Second research roadmap details priorities for AI in radiology
A new research roadmap published today details challenges for establishing AI in radiology and outlines top priorities for researchers to consider for introducing AI into current clinical workflows.
The report, put out by the Journal of the American College of Radiology, is a companion roadmap to part one which was published April 16 in Radiology.
Like the first installment, part two consists of conclusions taken from an August 2018 workshop held by the National Institutes of Health (NIH), but includes contributions from the National Institute of Biomedical Imaging and Bioengineering (NIBIB), part of the NIH. The event was co-sponsored by The Academy for Radiology and Biomedical Imaging Research, RSNA and the ACR.
"Although advances in foundational research are occurring rapidly, translation to routine clinical practice has been slower because we must ensure AI in medical imaging is useful, safe, effective and easily integrated into existing radiology workflows before they can be used in routine patient care,” Bibb Allen, MD, workshop co-chair and chief medical officer of the ACR Data Science Institute, said in a prepared statement from the ACR.
The four priority areas highlighted in the research roadmap are:
1. Create structured AI use cases and “defining and highlighting clinical challenges” that AI might solve.
2. Create avenues that encourage data sharing for training and testing algorithms that can be used across clinical practices.
3. Develop validation and performance monitoring tools to kick-start regulatory approval.
4. Focus on establishing standards and “common data elements” for integrating AI tools into established clinical workflows.
“Our companion paper gave a roadmap to advance foundational machine learning research. But for foundational research to benefit patients, novel algorithms must be evaluated and deployed in a safe and effective manner. This new roadmap paper gives guidance for the clinical translation of AI innovation,” said Curtis P. Langlotz, MD, PhD, report co-author and professor of radiology and biomedical informatics at Stanford University, in the same statement. “Together, these two connected roadmaps show us how AI not only will transform the work of radiologists and other medical imagers, but also will enhance the delivery of care throughout the clinical environment.”
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.