Artificial intelligence (AI) is improving radiology workflow by helping to shorten reading time, detect unexpected findings, provide measurements, and enable triage of abnormal examinations, according to Dr. Philippe Grenier, a past ECR president and consultant chest radiologist at Foch Hospital in Suresnes, France.
A particular benefit of AI is the decrease interpretation time for plain radiographs and CT scans, thanks to, for example, the capability to provide automated detection of pulmonary emboli on CT and to automatically detect and volumetrically measure pulmonary nodules, said Grenier during an interview published on 7 December by the European Society of Radiology (ESR) on its AI blog.
Dr. Philippe Grenier.Algorithms can also detect unexpected abnormalities and calculate measurements of anatomically segmented structures or detected lesions. What's more, AI tools can simplify and shorten reading times by triaging abnormal plain chest radiographs.
"However, on the opposite end, more complex AI tools are developed to predict the outcome, prognosis, probability of malignancy of a detected lesion, or to predict treatment response to this lesion," he said. "This provides an improvement of the quality of the medical decision, particularly for optimized treatment, but does not affect clinical workflow in hospitals."
AI also could play a role in addressing the rapid increase in imaging exam volume.
"The development of AI-based algorithms to control and validate the clinicians' requests for imaging examinations according to the referral guidelines should be widely appreciated," Grenier said. "Such tools already exist in hospitals to analyze clinicians' prescriptions in order to facilitate the pharmacists' validation."
Some hospitals and staff members may be resistant to adoption of radiology AI tools, but the way to overcome this challenge is by validating the technology.
"In other words, the high value of confidence index of AI algorithms performances have to be demonstrated before integration into the workflow," he continued. "For instance, the demonstration of a very high predictive negative value of AI-based chest radiographs interpretations is necessary to make emergency physicians and radiologists confident enough to use such AI tools."
Financial barriers do remain, although for the moment the cost of implementing these AI tools is limited as they tend to be provided by small startup companies, according to Grenier. That may not be the case when big firms supply global AI offerings.
"At this point in time, implementation of AI global solutions should necessitate improvement of hospital productivity and of reduction of expenses, including human resources, to ensure financial balance," he concluded.
