Meeting Banner
Abstract #1037

Fully Automated Deep Learning Pipeline for Meniscus Segmentation and Lesion Detection

Berk Norman1, Valentina Pedoia1, Thomas Link1, and Sharmila Majumdar1

1Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States

Damage to the meniscus is a physically limiting injury that can lead to further medical complications. Automatically classifying this type of meniscal damage poses the advantage for quicker and more accurate diagnosis at the time of an MRI scan. Using a fully automated deep learning pipeline we identify the region around the 4 meniscal horns and then classify if a lesion exists and if so, its severity based on WORMS grading. Lesion detection achieved 89.81% specificity and 81.98% sensitivity. This algorithm has the ability to quickly identify meniscal lesions from MRI and filter higher risk lesion subjects.

This abstract and the presentation materials are available to members only; a login is required.

Join Here