May 29, 2024 — The Radiological Society of North America (RSNA) has launched the 2024 RSNA Lumbar Degenerative Disease Classification AI Challenge. RSNA is hosting the AI Challenge to spur development of AI tools for radiology and improve patient care.
The competition, developed in collaboration with the American Society of Neuroradiology (ASNR) and the American Society of Spine Radiology (ASSR), challenges researchers to develop models that can detect and classify degenerative spine disease using lumbar MR images. The carefully curated and annotated dataset includes images from eight locations across five continents.
“With the help of more than 50 volunteer expert annotators from the ASNR and ASSR, our task force is excited to introduce the largest open source annotated lumbar spine MRI dataset of its kind,” said Jason Talbott, MD, co-leader of the Challenge Task Force and professor in the Department of Clinical Radiology and Biomedical Imaging at the University of California, San Francisco (UCSF) and San Francisco General Hospital.
According to the World Health Organization, back pain is the leading cause of disability worldwide, with 619 million people suffering from back pain in 2020, a 60% increase since 1990.
Most people experience back pain at some point in their lives, with its frequency increasing with age. Pain and limited mobility are often symptoms of spondylosis. Spondylosis encompasses a range of degenerative spinal disorders, including degenerative discs and narrowing of the spinal canal (spinal stenosis), that put pressure on the nerves in the lower back.
MRI provides detailed images of the lumbar spine, discs, and nerves, allowing radiologists to evaluate the presence and severity of these conditions. Proper diagnosis and treatment can improve a patient's overall health and quality of life.
“Lumbar spine pain accounts for a large portion of medical imaging diagnostics, so developing a powerful algorithm has the potential to have a significant impact on the vast number of patients suffering from this common condition,” said Tyler Richards, MD, co-leader of the Challenge Task Force and assistant professor of Neuroradiology at the University of Utah Hospital.
AI provides tools to improve the efficiency and accuracy of MRIs. To build these tools, AI researchers need access to large amounts of imaging data that has been annotated by expert radiologists. Data challenges such as the RSNA Lumbar Degenerative Disease Classification AI Challenge encourage the radiology community to develop such datasets to provide a standard of truth when training AI systems to perform tasks related to diagnostic imaging, according to a news release detailing the RSNA challenge.
“This AI Challenge addresses a clinical problem that has been shown to be difficult for radiologists to agree on, and this could play a potential role for AI in providing standardization and consistency that has been difficult for radiologists to address,” Richards said.
In the competitions, researchers compete to see how well their AI models perform specific tasks, such as detecting, locating and classifying abnormal features, according to defined performance criteria. Each AI competition explores and demonstrates how AI can benefit radiology and improve patient care.
The 2024 RSNA Lumbar Degenerative Disease Classification AI Challenge will be run on a platform provided by Kaggle, Inc. and is open to all researchers. The top nine participants will share a total prize pool of $50,000.
“Participants will have the opportunity to build algorithms that improve our ability to accurately and rapidly identify and classify lumbar degenerative disease. Such tools have the potential to improve on the current state of the art and change the future of degenerative spine diagnosis and management,” said Dr. Talbot.
To date, the RSNA AI Challenge has attracted over 1,000 teams of AI experts from around the world, with the best-performing models being released under an open license to foster further research.
The competition will run through September 2024. Winners will be announced in November, and the winning teams will be honored in the AI Theater at RSNA's 110th Scientific Assembly & Annual Meeting (RSNA 2024), December 1-5 at McCormick Place in Chicago, Illinois.
More information: RSNA.org/AI-image-challenge
