Nuance “AI in Practice” demonstration shows the benefits of standards-based data-sharing
Informatics experts from the RSNA and the American College of Radiology (ACR) have been collaborating with IT vendors for the past five years to unlock the clinical value of vast amounts of data created within radiology. Now in an “AI in Practice” demonstration applying Common Data Elements (CDE) standards, project collaborators are showing that radiologists can at last begin to access and use that data effectively to improve patient outcomes and the practice of radiology itself.
Alkasab
Kohli
Kovalenko
The CDE Steering Subcommittee has developed a growing CDE catalog for structured radiology reporting, available at RadElement.org. The group also has been working closely with Nuance and other vendors to implement CDE for generating and sharing actionable radiology reporting data across disparate AI services, PACS, EHR, and other downstream systems.
“We are coming out of the newborn phase and into the toddler phase,” said Marc Kohli, MD, associate professor, associate chair of clinical informatics at the University of California, San Francisco, and the RSNA co-chair of the initiative. “I don’t think we’re quite walking yet, but we’re getting close.”
The use of CDE in structured radiology reporting makes it possible to initiate and guide clinical action by physicians and other healthcare professionals. For example, structured data can automatically prompt the inclusion of evidence-based clinical guidance and trigger communication between radiologists and referring physicians to reduce the incidence of failed follow-up and adverse outcomes. The data can be used to optimize radiology workflows, improve report consistency and accuracy, evaluate clinical and financial performance, or inform population health initiatives. CDE and standards also add consistency and allow for the integration of the data produced by radiology AI models.
“Radiology is moving away from the free-form, narrative report a radiologist dictates,” said Tarik K. Alkasab, MD, PhD, a radiologist at Massachusetts General Hospital and the ACR co-chair of the initiative. “Now, it is about creating real, valuable data from the narrative report that can be used in downstream systems.”
The CDE Steering Subcommittee will continue defining and expanding the use cases for CDE and working with vendors including Nuance to implement the standard in their radiology solutions.
“The implementation of CDE standards allows health systems or anyone else that has a reporting system to unlock value from vast amounts of diagnostic imaging and radiology reporting data,” said Kathryn Kovalenko, director of diagnostic AI and reporting at Nuance. “This is a collaborative effort to put AI into practice – and derive value from the sea of radiology data to improve patient outcomes.”
At its virtual booth during RSNA 2020, Nuance will demonstrate a “Radiology Experience of the Future,” using scenarios in which the combination of CDE, conversational AI, and natural language processing will automatically generate structured data as the radiologist dictates and then drives appropriate clinical actions.
Visit the Nuance Communications virtual booth at RSNA 2020.
