AI-based Physiotherapy Evaluation System for Range of Motion in Oral Cancer Patients

Validity and Reliability of an AI-based Physiotherapy Evaluation System for Oromandibular and Neck-Shoulder Range of Motion in Oral Cancer Patients

This study aims to evaluate the validity and reliability of a novel AI-based physiotherapy evaluation system for measuring oromandibular and neck-shoulder range of motion (ROM). Traditional ROM assessments rely on manual measurements, which may be influenced by rater experience and variability. The proposed AI system uses automated keypoint tracking to provide objective and standardized measurements.

In this cross-sectional study, healthy adult participants will perform standardized ROM tasks. Measurements obtained from the AI system will be compared with those from two independent raters using conventional clinical tools. Repeated measurements will be conducted to assess intra-rater and inter-rater reliability. The agreement between the AI system and human raters will be evaluated to determine the system's clinical applicability.

Study Overview

• Status: Recruiting
• Conditions: Oral Cancer; AI (Artificial Intelligence)

Detailed Description

This study is a cross-sectional measurement study designed to evaluate the reliability and concurrent validity of an AI-based physiotherapy evaluation system for assessing oromandibular and neck-shoulder range of motion (ROM). Participants will be healthy adults aged 20 to 70 years who meet predefined inclusion and exclusion criteria. After providing informed consent, participants will perform standardized movements, including mouth opening and cervical and shoulder ROM tasks.

Each participant will undergo three repeated measurements for each movement. ROM will be assessed using three methods: (1) an AI-based system utilizing real-time keypoint tracking and automated angle calculation, (2) manual measurement by Rater 1, and (3) independent manual measurement by Rater 2 using a goniometer or TheraBite ROM scale.

To minimize measurement bias and fatigue effects, the order of the three assessment methods will be randomized for each participant. Raters will be blinded to each other's measurements and to the AI-generated results.

The primary outcomes include inter-rater reliability and intra-rater reliability of the AI system, as well as agreement between AI-based and manual measurements. Reliability will be assessed using intraclass correlation coefficients (ICC), while agreement will be evaluated using Bland-Altman analysis and mean absolute error (MAE).

This study is expected to provide evidence supporting the clinical applicability of AI-based physiotherapy assessment tools, particularly for standardized and scalable musculoskeletal evaluations.

• Study Type: Observational
• Enrollment (Estimated): 20

Contacts and Locations

• Study Contact: Name: Yueh-Hsia Chen, Ph.D.; Phone Number: +886-2-33668133; Email: yuehhsiachen@ntu.edu.tw

Study Locations

• Taiwan
  • Taipei, Taiwan, 100
    • Recruiting
    • School and Graduate Institute of Physical Therapy, College of Medicine, National Taiwan University
    • Contact: Yueh-Hsia Chen, Ph.D

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of healthy adult volunteers aged 20 to 70 years recruited through non-probability sampling. Participants without a history of trismus, head and neck cancer, or musculoskeletal conditions affecting the head, neck, or shoulder regions are eligible. All participants are capable of following instructions and performing standardized movement tasks. This population is selected to establish baseline measurement performance and to evaluate the reliability and validity of the AI-based physiotherapy evaluation system under controlled conditions.

Description

Inclusion Criteria:

• Healthy adults aged 20 to 70 years
• No trismus
• No history of head, neck, or shoulder injury or surgery
• No history of head and neck cancer-related radiotherapy or chemotherapy

Exclusion Criteria:

• Inability to communicate or follow instructions
• Any condition that may affect movement performance

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Healthy group; Healthy adults aged between 20 and 70 years without a history of trismus, head, neck or shoulder injury or surgery, HNC-related radiotherapy or chemoradiotherapy were recruited.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Agreement Between AI and Manual Measurements	Agreement between AI-based and manual measurements assessed using Intraclass correlation coefficients (ICC) and Bland-Altman analysis	Baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Mean Absolute Error (MAE)	Average absolute difference between AI measurements and manual measurements	Baseline
Intra-rater reliability of human raters	Consistency of manual measurements by Rater 1 and Rater 2 across repeated trials using intraclass correlation coefficients (ICC)	Baselinte
Inter-rater reliability among all raters	Agreement among measurements obtained from the AI system, Rater 1, and Rater 2 will be assessed using intraclass correlation coefficients (ICC)	Baseline
Intra-rater reliability of AI system	Consistency of AI-based measurements across three repeated trials using intraclass correlation coefficients (ICC)	Baseline
Systematic measurement bias	Mean difference between AI-based and manual measurements	Baseline

Collaborators and Investigators

• Sponsor: National Taiwan University Hospital
• Investigators: Principal Investigator: Yueh-Hsia Chen, Ph.D., School and Graduate Institute of Physical Therapy, College of Medicine, National Taiwan University

Publications and helpful links

General Publications

• Agarwal P, Shiva Kumar HR, Rai KK. Trismus in oral cancer patients undergoing surgery and radiotherapy. J Oral Biol Craniofac Res. 2016 Nov;6(Suppl 1):S9-S13. doi: 10.1016/j.jobcr.2016.10.004. Epub 2016 Oct 22.
• Deb S, Islam MF, Rahman S, Rahman S. Graph Convolutional Networks for Assessment of Physical Rehabilitation Exercises. IEEE Trans Neural Syst Rehabil Eng. 2022;30:410-419. doi: 10.1109/TNSRE.2022.3150392. Epub 2022 Feb 23.

Study record dates

Study Major Dates

• Study Start (Actual): April 7, 2026
• Primary Completion (Estimated): May 31, 2026
• Study Completion (Estimated): December 31, 2027

Study Registration Dates

• First Submitted: March 29, 2026
• First Submitted That Met QC Criteria: March 29, 2026
• First Posted (Actual): April 3, 2026

Study Record Updates

• Last Update Posted (Actual): April 13, 2026
• Last Update Submitted That Met QC Criteria: April 6, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Physiotherapy; Artificial intelligence; Oral cancer; Range of motion; Maximum interincisal opening; Neck and shoulder
• Additional Relevant MeSH Terms: Mouth Diseases; Stomatognathic Diseases; Neoplasms by Site; Neoplasms; Head and Neck Neoplasms; Mouth Neoplasms
• Other Study ID Numbers: 202603023RIND

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Individual participant data (IPD) will not be shared due to privacy considerations and institutional regulations. Although the AI system processes de-identified keypoint data, the dataset may still contain information that could potentially be re-identified. Data sharing may be considered upon reasonable request and subject to institutional review and data protection policies.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No