In a small, multi-institutional study, an artificial intelligence-based system improved provider assessments of whether patients with bladder cancer had a complete response to chemotherapy before radical cystectomy (surgery removal of the bladder).
Still, the researchers caution that AI is no substitute for human expertise, and their tool should not be used as such.
“If you use the tool smartly, it can help you,” said Lubomir Hadjiyski, Ph.D., professor of radiology at the University of Michigan Medical School and lead author of the study.
When patients develop bladder cancer, surgeons often remove the entire bladder to prevent the cancer from coming back or spreading to other organs or areas. A growing body of evidence is accumulating, however, that surgery may not be necessary if a patient has no evidence of disease after chemotherapy.
However, it is difficult to determine whether the lesion left after treatment is simply necrotic or scarred tissue as a result of treatment or whether the cancer persists. Researchers wondered if AI could help.
“The big question was when you have such an artificial device next to you, how is it going to affect the doctor?” said Hadjiyski. “Will that help? Will it confuse them? Will it boost their performance or will they just ignore it? »
Fourteen physicians from different specialties – including radiology, urology and oncology – along with two fellows and one medical student reviewed the before and after treatment scans of 157 bladder tumours. Providers gave scores for three measures that assessed the level of response to chemotherapy as well as a recommendation for each patient’s next treatment (radiation therapy or surgery).
Next, providers reviewed a score calculated by the computer. Lower scores indicated a lower likelihood of complete response to chemo and vice versa for higher scores. Suppliers could revise their ratings or leave them unchanged. Their final scores were compared to tumor samples taken during their bladder removal operations to assess accuracy.
Across different specialties and experience levels, providers saw improvements in their assessments with the AI system. Those with less experience had even more gains, so much so that they were able to make diagnoses at the same level as the experienced participants.
“It’s the distinct part of this study that showed some interesting insights into the audience,” Hadjiyski said.
The tool helped providers in academic institutions more than those working in health centers focused only on clinical care.
The study is part of an NIH-funded project, led by Hadjiyski and Ajjai Alva, MD, associate professor of internal medicine at UM, to develop and evaluate biomarker-based tools for decision support. response to bladder cancer treatment.
In more than two decades of AI-based studies to assess different types of cancer and their response to treatment, Hadjiyski says he’s observed that machine learning tools can be useful as second opinions to help doctors make decisions, but they can also make mistakes.
“One interesting thing we discovered is that the computer makes errors on a different subset of cases than a radiologist would,” he added. “Which means that if the tool is used correctly, it gives a chance to improve but does not replace the doctor’s judgement.”
Other authors include Di Sun, Ajjai Alva, Heang-Ping Chan, Richard H. Cohan, Elaine M. Caoili, Wesley T. Kerr, Matthew S. Davenport, Prasad R. Shankar, Isaac R. Francis, Kimberly Shampain, Nathaniel Meyer, Daniel Barkmeier, Sean Woolen, Phillip L. Palmbos, Alon Z. Weizer, Ravi K. Samala, Chuan Zhou, and Martha Matuszak of UM; Yousef Zakharia, Rohan Garje and Dean Elhag from the University of Iowa; Monika Joshi and Lauren Pomerantz of Pennsylvania State University; Kenny H. Cha of the United States Food and Drug Administration’s Center for Devices and Radiological Health and Galina Kirova-Nedyalkova of the Acibadem City Clinic at Tokuda Hospital in Sofia, Bulgaria.
The title of the article
Computerized decision support for the assessment of bladder cancer treatment response in computed tomography urography: effect on diagnostic accuracy in a multi-institutional multi-specialty study
Publication date of articles
Warning: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.