A chest X-ray might be flagged for potential problems by artificial intelligence, which could also aid in a variety of other medical procedures. Another excellent example is checking for colorectal cancer during a colonoscopy.
With the use of artificial intelligence, or AI, a colonoscopy—which is advised for Americans at average cancer risk starting at age 45—won’t be much different for patients. However, AI may be working behind the scenes to increase the likelihood of finding any malignant tumours or precancerous polyps.
A gastroenterologist at the Center for Advanced Endoscopy at Beth Israel Deaconess Medical Center in Boston, Tyler M. Berzin, MD, claims that AI-enhanced colonoscopy “effectively turbocharges the physician’s ability to find even the most subtle precancerous polyps.”
The technology is made to indicate whatever the computer “sees” as odd, but it cannot take the place of a gastroenterologist’s education and experience. Even with AI, clinicians still carry out the procedure while standing by the patient’s side.
According to Prateek Sharma, MD, a gastroenterologist and professor of medicine at the University of Kansas School of Medicine in Kansas City, KS, the doctor retains complete control. Colon polyps, which are precancerous lesions in the colon, are being helped and identified by AI so that the doctor can remove them.
With polyps, size, height, and quantity are important. Lesions 10 millimetres and larger are typically removed or biopsied by a doctor.
However, there is still some disagreement over how to handle smaller polyps.
Berzin and co-authors stated in a prestigious gastroenterology publication in May 2020 that “the clinical importance of finding and eradicating small (5 to 9 mm) or diminutive (less than 5 mm) adenomas is a subject of continuous discussion.”
For instance, Berzin cites “the possibility of removing a higher number of tiny or hyperplastic polyps, which increases cost and risk, without any benefit to the patient,” as one of the potential drawbacks of utilising AI polyp tools.
Precancerous colon polyps can be easily detected and removed by skilled gastroenterologists, according to Berzin. The ability of AI computer vision tools to simultaneously evaluate every pixel of the endoscopy monitor without being distracted or worn out for even a millisecond is a major advantage for a gastroenterologist using a polyp identification tool, however.
According to Sharma, who is also the chair of the Artificial Intelligence Task Force of the American Society for Gastrointestinal Endoscopy, the benefit for patients is “another set of eyes looking for polyps and supporting the doctor.”
What It Does
Algorithms are sets of computer instructions that teach computers to distinguish between worrying and benign colonoscopy images and movies. Machine learning is the method by which AI improves over time. The device highlights a potential area of concern when an AI system detects it by framing it within a box on the screen. Aural alarms are also used by some systems.
According to Sravanthi Parasa, MD, a gastroenterologist at the Swedish Health Services in Seattle, “We are seeing greater interest in employing these algorithms since they will standardise endoscopists’ polyp diagnosis and, consequently, minimise the number of colon malignancies overlooked.”
These goods are gradually gaining popularity. Patients should inquire if their endoscopist has access to augmented diagnostic tools when arranging a colonoscopy, according to the expert.
The technology is not always correct; occasionally, a bubble in the colon, for instance, may be falsely flagged as potentially hazardous by the system. That’s just one of the reasons why physicians still have the final say in whether or not a polyp is suspicious.
Regardless of artificial intelligence, according to Berzin, an associate professor of medicine at Harvard Medical School, “colonoscopy has long been our most effective tool for preventing colon cancer, detecting precancerous polyps earlier than any other screening method.”
AI Can Be Expensive
Self-driving cars, speech recognition software, and “smart” products like smartphones, smartwatches, and speakers already use AI and machine learning. However, the application of AI in medicine is still very young. It is also pricey, like many revolutionary technologies. The necessary and pricey AI technology must be obtained, according to Sharma.
In the current health care environment, some facilities may find the expense of the algorithms to be prohibitive, acknowledges Parasa. As with other software solutions, the price is probably going to decrease as additional algorithms are introduced to the GI market.
The Prevalence of Colorectal Cancer
The fourth most frequent cancer in Americans after some forms of skin cancer is colorectal cancer. According to the CDC, it is also the fourth most common reason for cancer-related fatalities in the US. According to data from the National Cancer Institute, more than 150,000 Americans will be given a colorectal cancer diagnosis in 2022, and more than 50,000 will pass away from the disease.
The interaction between people and technological technology requires more study, according to Berzin. The most intriguing study in this area will be centred on figuring out the intricacies of “physician plus AI,” rather than comparing “physician vs AI.”
According to Parasa, there are currently at least three AI algorithms for polyp diagnosis in the United States that have received FDA approval.
“In addition, other applications that are already offered on the European market, such as polyp characterisation, may become available on the American market in the near future.”
She continues, “As the area develops, we’ll probably see more AI augmentation tools that will help us identify and diagnose GI issues in real time.” “This kind of algorithm suite will undoubtedly enhance patient outcomes and care.”
The combination of doctors and AI technology, according to Berzin, “will translate into the highest possible protection from colon cancer in the long term,” despite the fact that AI in medicine is still somewhat of a work in progress.