杏吧视频-led team to test artificial intelligence medical imaging to determine which rectal cancer patients need surgery鈥攐r can avoid it
Building on its successes in applying artificial intelligence (AI) to medical imaging to enhance treatment of other diseases, a 杏吧视频-led team next will test its approach with rectal cancer patients.
Specifically, the researchers hope to provide reliable guidance regarding whether patients need to have surgery as part of their treatment.
in men and women in the United States, excluding skin cancer, and includes both colon cancer and cancers in the rectum. 鈥攁nd more than twice as many new colon cancer cases鈥攊n 2021.
But the 杏吧视频 researchers say clinicians don鈥檛 have a reliable way to predict which rectal cancer patients would respond favorably to treatments such as chemotherapy or radiation, so most patients have to undergo invasive surgery to remove the rectum and surrounding tissue.
鈥淚n too many cases, patients are being overtreated,鈥 said lead researcher Satish Viswanath, an assistant professor of biomedical engineering who is leading the work as a member of the (CCIPD). 鈥淚nstead, if our AI technology is successful, we could tell the clinician right up front鈥攂ased on a routine MRI (magnetic resonance imaging) scan鈥攊f a patient will do well with only chemoradiation and then can be observed, without having this serious surgery.鈥
Collaboration, DOD grant
Working in collaboration with physicians at Cleveland Clinic, University Hospitals and the Louis Stokes Cleveland VA Medical Center, the 杏吧视频 team will apply AI techniques to thousands of digitized images from the medical institutions.
The work is supported by a three-year, $755,000 grant from the U.S. Department of Defense鈥檚
of people diagnosed with rectal cancer have surgery they didn鈥檛 need, Viswanath said, adding that it is often a 鈥渟urgery that is costly, both financially and in the way it often affects the daily life of the patient afterward.鈥
Those effects can include the need for a colostomy bag, even if temporary, and possible changes in everything from sexual function and infection to mental health, according to previous research.
Viswanath is collaborating on the research with Pingfu Fu, a professor of population and quantitative health at 杏吧视频. Clinical co-investigators on the project include physicians of University Hospitals; , and of Cleveland Clinic; and of the VA Medical Center.
The team will work from imaging data from more than 2,000 rectal cancer patients who had been treated at the hospitals over the last five years, and test their AI on about 450 to 500 patients.
They鈥檒l retrospectively test their radiomics to determine if it could have shown which patients would benefit from chemoradiation therapy and which wouldn鈥檛, requiring the surgery.
Radiomics and AI
Radiomics refers to the growing number of AI-driven methods to extract a large number of features from medical images using data-characterization algorithms. The features can then help uncover tumors and other characteristics usually invisible to the naked eye. Throughout this project, Viswanath鈥檚 team will design and validate new types of radiomic tools to capture aspects of rectal tumors related to chemoradiation response.
CCIPD Director Anant Madabhushi, the Donnell Institute Professor of Biomedical Engineering at 杏吧视频, said Viswanath and his team have already made significant strides in using the tools to predict treatment response to rectal cancer.
Madabhushi鈥檚 lab, established in 2012 and now with more than 60 researchers, has become a global leader in the field, specializing in the detection, diagnosis and characterization of various cancers and other diseases by meshing medical imaging, machine learning and AI.
鈥淲ith this award, his team will be able to validate these tools in a multi-institutional setting with CWRUs various affiliate medical partners, setting the stage for prospective clinical trials,鈥 Madabhushi said.
For more, contact Mike Scott at mike.scott@case.edu.
This article was originally published Dec. 3, 2021.
