This study aims to evaluate the feasibility, safety, and acceptability of a newly developed artificial intelligence (AI)-assisted physiotherapy system for oromandibular and neck-shoulder range of motion (ROM) in patients who have undergone treatment for oral cancer.

In this single-group, prospective, non-randomized pilot study, recruited participants will receive 4 to 6 weeks of AI-assisted physiotherapy. Participants will undergo a comprehensive clinical evaluation at baseline and post-intervention. During the intervention period, the AI system will perform a daily automated assessment to dynamically generate and adjust personalized exercise programs. Participants will perform these prescribed programs 4 to 6 times daily. Pre- and post-intervention changes, along with key feasibility parameters, acceptability, and safety metrics, will be statistically analyzed to inform future definitive trials.

This is a single-group, prospective, non-randomized pilot feasibility study. Oral cancer patients with trismus or neck-shoulder impairment will be trained to exercise their jaw or neck-shoulder muscles using an AI-assisted physiotherapy assessment system, which is reliable and valid. The study aims to recruit approximately 15 patients who meet the inclusion and exclusion criteria. This sample size is sufficient for the primary goal of assessing study procedures, recruitment rates, and study adherence of this supportive care intervention, rather than determining definitive treatment efficacy.

Participants will undergo a comprehensive clinical evaluation at baseline and post-intervention. The 4-to-6-week intervention features a daily automated assessment using a previously validated AI-assisted physiotherapy assessment system. Based on these assessments, an established system clinical workflow then dynamically generates a personalized exercise curriculum for that day. Participants will perform these exercise programs 4 to 6 times daily.

Key feasibility metrics, including recruitment rate, retention rate, intervention adherence, and system acceptability (SUS score), will be summarized using descriptive statistics. Adherence will be monitored automatically via server log data. Following the Intention-to-Treat (ITT) principle, technical dropouts who transition to face-to-face therapy will be documented as a feasibility outcome to identify technical barriers.

Preliminary efficacy will be evaluated by analyzing functional changes from baseline to endpoint, including Maximal Interincisal Opening (MIO), cervical and shoulder ROM, upper extremity function questionnaires, and quality-of-life questionnaires. Depending on data distribution, changes will be analyzed using the Paired t-test or the Wilcoxon Signed-Rank Test. Effect size estimates (Cohen's d) and their 95% confidence intervals will be calculated.

This study is expected to provide essential feasibility and safety evidence of an AI-assisted home-based exercise protocol, establishing a foundation and providing sample size calculations for a subsequent randomized controlled trial.

Inclusion Criteria:

• Oral cancer patients with trismus, clinical signs of neck or shoulder joint impairment after oral cancer surgery or radiotherapy in recent 12 months
• Age between 20 and 70 years

Exclusion Criteria:

• Could not communicate
• Had any disorder that could influence movement performance (e.g., stroke, Parkinsonism, head injury)

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.