressure and pruritis. DISCUSSION: Findings from this exploratory study demonstrated preliminary feasibility and safety for a customized 3DP HPPD for temporary post-DC head protection over 8 weeks of follow-up. Monitoring and regular rest breaks during HPPD wear were important to prevent skin complications. CONCLUSION: This study suggests the potential for wider 3DP technology applications to provide cranial protection for this vulnerable population.

The Validity of the Montreal Cognitive Assessment for Moderate to Severe Traumatic Brain Injury Patients: A Pilot Study.

OBJECTIVE: The aim of the study was to pilot the use of Montreal Cognitive Assessment as a quick clinical screen for cognitive assessment in traumatic brain injury patients. DESIGN: The study recruited 61 participants with moderate to severe traumatic brain injury presenting to a tertiary rehabilitation center under the Brain Injury Program. A Montreal Cognitive Assessment questionnaire and neuropsychological battery (Repeatable Battery for the Assessment of Neuropsychological Status and Color Trails Test) were administered to participants who had completed inpatient rehabilitation. RESULTS: Receiver operating characteristic analysis for the Montreal Cognitive Assessment revealed an optimal balance of sensitivity and specificity at 24/25 to discriminate participants who were classified as less than 5th centile on the Total Scale Index on the Repeatable Battery for the Assessment of Neuropsychological Status. This achieved a sensitivity, specificity, PPV, and NPV of 73.9%, 86.5%, 77.3%, and 84.2%, respectively. Receiver operating characteristic analysis for the trail making subtest of the Montreal Cognitive Assessment achieved a sensitivity, specificity, PPV, and NPV of 79.4%, 74.1%, 79.4%, and 74.1% in identifying patients classified as less than 5th centile on Color Trail Test part 2. CONCLUSIONS: The use of Montreal Cognitive Assessment displayed good validity in identifying patients with clinically significant impairment on a standard neuropsychological assessment battery in the study population. However, it may lack sensitivity for estimating mild levels of impairment.