Although I frequently make fun of artificial intelligence due to some of its inherent limitations for certain situations, the fact remains that when AI works well, it works extremely well. Additionally, as big data gets bigger, AI becomes stronger. This is proven by current trends in healthcare technology.
No one would deny that the screening colonoscopy is an invasive procedure. That is exactly why many people search for every opportunity to use some kind of a substitute process. Researchers have been attempting to satisfy this demand. An Israeli health-tech company, Medial EarlySign has created a blood test called ColonFlag that capitalizes on AI and big data to determine colon cancer risk. The blood test performs rather well as Paul Tullis reports (“Colonoscopy? I’ll Take a Blood Test” Bloomberg Businessweek. January 15, 2018, pp. 24–25):
“ColonFlag . . . predicts colon cancer twice as well as the fecal exam that’s the industry-standard colonoscopy alternative. . . . ColonFlag compares new blood tests against a patient’s previous diagnostics, as well as Medial’s proprietary database of 20 million anonymized tests spanning three decades and three continents, to evaluate the patient’s likelihood of harboring cancer.” (p. 24)
Another excellent example of AI’s value in healthcare is mammogram interpretation. Houston Methodist Hospital is using AI software for this task with much more effectiveness compared to human doctors. Authors Frank Malcolm, Paul Roehrig, and Ben Pring summarize (What to Do When Machines Do Everything: How to Get Ahead in a World of AI, Algorithms, Bots, and Big Data. Hoboken, New Jersey: John Wiley & Sons, Inc., 2017):
“AI software . . . interprets [mammogram results] 30 times faster than doctors and with 99% accuracy. By contrast, mammograms reviewed by humans result in unnecessary biopsies nearly 20% of the time.” (p. 2)
You’ve heard of safety in numbers. AI and big data are proving the concept in exciting new dimensions.