Artificial intelligence helps UH doctors better detect colon cancer

Cleveland, Ohio – A colonoscopy is the best way to detect colorectal cancers, and a doctor generally uses experience to identify areas of concern and send tissue samples for testing.

However, doctors at university hospitals do get some help – from a computer designed to look at images of the colon and spot subtle changes that even a pair of well-trained human eyes might miss.

The system is called genius ji, is a software unit that works like the second set of doctors’ eyes. It constantly monitors and analyzes images during a colonoscopy and highlights suspicious polyps with a visible sign in real time – alerting the doctor to places where he thinks a closer look is warranted. And you find them 99.7% of the time, according to data from Medtronic, which makes the unit.

The data was published in the Journal of Gastroenterology This past July showed that the technique reduced the number of colorectal tumors missing on standard colonoscopy by up to 50%.

“Every 1% increase in the detection rate of adenoma reduces the risk of colorectal cancer.” 3%,” said Dr. Gerard Eisenberg, chief medical quality officer at the UH Digestive Health Institute. “There is currently no other technology that achieves this rate of success.”

UH began using the AI ​​unit in March on its main campus and says it is the first hospital in Ohio to fully integrate the technology into its endoscopy services.

This fall, UH will receive five more units as part of an award through Medtronic Health Equity Assistance Program For colorectal cancer screening. Medtronic and Amazon Web Services donate units to facilities that provide care to medically disadvantaged communities with low screening rates for colorectal cancer or where access to this cutting-edge technology is not currently available.

UH is one of 62 facilities in the United States to receive donated GI Genius units. The hospital plans to place one of the new units in each of its general endoscopy rooms, and one at UH Ahuja Medical Center. Ultimately, UH plans to get 20 across its system

Colorectal cancer is the third most common and second deadliest cancer in the United States, but when caught early, the survival rate for certain types of colorectal cancers can be over 90%.

death rates of colorectal cancer is higher in northeastern Ohio than nationally, and rates of diagnosis among people under 50 are increasing locally as well as according to statistics from the Ohio Cancer Surveillance System. In addition, African Americans are disproportionately affected, and have a lower five-year survival rate.

The high death rates in northeastern Ohio are concerning, and underscores the need for better screening capabilities in our region, Isenberg says.

“This statistic is alarming and explains why we need to start screening at age 45 in people without risk factors as recommended by many national organizations now, and even people with risk factors earlier,” Isenberg said.

“This AI module directly impacts our ability to improve colon cancer detection and prevention and provides a technology that is likely to reduce healthcare disparities in our community.”

On Monday, Jeff Martha, CEO of Medtronic, the company that developed the GI Genius unit, spoke to a crowd of biomedical engineering students at Case Western Reserve and faculty about the role of artificial intelligence in the future of medical technology.

“It sneaks into everything we do,” said Martha, who discussed the company’s use of artificial intelligence in spinal surgery, deep brain stimulation to treat Parkinson’s symptoms, and diabetes management. Martha said that many aspects of medicine that depend on the skill and experience of a doctor or surgeon can be improved by AI. “We are bringing technology into that and turning it from art to science,” he said.

“Colon and rectal cancer is often bad news,” Isenberg said. “We look forward to helping our community by providing access to a tool that helps detect pests before they turn into cancers and hopefully prevent more cases.”

Leave a Comment