A study by the Technical University of Munich researchers evaluated the use of a deep learning algorithm to differentiate between colon cancer and acute diverticulitis on CT images and its impact on radiologists’ performance.
The 3-D convolutional neural network reached a sensitivity of 83.3% and specificity of 86.6% compared to the average reader sensitivity of 77.6% and specificity of 81.6%.
The radiology readers with the algorithm support improved their performance significantly. With a sensitivity increase from 77.6% to 85.6% and a specificity increase from 81.6% to 91.3%. The AI support also reduced false-negative and false-positive findings.
The study suggests that a deep learning model as a support system may significantly improve radiologists’ diagnostic performance and patient care.
Ziegelmayer S, Reischl S, Havrda H, et al. Development and Validation of a Deep Learning Algorithm to Differentiate Colon Carcinoma From Acute Diverticulitis in Computed Tomography Images. JAMA Netw Open. 2023;6(1):e2253370. doi:10.1001/jamanetworkopen.2022.53370
Leave a Reply