Abstract #4014

Self-supervised Pretraining on OAI data for 3D Knee MRI Analysis

Xinxin Wang^1,2,3, Liam Hazan⁴, Mingrui Yang^1,2, Simona Rabinovici-Cohen⁴, and Xiaojuan Li^1,2,5

¹Program of Advanced Musculoskeletal Imaging (PMAI), Cleveland Clinic, Cleveland, OH, United States, ²Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States, ³Department of Computer and Data Sciences, Case Western Reserve University, Cleveland, OH, United States, ⁴IBM Research Labs, IBM Research, Haifa, Israel, ⁵Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States

Synopsis

Keywords: Diagnosis/Prediction, Machine Learning/Artificial Intelligence, Knee MRI; Pretrained model

Motivation: Self-supervised pretraining is efficient and requires no labeled data, yet it is understudied for 3D knee MRI analysis.

Goal(s): Our goal is to develop a self-supervised pretraining model and explore its potential for 3D knee MRI analysis.

Approach: We use DINO pipeline for self-supervised training on OAI data and apply the pretrained model to downstream tasks, comparing it with training from scratch and a supervised pretrained model.

Results: Our OAI-DINO pretrained model significantly outperforms training from scratch on downstream tasks, offers comparable segmentation results and improves classification performance over the supervised pretrained model.

Impact: Our study has leveraged the OAI database and demonstrated the effectiveness of self-supervised pretraining for 3D knee MRI. Our approach enhances downstream task performance, inspiring further study on advancing automated 3D medical imaging analysis without labeled data.

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.

Click here for more information on becoming a member.