Temporomandibular Disorders (TMD), TMJ Disorder, TMJ Pain
Conditions
Keywords
Machine Learning, Cone-Beam Computed Tomography, CBCT, Magnetic Resonance Imaging, MRI, Proprioceptive Training, low level laser, Laser Acupuncture
Brief summary
This study aims to improve clinical efficiency and effectiveness in treating Temporomandibular Disorders (TMD). The researchers will integrate machine learning technology with clinical examination data, CBCT, and MRI to identify the core underlying issues of each patient. Based on this precise, AI-assisted diagnosis, the study will implement a tailored treatment framework. The final goal is to develop predictive algorithms that can determine MRI necessity while ensuring that clinical interventions are highly targeted, efficient, and successful.
Detailed description
This study integrates machine learning with a clinical trial framework to optimize the evaluation and management of Temporomandibular Disorders (TMD). A total of 180 eligible participants are expected to be enrolled in the study. To ensure a homogenous sample, strict screening will be conducted using standardized psychometric scales; patients with significant psychosocial confounders (defined as GAD-7 ≥10, PHQ-9 ≥15, PHQ-15 ≥10, or PSQI \>5) or comorbidities such as fibromyalgia, rheumatic autoimmune diseases, and trauma history will be excluded. Prior to treatment, baseline clinical examination, cone beam computed tomography (CBCT), and magnetic resonance imaging (MRI) data will be collected. All clinical data and longitudinal pre- and post-treatment MRI findings will be utilized to train advanced multi-modal diagnostic models; based on this foundational intelligence, predictive algorithms will then be developed to determine the necessity of an MRI scan using only clinical data and CBCT images. The enrolled participants will be divided by active maximum mouth opening into two subgroups (≤30 mm and \>30 mm) and randomly assigned to three treatment arms: (1) Proprioceptive Training + Sham + LLLT (2 weeks of sham laser followed by 2 weeks of active laser acupuncture); (2) Proprioceptive Training + LLLT + LLLT (4 weeks of active laser acupuncture); and (3) Waiting List Control (receiving standard conservative care only).Proprioceptive training includes gentle teeth tapping, tongue-to-palate contact, and neutral head-neck posture exercises, performed three times daily. LLLT targets the Xiaguan (ST7), Yifeng (SJ17), and bilateral Fengchi (GB20) points once weekly. Assessments will be conducted biweekly during treatment, with final evaluation at 4 weeks and follow-ups at 1, 3, 6, 12, and 24 months post-treatment. Post-treatment MRI will be performed at the 4-week endpoint to evaluate objective intra-articular and perimuscular soft-tissue changes, correlating with improvements in pain resolution efficiency and mandibular range of motion recovery.
Interventions
DEVICELow-Level Laser Acupuncture
Laser acupuncture targets the Xiaguan (ST7), Yifeng (SJ17), and bilateral Fengchi (GB20) points once weekly. For the LLLT+LLLT group, active laser is given for 4 weeks. For the Sham+LLLT group, sham laser is given for the first 2 weeks followed by 2 weeks of active laser.
BEHAVIORALProprioceptive Training
Includes gentle teeth tapping, tongue-to-palate contact, and neutral head-neck posture exercises, performed three times daily for 4 weeks.
Standard conservative clinical care utilizing a hard acrylic flat-plane stabilization splint (centric stabilization splint) for TMD management. The splint is customized to provide even, bilateral occlusal contact in the musculoskeletal stable position during the 4-week trial period.
Sponsors
En Chu Kong Hospital
National Yang Ming Chiao Tung University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
QUADRUPLE (Subject, Caregiver, Investigator, Outcomes Assessor)
Masking description
This is a double-blind trial using coded laser devices (e.g., Device A and Device B) provided by the manufacturer. Both devices have identical appearances and sounds. The investigator, who acts as the care provider and outcomes assessor, knows the group assignment but remains blind to which code represents the active laser energy until the database is locked.
Eligibility
Sex/Gender
ALL
Age
18 Years to 70 Years
Healthy volunteers
No
Inclusion criteria
* Patients diagnosed with temporomandibular disorders (TMDs) screened by the Department of Dentistry, Division of Oral and Maxillofacial Surgery at En Chu Kong Hospital, who have failed to respond to analgesic treatment for at least one week. * Patients clinically examined and identified as having pain predominantly originating from muscles associated with mandibular movement. * Patients aged between 18 and 70 years (inclusive) with full legal capacity. * Patients who are willing to participate in the study and provide written informed consent.
Exclusion criteria
* Patients meeting any of the following clinical scale thresholds: Generalized Anxiety Disorder-7 (GAD-7)≥ 10, Patient Health Questionnaire-9 (PHQ-9) ≥ 15, Patient Health Questionnaire-15 (PHQ-15) ≥ 10, or Pittsburgh Sleep Quality Index (PSQI) \> 5. * Patients with tumors of the temporomandibular joint (TMJ) or head and neck region. * Patients with acute or chronic infections. * Patients with a history of neuropathy or stroke. * Patients with a history of head and neck trauma. * Patients diagnosed with fibromyalgia. * Patients diagnosed with autoimmune or rheumatic diseases. * Patients diagnosed with irritable bowel syndrome (IBS). * Patients with claustrophobia (unable to undergo MRI examination). * Women who are pregnant or lactating, or parents currently in the infant care/nursing period. * Patients who have received invasive treatment for temporomandibular joint disorders within the past 3 months.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Tenderness of Masticatory Muscles and Periarticular Temporomandibular Joint Pain | Baseline, 2 weeks, 4 weeks (endpoint), and long-term follow-ups at 1, 3, 6, 12, and 24 months. | Evaluated using the Numeric Rating Scale (NRS) 0-10, to measure pain intensity upon palpation of the masticatory muscles and the periarticular structures of the temporomandibular joint. |
| Pain-Free Mandibular Active Range of Motion | Baseline, 2 weeks, 4 weeks (endpoint), and long-term follow-ups at 1, 3, 6, 12, and 24 months. | Measured as the maximum active mouth opening (maximum unassisted pain-free opening) in millimeters using a calibrated ruler. |
| Jaw Functional Limitation Scale - 20 (JFLS-20) | Baseline, 4 weeks (endpoint), and long-term follow-ups at 1, 3, 6, 12, and 24 months. | The Jaw Functional Limitation Scale-20 (JFLS-20) is a 20-item patient-reported questionnaire assessing jaw functional limitation during the previous month. Each item is rated from 0 (no limitation) to 10 (severe limitation). Domain scores are calculated as the mean of completed items for mastication (items 1-6), vertical jaw mobility (items 7-10), and verbal and non-verbal communication (items 11-20). The overall JFLS-20 score is the mean of the three domain scores. Scores range from 0 to 10; higher scores indicate greater jaw functional limitation and a worse outcome. A domain score is calculated when no more than 2 mastication items, 1 mobility item, or 2 communication items are missing. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Cervical active range of motion (AROM) limitation | Baseline, 2 weeks, 4 weeks (endpoint), and long-term follow-ups at 1, 3, 6, 12, and 24 months. | Cervical AROM of flexion, extension, left/right rotation, and left/right lateral flexion is measured using a plastic universal goniometer in degrees from 0-80/40/45 etc. For reporting purposes, each movement is categorized as "normal", "limited", or "no movement", based on predefined cutoffs (e.g., rotation ≥80° = normal; 40-79° = limited; \<40° = no movement). The outcome is the proportion of participants with limited cervical AROM at each time point. |
| MRI T2-Weighted Soft Tissue Assessment | Baseline (pre-treatment) and 4 weeks (post-treatment endpoint). | T2-weighted MRI scans were compared with T1-weighted images to identify joint effusion, disc displacement, perimuscular/intramuscular edema, and muscle hemorrhage. |
Contacts
CONTACTTing-Yu Huang, D.D.S.
STUDY_DIRECTORChing-Po Lin, PhD
Institute of Neuroscience, Brain Research Center, National Yang Ming Chiao Tung University, Taipei, Taiwan. Department of Education and Research, Taipei City Hospital, Taipei, Taiwan.
STUDY_DIRECTORChing-Mao Chang, MD, PhD
Center for Traditional Medicine, Taipei Veterans General Hospital, Taipei, Taiwan. Institute of Traditional Medicine, and Schools of Chinese Medicine, College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan.
Outcome results
None listed