July 29, 2022 — The Radiological Society of North America (RSNA), in collaboration with the American Society of Neuroradiology (ASNR) and the American Society of Spine Radiology (ASSR)has launched the “RSNA Cervical Spine Fracture AI Challenge” to explore whether artificial intelligence (AI) can be used to aid in the detection and localization of cervical spine injuries.
The international imaging dataset being compiled and selected for the challenge is one of the largest and most diverse of its kind, including detailed clinical labels, radiologist annotations, and segmentations.
“A unique aspect of this year’s RSNA AI Challenge is the sheer diversity of data,” said Errol Colak, MD, FRCPC, assistant professor in the Department of Medical Imaging at the University of Toronto in Ontario, Canada. “Our team has compiled a large dataset of cervical spine CT scans from 12 institutions in nine countries on six different continents. In addition, this year’s competition will feature annotated data in multiple ways, including exam level labels, vertebral body segmentation, and image level bounding boxes.”
More than one million vertebral fractures and more than 17,000 spinal cord injuries occur annually in the United States. The most common site of spinal fracture is the cervical spine, located in the neck. Elderly populations are particularly vulnerable and fractures may be more difficult to detect on imaging due to overlapping degenerative diseases and osteoporosis.
Imaging of spinal fractures in adults is now performed almost exclusively with computed tomography (CT) scans rather than x-rays. Prompt detection of the location of any vertebral fracture is essential to prevent neurological deterioration and paralysis after surgery. trauma. The researchers hope that AI can help quickly identify and locate fractures.
To create the actual data set, the Challenge Planning Task Force collected imaging data from 12 sites on six continents, including more than 1,400 CT scans with diagnosed cervical spine fractures and an approximately equal number of negative scans. ASNR and ASSR spinal radiology specialists provided expert image level annotations on these images to indicate the presence, vertebral level, and location of any cervical spine fractures.
For the challenge competition, contestants will attempt to develop machine learning models that match the performance of radiologists in detecting and locating fractures within the seven vertebrae that make up the cervical spine.
“The machine learning models that are developed as part of this challenge can help improve patient care by helping radiologists and other clinicians detect fractures, which can be a difficult task,” said Dr. Colak. “These models may be of particular value in underprivileged areas with limited access to expert neuroradiologists. Additionally, these models can aid patient care by prioritizing positive CT scans for radiologist review in high-volume clinical settings.”
The RSNA Cervical Spine Fracture AI Challenge takes place on a platform provided by Kaggle, Inc., and is open to everyone. The competition phase will end in October. The top 10 performing competitors will receive a total of $30,000.
Winners will be recognized at the AI Showcase during the RSNA 108th Annual Meeting and Scientific Assembly at McCormick Place Chicago (RSNA 2022, Nov. 27-Dec. 1).
To learn more about the challenge, visit RSNA.org/AI-image-challenge