Feasibility and safety of a remotely administered Five Times Sit-to-Stand test and its relationship to capacity and performance in stroke survivors

Title: Dataset for “Feasibility and safety of a remotely administered Five Times Sit-to-Stand test and its relationship to capacity and performance in stroke survivors” Description:This repository contains the de-identified dataset and analysis materials for a study examining the feasibility and safety of remote functional mobility assessment using wearable sensors and an AI-based video analysis pipeline. The study evaluates four complementary measures of sit-to-stand performance:(1) Capacity – Five Times Sit-to-Stand Test (FTSTS) assessed in the clinic using APDM® wearable sensors;(2) Capacity (AI) – FTSTS recorded in the clinic using smartphone video and processed through an AI-based backend pipeline;(3) Performance – real-world sit-to-stand activity measured using ActivPAL® over three consecutive days;(4) Performance (AI) – sit-to-stand performance recorded in the home environment using smartphone video and processed through the same AI-based pipeline. The dataset includes participants with stroke and age-matched controls. Variables include demographic characteristics, clinical measures (e.g., MoCA, Fugl-Meyer), capacity and performance metrics, and feasibility and safety indicators. Files included: data_clean.csv: de-identified dataset used for analysis codebook.pdf: detailed variable definitions and coding scheme analysis_script.R: statistical analysis script used to generate results Data processing and analysis:Statistical analyses include descriptive statistics, between-group comparisons, correlation analyses, Bland–Altman agreement analysis, and sensitivity analyses using multiple approaches to handle missing data. Ethics and data sharing:All data have been de-identified in accordance with institutional review board (IRB) approval and ethical guidelines. Raw video recordings are not included due to privacy and confidentiality restrictions. Funding:This work was supported by a Pilot Project Award from the Restore Center at Stanford University (Grant No. 226352158A). The Restore Center is funded by the National Institutes of Health (Grant No. P2CHD101913). Reuse:The dataset is provided for research and educational purposes. Users are encouraged to cite the associated publication when using these data. Contact:Camila Torriani-Pasin, University of Texas at El Paso [e-mail]