Radiology Partners reaches ‘milestone,’ deploying clinical AI across more than 20 million exams
Radiology Partners has reached a new “milestone,” the country’s largest imaging group announced Monday, deploying clinical artificial intelligence across more than 20 million patient exams so far this year.
El Segundo, California-based RP estimated that 2,600 of its 3,600 radiologists (or 72%) are now using at least one AI tool in their daily workflow. Altogether, the practice has processed more than 30 million medical images with AI and 116 million reports via natural language processing, representing “unprecedented scale.”
“AI will not replace radiologists, but radiologists using AI are already revolutionizing how exams are interpreted, resulting in improved and faster diagnoses and providing superior information for our physician partners,” Rich Whitney, MBA, RP’s board chair and CEO, said in a Nov. 13 prepared statement. “AI is just a tool, but it is a powerful tool and a differentiator in the advancement of our role within healthcare.”
The announcement comes more than two years after Rad Partners inked a strategic partnership with vendor Aidoc. Neither would disclose terms of the deal at the time but said their alliance goes beyond a simple software installation, with the two working together to develop and refine AI tools using RP’s vast trove of data. Rad Partners said the partnership has proven “pivotal” in paving the way for the implementation of a wide variety of AI solutions. The company’s radiologists are now using Aidoc solutions (with corresponding improved detection rates) for intracranial hemorrhage (12.6%), pulmonary embolism (18.1%), incidental PE (35.8%), and fractures of the cervical spine (16.4%) or rib (60.5%).
In addition, RP also is utilizing solutions from vendors such as Rad AI (auto generating customized report impressions), Subtle AI (minimizing image-capture time), Qure.ai (triaging intracranial hemorrhage) and iCAD (detecting breast cancer). The company has a team of 13 AI-trained radiologist leaders and a data set spanning 50 million exams each year across 50 states. RP also has created best practices for evaluating AI and disseminated them across the organization.
“By augmenting radiologists’ capabilities, we can further elevate the quality of our work, transitioning from purely imaging experts to information experts,” Nina Kottler, MD, Rad Partners’ associate chief medical officer for clinical AI, said in the announcement. “AI tools are positively impacting care pathways and care coordination for patients, enhancing value for referring providers, hospitals and health systems.”
Kottler is leading a multi-year, multimillion-dollar effort to vet AI algorithms cleared by the U.S. Food and Drug Administration. She spoke about this work and RP’s AI efforts in an article published by the New York Times last month.
Marty Stempniak has covered healthcare since 2012, with his byline appearing in the American Hospital Association's member magazine, Modern Healthcare and McKnight's. Prior to that, he wrote about village government and local business for his hometown newspaper in Oak Park, Illinois. He won a Peter Lisagor and Gold EXCEL awards in 2017 for his coverage of the opioid epidemic.