How could a person not be curious about a New York Times article titled “The A.I. Diet” — A.I. of course being short for Artificial Intelligence. This was written by cardiologist and molecular medicine professor Eric Topol, who is also executive vice president of Scripps Research.
He used a smartphone app to keep track of every food, beverage, and medication he ingested, along with their amounts, for two weeks, along with charting all his sleep and exercise times. Various sensors and lab tests were also involved. If this sounds extreme, consider that more than 1,000 other subjects were doing it too, for a study on how to achieve a long and healthy lifespan. As the saying goes, hilarity ensued. Dr. Topol writes,
In the sweets category: Cheesecake was given an A grade, but whole-wheat fig bars were a C -. In fruits: Strawberries were an A+ for me, but grapefruit a C. In legumes: Mixed nuts were an A+, but veggie burgers a C.
This abundantly credentialed academic concluded, “[W]e know surprisingly little about the science of nutrition.” Much can be said about that proposition, one way or another, but in Topol’s estimation, the results point to a “central flaw in the whole premise […] the idea that there is one optimal diet for all people.” Furthermore, we have A.I. to thank for the ability to figure out even this much. Only humans would be “simplistic and naive” enough to believe in the universal optimal diet which the author says is “both biologically and physiologically implausible.” Why?
It contradicts the remarkable heterogeneity of human metabolism, microbiome and environment, to name just a few of the dimensions that make each of us unique. A good diet, it turns out, has to be individualized.
The bottom line is, some things work for some people and other things work for other people. Okay, great, why don’t we just figure out, one by one, what each individual ought to be eating in order to live their best life? Because…
Coming up with a truly personalized diet would require crunching billions of pieces of data about each person. In addition to analyzing the 40 trillion bacteria from about 1,000 species that reside in our guts […] it would need to take into account all of the aspects of that person’s health, including lifestyle, family history, medical conditions, immune system, anatomy, physiology, medications and environment.
In other words, pretty much like a life insurance application.
The earliest serious results in this area were obtained by studying glycemic responses in relation to developing diabetes, and again, different people are different. But the pioneering Weizmann Institute of Science studies of blood glucose levels in relation to substances consumed showed one thing: “the first objective proof that we do indeed respond quite differently to eating the same foods in the same amounts.” This conclusion was derived from more than a million and a half glucose measurements. As Topol noted, “That’s a big data set.”
There are ways to gather the most elusive information, and ways to make sense of it, to the point where he calls it “eminently doable” and offers both a warning and a positive prediction:
A number of companies have been marketing “nutrigenomics,” or the idea that a DNA test can provide guidance for what foods you should eat… but they don’t have the data to back their theory up… This would require developing an artificial intelligence more sophisticated than anything yet on the market.
In the next few years, you could have a virtual health coach that is deep learning about your relevant health metrics and providing you with customized dietary recommendations.
Your responses and feedback are welcome!
Source: “The A.I. Diet,” NYTimes.com, 03/02/19
Image by Steve/CC BY-SA 2.0