Keywords: Segmentation, Segmentation
Motivation: Accurate disc segmentation in TMJ MRI is essential for diagnosing TMD; however, it is challenged by inter-observer variability and limited expert availability.
Goal(s): Our goal is to develop a semi-supervised deep learning model for TMJ disc segmentation to reduce manual labeling and improve diagnostic consistency.
Approach: Using SAM, one of the foundation models, as a baseline, we employed a semi-supervised learning method that incorporates Dice loss and cosine similarity between feature maps of the labeled and unlabeled data.
Results: The model achieved performance comparable to fully supervised models, demonstrating robust segmentation in both open- and closed-mouth positions, regardless of disc position.
Impact: The proposed model has the potential to assist dental clinicians by improving the accuracy and consistency of TMJ MRI interpretation while reducing manual effort. This approach could facilitate further exploration into automated diagnostics across various imaging modalities, enhancing clinical workflows.
How to access this content:
For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.
After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.
After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.
Keywords