Exercise Prescription Generation Using Large Language Models (LLMs) with Motion Tracking Inputs

Gangmin Li; Xuming Bai

Exercise Prescription Generation Using Large Language Models (LLMs) with Motion Tracking Inputs

Gangmin Li
, Xuming Bai

Research output: Contribution to conference › Paper › peer-review

Abstract

This paper reports our first attempt using Large Language Models (LLMs) to generate personalized physical prescriptions using real-time motion tracking system inputs for gait analysis. We investigate the feasibility of such a system through relevance, accuracy, and potential health outcomes of the generated prescriptions compared to those created by humans. The study highlights the benefits, challenges, and future directions of integrating sensor-based real-time inputs into LLMs for generating physical prescriptions for physical treatment and medical rehabilitation.

Original language	English
Publication status	Published - Aug 2025
Event	The 30th International Conference on Automation and Computing (ICAC 2025) - University of Loughborough, Loughborough, United Kingdom Duration: 27 Aug 2025 → 29 Aug 2025 Conference number: 30th https://cacsuk.co.uk/icac/

Conference

Conference	The 30th International Conference on Automation and Computing (ICAC 2025)
Abbreviated title	ICAC 2025
Country/Territory	United Kingdom
City	Loughborough
Period	27/08/25 → 29/08/25
Internet address	https://cacsuk.co.uk/icac/

Keywords

Physical Prescription Generation
large language model
Motion Tracking Systems
Physical Therapy
Rehabilitation
Sensor-Based Data Integration

Cite this

@conference{9287270cce6d4b52b9fa90e004481356,

title = "Exercise Prescription Generation Using Large Language Models (LLMs) with Motion Tracking Inputs",

abstract = "This paper reports our first attempt using Large Language Models (LLMs) to generate personalized physical prescriptions using real-time motion tracking system inputs for gait analysis. We investigate the feasibility of such a system through relevance, accuracy, and potential health outcomes of the generated prescriptions compared to those created by humans. The study highlights the benefits, challenges, and future directions of integrating sensor-based real-time inputs into LLMs for generating physical prescriptions for physical treatment and medical rehabilitation.",

keywords = "Physical Prescription Generation, large language model, Motion Tracking Systems, Physical Therapy, Rehabilitation, Sensor-Based Data Integration",

author = "Gangmin Li and Xuming Bai",

year = "2025",

month = aug,

language = "English",

note = "The 30th International Conference on Automation and Computing (ICAC 2025), ICAC 2025 ; Conference date: 27-08-2025 Through 29-08-2025",

url = "https://cacsuk.co.uk/icac/",

}

TY - CONF

T1 - Exercise Prescription Generation Using Large Language Models (LLMs) with Motion Tracking Inputs

AU - Li, Gangmin

AU - Bai, Xuming

N1 - Conference code: 30th

PY - 2025/8

Y1 - 2025/8

N2 - This paper reports our first attempt using Large Language Models (LLMs) to generate personalized physical prescriptions using real-time motion tracking system inputs for gait analysis. We investigate the feasibility of such a system through relevance, accuracy, and potential health outcomes of the generated prescriptions compared to those created by humans. The study highlights the benefits, challenges, and future directions of integrating sensor-based real-time inputs into LLMs for generating physical prescriptions for physical treatment and medical rehabilitation.

AB - This paper reports our first attempt using Large Language Models (LLMs) to generate personalized physical prescriptions using real-time motion tracking system inputs for gait analysis. We investigate the feasibility of such a system through relevance, accuracy, and potential health outcomes of the generated prescriptions compared to those created by humans. The study highlights the benefits, challenges, and future directions of integrating sensor-based real-time inputs into LLMs for generating physical prescriptions for physical treatment and medical rehabilitation.

KW - Physical Prescription Generation

KW - large language model

KW - Motion Tracking Systems

KW - Physical Therapy

KW - Rehabilitation

KW - Sensor-Based Data Integration

M3 - Paper

T2 - The 30th International Conference on Automation and Computing (ICAC 2025)

Y2 - 27 August 2025 through 29 August 2025

ER -

Exercise Prescription Generation Using Large Language Models (LLMs) with Motion Tracking Inputs

Abstract

Conference

Keywords

Fingerprint

Cite this