AI in Rehabilitation: An Umbrella Review of Clinical Effectiveness and Safety

An umbrella review synthesizing evidence on AI and device-based rehabilitation interventions, authored by Nafisa Abdalla and published on March 18, 2026. The 17.6 KB document analyzes clinical effectiveness, real-world performance, safety, and equity across modalities and settings, based on a review of literature from biomedical, allied health, and engineering databases up to September 1, 2025.

Use Cases

• Benchmarking AI-enabled intervention performance against technology-assisted modalities based on the described synthesis of outcomes.
• Identifying research gaps in safety, usability, and equity reporting for home-based rehabilitation delivery.
• Informing clinical trial design for pragmatic, multi-site, assessor-blinded studies based on the review's methodological findings.
• Assessing the lab-to-clinic performance drop for AI models like brain-computer-interface classifiers and computer-vision movement evaluation.

Strengths

• License is CC-BY-4.0, permitting broad reuse and adaptation.
• The review uses a structured Population–Exposure–Outcome framework for evidence synthesis.
• Analysis distinguishes between AI-enabled (ML/DL) and technology-assisted interventions.
• Covers multiple outcome domains including impairment, activity, independence, usability, safety, equity, and economics.

Limitations

• The dataset is a 17.6 KB DOCX file, indicating a limited scope as a review document, not a primary data collection.
• Row and column counts are unknown, and no sample data is available for inspection.
• The description is detailed but column-level documentation for any underlying data is absent.

Provenance

Source

figshare, authored by Nafisa Abdalla.

Collection Method

Umbrella review of reviews using a Population–Exposure–Outcome framework, searching multiple databases.

Time Range

Literature reviewed from database inception to September 1, 2025.

Freshness

Last updated 2026-03-18 11:12:59.