Related Experiment Video
Updated: Jun 20, 2026

Robotic Mirror Therapy System for Functional Recovery of Hemiplegic Arms
Published on: August 15, 2016
Automated prescription of therapeutic exercise for shoulder impingement syndrome using literature-driven rule
1Department of Physical Therapy, College of Software and Digital Healthcare Convergence, Yonsei University, 1, Yonseidae-gil, Heungeop-myeon, Wonju-si, Gangwon-do, 26493, South Korea.
Background:
Shoulder impingement syndrome (SIS) is a common musculoskeletal disorder that requires individualized therapeutic exercise prescription. However, it is challenging to provide individualized prescriptions based on the latest evidence.
Objectives:
This study aimed to develop and evaluate an automated literature-driven architecture that generates personalized therapeutic exercise prescriptions for SIS by integrating real-time clinical literature with patient-specific shoulder range of motion (ROM) data.
Design:
a cross-sectional secondary data analysis.
Methods:
A generative artificial intelligence (AI) model (Gemini 1.5) was used to retrieve, extract, and structure the evidence-based exercise recommendations from recent PubMed abstracts. Information linking specific shoulder ROM impairments to clinically relevant symptoms and therapeutic exercises was converted into executable decision rules. These rules were applied to anonymized shoulder ROM data from 35 patients with SIS. ROM deficits were calculated as percentage deviations from the normative values, prioritized by severity, and used to generate individualized exercise prescriptions.
Results:
The architecture successfully extracted six evidence-based decision rules associating distinct shoulder ROM limitations with the corresponding symptoms and targeted therapeutic exercises. When applied to patient data, the system generated personalized exercise prescriptions that reflected prioritized ROM impairments and included symptom descriptions and detailed exercise instructions based on current literature.
Conclusion:
This study demonstrated the feasibility of the literature-driven rule generation architecture for the automated prescription of personalized therapeutic exercises for SIS. By integrating real-time clinical evidence with individual biomechanical profiles, the proposed architecture offers a scalable and adaptive framework for evidence-based rehabilitation planning and has potential applicability in other musculoskeletal conditions.
More Related Videos
07:44Author Spotlight: Implementation of BIVA for Analyzing Disease Risk Factors in Patients with Low Body Cell Mass
Published on: July 14, 2023
04:49Author Spotlight: Enhancing Post-Stroke Upper Limb Rehabilitation with Robotic Technologies for Improved Motor Recovery and Functional Outcomes
Published on: September 6, 2024
Related Concept Videos
Pre-Procedural Guidelines for Assessing Blood Pressure
Generation of Action Potential in Skeletal Muscles
Like neurons, muscle cells are also regarded as excitable due to their capacity to change in response to stimuli, primarily due to voltage-gated ion channels embedded in their plasma membranes, which get activated by alterations in the cell's...