The $4.3-million NIDCR grant will help clinicians build upon previous work to improve patient outcomes for those suffering with TMD pathology.

THERE ARE FEW JOINTS AS COMPLEX as the jaws — they allow people to move the mandible forward and backward, up and down and side to side. But the simple act of chewing, talking or even smiling can be excruciating for the estimated 5 to 12 percent of people living with temporomandibular disorders (TMD), a group of conditions affecting the jaw joint.

A new $4.3-million, 5-year federal grant from the National Institute of Dental and Craniofacial Research (NIDCR-NIH) will help Ostrow researchers and their colleagues build upon work already completed to conduct Broader Evaluation of TMD Treatment Efficacy and Response (BETTER-TMD), a part of the TMD Collaborative for Improving Patient-Centered Translational Research (TMD IMPACT).

BETTER-TMD will follow up to 5,000 patients over time to understand which treatments work best for which types of patients.

“This is an understudied disease,” says Jianfu Chen, the study’s co-principal investigator and professor at Ostrow. “There are about 30 different categories of temporomandibular joint disorders. It’s a really big deal.”

USC is one of three U.S. research teams selected for the TMD Impact Collaborative, a nationwide effort supported by federal funds to improve patient-centered, translational research in the condition.

Enter Artificial Intelligence

Today, many TMD treatment choices are largely guided by a clinician’s experience rather than robust data. That’s a problem in a condition as complex and variable as TMD, where patients can present with very different symptoms, underlying causes and responses to therapy.

“The goal is to develop this data‑driven tool that helps clinicians make more evidence‑based, personalized treatment decisions,” Chen says. “Oftentimes, there’s misdiagnosis or delayed treatment, and the decision for a certain treatment is based on clinician experience — it’s not based on data or evidence.”

This project aims to address this critical gap by collecting a large, structured and longitudinal data set that can let clinicians eventually match specific patient subpopulations with the treatments that are most likely to be effective.

The study will draw from real-world data from USC as well as private practice sites across the United States. Clinicians will record detailed clinical signs, symptoms and diagnoses in a custom-developed app called Smart Note, while patients report their treatment outcomes over time. Those data streams will feed into algorithms designed to help clinicians choose the most effective, individualized care.

Part of the challenge, researchers say, is the jaw joint itself. The temporomandibular joint is small but intricate, involving bone, cartilage, discs, synovial tissue and muscle, and is heavily used whenever people eat, speak or smile.

On top of that biological complexity, factors such as anxiety, depression, stress, sex, aging and environment can all influence pain and function.

At USC, the project is led clinically by Associate Dean of Distance Learning and Telehealth Glenn Clark, an internationally recognized expert in TMD care, and Anette Vistoso MS ’20, assistant professor of clinical dentistry, director of the Oral Care Precancer and Pain Clinic and director of distance learning and telehealth. All data is collected through AI-enabled software developed by Ostrow staff member, software engineer and data scientist Nicolas Veas. The effort spans USC academic units, with faculty member Gerald Loeb from the USC School of Engineering offering guidance and support.

Dean Yang Chai PhD ’91, DDS ’96 serves as a key co-investigator and advisor, providing school-wide leadership and infrastructure support. “None of the other places can do such a comprehensive, large‑scale clinic study, because it’s not something a few scientists or clinicians can do — it’s really a school‑wide effort,” Chai said. “I think it’s really a patient‑centered clinic study.”