Meeting Banner
Abstract #1147

A Report on the International Workshop on Osteoarthritis Imaging Segmentation Challenge: A Multi-Institute Evaluation on a Standard Dataset

Arjun D. Desai1,2, Francesco Caliva3, Claudia Iriondo3, Naji Khosravan4, Aliasghar Mortazi4, Sachin Jambawalikar5, Drew Torigian6, Jutta Ellerman7, Mehmet Akçakaya8, Ulas Bagci4, Radhika Tibrewala3, Io Flament3, Matt O'Brien3, Sharmila Majumdar3, Mathias Perslev9, Akshay Pai9, Christian Igel9, Erik B. Dam9, Sibaji Gaj10, Mingrui Yang10, Kunio Nakamura10, Xiaojuan Li10, Cem M. Deniz11, Vladimir Juras12, Ravinder Regatte11, Garry E. Gold2, Brian A. Hargreaves2, Valentina Pedoia3, and Akshay S. Chaudhari2
1Electrical Engineering, Stanford University, Stanford, CA, United States, 2Radiology, Stanford University, Stanford, CA, United States, 3Radiology, University of California San Francisco, San Francisco, CA, United States, 4University of Central Florida, Orlando, FL, United States, 5Radiology, Columbia University Medical Center, New York, NY, United States, 6Radiology, University of Pennsylvania, Philadelphia, PA, United States, 7Radiology, University of Minnesota, Minneapolis, MN, United States, 8Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, United States, 9Computer Science, University of Copenhagen, Copenhagen, Sweden, 10Biomedical Engineering, Cleveland Clinic, Cleveland, OH, United States, 11Radiology, New York University Langone Health, New York, NY, United States, 12Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria

Cartilage thickness can be predictive of joint health. However, manual cartilage segmentation is tedious and prone to inter-reader variations. Automated segmentation using deep-learning is promising; yet, heterogeneity in network design and lack of dataset standardization has made it challenging to evaluate the efficacy of different methods. To address this issue, we organized a standardized, multi-institutional challenge for knee cartilage and meniscus segmentation. Results show that CNNs achieve similar performance independent of network architecture and training design and, given the high segmentation accuracy achieved by all models, only a weak correlation between segmentation accuracy metrics and cartilage thickness was observed.

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

Join Here

Abstract Presentation Video