Site Editor

Thomas Flaig, MD


Can Artificial Intelligence–Based Tool Help to Assess Treatment Response in Bladder Cancer?

By: Victoria Kuhr, BA
Posted: Monday, May 16, 2022

Lubomir Hadjiiski, PhD, of the University of Michigan, Ann Arbor, and colleagues reported that the computerized decision support system (CDSS-T) has the potential to improve the accurate assessment of a patient’s response to neoadjuvant chemotherapy for muscle-invasive bladder cancer. Additionally, physicians who used the CDSS-T aid demonstrated less variability and better agreement in their diagnoses for both the original and repeated evaluations of postchemotherapy computerized tomography (CT) urography scans, the investigators found. These findings were published in the journal Tomography.

“The CDSS-T aid was able to assist the inexperienced observers in making diagnoses at a level comparable to that of experienced observers,” said the study authors.

The study received approval from an institutional review board to have 17 observers rate pre– and post–CT urography scans of 123 patients (157 pre- and post-treatment cancer pairs) for their level of response to chemotherapy. The observers included 14 physicians from various specialties, 2 fellows, and 1 medical student. The lower the score on the CDSS-T, the lower the likelihood of completed response to treatment and vice versa for higher scores.

The scans indicated that 40 of 157 lesion pairs had a complete response after chemotherapy, and 117 lesion pairs had an incomplete response after chemotherapy.

The average performance of the 17 observers was significantly improved when aided by the CDSS-T. The cancer case difficulty, institution affiliation, specialty, and assessment times influenced the observers’ performance without the CDSS-T. The increases in the area under curve (AUC) between the easy and difficult subsets were almost equal for most of the groups. The group of radiologists from the University of Michigan showed a 0.01 larger improvement for the easy subset than for the difficult subset. On average, oncologists had a larger gain in AUC than radiologists with the CDSS-T aid, suggesting the tool may be potentially useful for nonradiology physicians.

Disclosure: The study authors reported no conflicts of interest.

By continuing to browse this site you permit us and our partners to place identification cookies on your browser and agree to our use of cookies to identify you for marketing. Read our Privacy Policy to learn more.