ACR strengthens focus on AI development through new collaboration with the National Cancer Institute
The American College of Radiology has teamed up with the National Cancer Institute to create a one-stop shop for artificial intelligence developers.
As part of the new approach, the ACR’s Data Science Institute is now linking use cases to data within the NCI’s Cancer Imaging Archive (TCIA), the pair announced Monday. The move will give radiology AI creators access to clearly defined data elements and DICOM images for inputs, outputs, and training and testing models.
“Finding a good source of data … for model testing and validation is an ongoing challenge for those developing AI,” Bibb Allen Jr., MD, chief medical officer of the ACR DSI, said in a statement. “We’ve now made it easier for those developing AI for medical imaging to get good data, and we’re sharing it along with the guidelines provided by our use cases to be sure the algorithms developed have value for us as radiologists.”
Medical AI creators have been increasingly asking for better access to imaging data, according to the ACR. So TCIA datasets are now matched to ACR DSI cancer and non-cancer use cases based on a few categories. Those include body area, modality, and presence of secondary comorbidities.
Most data are available to use for free when developing commercial machine learning tools, the college noted.
Earlier this month the ACR launched a new repository known as the National Clinical Imaging Registry. The informatics tool brings in data from multiple national organizations with hopes of supporting AI development, clinical research, and many other goals.
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.