Monday, November 27 | 1:30 p.m.-1:40 p.m. | M6-SSNPM01-1 | Room E351
With help from an AI algorithm, radiologists can detect and classify lung nodules on routine clinical chest CT exams faster and more effectively, according to this new study.
A team led by cardiothoracic radiologist and researcher Dr. Catherine Jones of Monash University, the University of Sydney, and the I-MED Radiology Network in Australia sought in their study to assess the performance AView AI software (Coreline Soft) in a cohort of 130 CT studies, including 87 with up to five nodules and 43 without nodules. For this analysis, a retrospective modeling approach was employed to gauge the clinical and financial implications and encompassing cost savings and return-on-investment (ROI) calculations, the researchers noted.
They found that AI assistance led to a mean reading time of 235 minutes, down sharply from the 585 minutes without help from AI. In addition, the researchers observed statistically significant improvements in performance and an impressive ROI.
"The superior diagnostic performance facilitated by AI assistance has the potential to significantly enhance patient outcomes through the timely identification and judicious management of incidentally detected lung nodules," Jones et al wrote. "Moreover, the concomitant economic impact, characterized by optimized resource utilization, substantial cost savings, and an attractive ROI, positions AI as an indispensable asset for healthcare providers in the pursuit of diagnostic excellence in routine clinical settings
Stop by this early afternoon session on Monday to learn more.