Meeting Banner
Abstract #1414

Segmentation of contrast enhancing lesions in multiple sclerosis using deep learning and a large cohort study

Refaat E Gabr1, Ivan Coronado1, and Ponnada A Narayana1
1Diagnostic and Interventional Imaging, University of Texas Health Science Center at Houston, Houston, TX, United States

Multiple sclerosis (MS), a demyelinating disease of the central nervous system, affects more than two million people worldwide. Contrast enhancing lesions are thought to reflect active disease state and play a key role in MS management. Deep learning (DL) based on convolutional neural networks has reached state-of-art performance on semantic segmentation tasks. Using annotated images for 398 MS patients, we evaluated DL performance on segmentation of enhancing lesions. Our approach yielded Dice similarity coefficient of 0.78, true positive rate of 0.91, and false positive rate of 0.28 for test data. Network performance was excellent for enhancement volumes ≥70 µl.

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.

Keywords