“Prostate cancer (PCa) is the second most frequent type of cancer found in men worldwide and the fifth leading cause of death by cancer,” according to a paper published April 12, 2021, in Scientific Reports. Lead author Steven Korevaar and his team from Melbourne’s RMIT University School of Engineering and Melbourne’s St Vincent’s Hospital Radiology department were “[i]nspired by the recent success of deep convolutional neural networks (CNN) in computer[-]aided detection (CADe), [so] we propose a new CNN based framework for incidental detection of clinically significant prostate cancer (csPCa) in patients who had a CT scan of the abdomen/pelvis for other reasons.”

Though the authors say that CT’s inferior soft-tissue characterization typically makes it unsuitable for diagnosing PCa, “our evaluations on a relatively large dataset consisting of 139 clinically significant PCa patients and 432 controls show that the proposed deep neural network pipeline can detect csPCa patients at a level that is suitable for incidental detection.”

Significantly, their neural network achieved higher detection rates than radiologists. “The proposed pipeline achieved an area under the receiver operating characteristic curve (ROC-AUC) of 0.88 (95% Confidence Interval: 0.86–0.90) at patient level csPCa detection on CT, significantly higher than the AUCs achieved by two radiologists (0.61 and 0.70) on the same task.”

Early detection of prostate cancer is notoriously difficult, and the authors acknowledge that more work needs to be done, such as assessing the pipeline’s performance in detecting different forms of cancer and determining patient outcomes. This new AI-aided imaging technique is a promising breakthrough that may lead to thousands of more men surviving prostate cancer in the future.

Source:

Korevaar, S., Tennakoon, R., Page, M. et al. “Incidental detection of prostate cancer with computed tomography scans.” Sci Rep 11, 7956 (2021). https://doi.org/10.1038/s41598-021-86972-y

This article is published as Open Source under the Creative Commons Attribution 4.0 International License. You may view a copy of the license at: http://creativecommons.org/licenses/by/4.0/. Only minor editorial changes were made when quoting from this article.