A Case Western Reserve University-led team will test their next strategy with rectal cancer patients, following its success in applying artificial intelligence (AI), to medical imaging, to improve treatment for other diseases.
In particular, the researchers hope to provide reliable advice on whether patients should undergo surgery as part of their treatment.
Colorectal cancer is the third most frequent cancer that is diagnosed by women and men in the United States, excluding skin cancer. It encompasses colon cancer as well as cancers of the rectum. According to the American Cancer Society, there will be more than 45,000 cases of rectal cancer and more than twice as many cases of colon cancer by 2021.
However, the Case Western Reserve researchers state that doctors do not have a reliable method to predict which patients will respond to radiation or chemotherapy. Thus, the majority of patients must undergo invasive surgery in order to remove the rectum.
Satish Viswanath (an assistant professor of biomedical engineer) is the lead researcher. He is a member the Center for Computational Imaging and Personalized Diagnostics. “Instead in the event that our AI technology is successful, we can inform the physician in advance-based on an ordinary MRI (magnetic resonance imaging) scan-;if a patient will do well with only chemotherapy and be observed, without having this serious surgery.”
Collaboration and DOD grant
Working in collaboration with physicians at Cleveland Clinic, University Hospitals and the Louis Stokes Cleveland VA Medical Center, the Case Western Reserve team will apply AI techniques to digitize thousands of images from medical institutions.
The grant, which is three years in duration and valued at $755,000, was provided by the U.S. Department of Defense’s Congressionally Directed Medical Research Programs.
Other research has reported that up to 30% of those diagnosed with rectal cancer receive surgery they didn’t need, Viswanath said, adding that it’s a “surgery that is expensive, both financially and in the way that it affects the daily life of the patient afterward.”
The effects can include the necessity for a colostomy bag even if temporary, and possible changes in everything from sexual function mental health to infection, according to previous research.
Viswanath will work with Pingfu Fu (a professor of Population and Quantitative Health at Case Western Reserve), on the research. The research team includes three clinical co-investigators Sharon Stein, University Hospitals; Andrei Purysko, Sharon Stein, and Smitha Krishnamurthi, all from Cleveland Clinic; Eric Marderstein, VA Medical Center.
The team will work from images of more than 2,000 rectal cancer patients who were treated at the hospitals in the last five years and will test their AI on 450 to 500 patients.
They will then review their radiomics to determine whether it could have revealed which patients would most benefit from chemoradiation therapy, and which ones would not, thus requiring 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 algorithms can help to identify the presence of tumors and other signs that are normally not visible to the naked eye. Through the course of this research, Viswanath’s team will design and validate new radiomic tools to identify the rectal tumors that are related to chemoradiation.
Anant Madabhushi is the Director of CCIPD and also the Donnell Institute Professor in Biomedical Engineering at Case Western Reserve. He said Viswanath has already made significant progress making use of the tools to predict the response to rectal cancer.
Madabhushi’s lab, founded in 2012 and with more than 60 researchers, has grown into an international leader in the field, specializing in the detection, diagnosis and characterization of different cancers and other diseases through meshing medical imaging machine learning, AI, and machine learning.
His team will be able, with the support of CWRUs medical partners, to test these tools in a multidisciplinary setting. This will set the stage for future clinical studies.
Anant Madabhushi is the Donnell Institute Professor in Biomedical Engineering at Case Western Reserve University
Content Source: https://www.news-medical.net/news/20211204/AI-driven-medical-imaging-may-help-fight-against-rectal-cancer.aspx