Colon cancer is often preventable with timely screening. One way to be screened is through a procedure called colonoscopy, where a physician examines your colon with a camera to look for precancerous polyps. Unfortunately, some physicians are not as thorough as they should be and overlook growths they should be removing. It might be time to ask whether your doctor is ready to get help from AI.
To understand why AI might help, imagine you are choosing between two physicians to perform your colonoscopy. After explaining the procedure, the first reassures you not to worry, because pre-cancers are rare: “I discover polyps in patients like you only 10% of the time, and then remove them before they become cancerous.” The 10% figure is meant to reassure you about how unlikely you are to have cancer or pre-cancer growing in your colon.
The second physician also tries to reassure you, but with a notable difference – this physician explains that he discovers polyps “in patients like you 30% of the time.”
All else equal, which physician would you choose?
At first glance, it might feel like the first physician is best, given how much lower the risk of pre-cancer in their patients. However, I recommend you choose the second one. You see, some physicians are too quick in scanning the colon for polyps. Maybe they are not particularly thorough by nature. Or maybe they feel pressure to maximize their revenue by taking less time with their patients. Whatever the case may be, these physicians are more likely to miss polyps, and thus their patients are more likely to develop cancer later.
Here is a picture illustrating the relationship between polyp (i.e. adenoma) detection and the chance of later being diagnosed with colon cancer (what the cool kids call PCCRC: post-colonoscopy colorectal cancer). As a physician’s adenoma detection rate, or ADR, increases, the chance of their patients later experiencing a diagnosis of colon cancer declines.
Association Between Physician Adenoma Detection Rate and Risk of Postcolonoscopy Colorectal Cancer
In other words, physicians who find fewer polyps are missing growths they should be removing.
What should we do about these hasty colonoscopists? For policymakers, this finding suggests that quality metrics for physicians performing colonoscopies should identify ones who fail to uncover polyps very often in their patients. In addition, we should publish polyp detection rates, so patients can make informed decisions about which doctors they will seek out to perform this lifesaving test.
But there’s something else for patients to do. Ask your doctors whether they use AI assistance during the procedure. Studies in Italy and the U.S. show a significant increase in polyp detection when physicians use AI assistance. Essentially, they have a computer scan the camera images, live, and highlight suspicious areas for them to look at more closely.
I expect that use of AI-based computer copilots will soon become routine. In the meantime, you should consider asking your doctor whether they use such a program. Even the best of doctors makes mistakes. As any airline will tell you, sometimes it pays to have copilots.
Read the full article here