· 02 February 2018

Artificial intelligence can already predict if you are going to have a long and healthy life

Por cenie
La inteligencia artificial ya puede predecir si vas a tener una vida larga y saludable - Innovación, Investigación

With a simple blood sample, that`s how everything begins. Yes, one of those small test tubes that are used to perform a lot of routine medical tests. Then we must take into account some parameters such as diet, exercise, environmental exposures ... All well mixed in a cocktail of data that will be analyzed by the new artificial intelligence (AI) that are developing rapidly and Voilà! An aging clock sufficiently developed to predict whether a person will have a long and healthy life.

This AI developed by the Centre for Healthy Aging at Copnehague University may reveal, according to researchers in a study published in the journal The Journals of Gerontology, whether changes in lifestyle and medications can increase their chances of living.

"Historically, a lot of money has been spent trying to identify biomarkers of aging and, in general, all attempts have failed. But by using these large sets of complex data that contain a series of interacting parameters, we can create a fairly accurate picture of a person's biological age, "says Dr. Morten Scheibye-Knudsen.

The developed programs -which they have called Aging.AI- help determine if an individual's body age corresponds to their real age, in addition to determining the risk of developing diseases related to aging. "If the algorithm finds that you are younger than the age that shows your ID, then you have a better chance of living a long life," according to the study.

Scheibye-Knudsen and his team have collaborated with scientists from the United States, South Korea and Canada to develop an algorithm capable of determining the age - "fairly close to the real" - of a person based on a standard blood sample, like those that your doctor asks you or that are made every morning in hospitals.

"Artificial intelligence is as good at predicting your age as if you were looking at a picture of the person and you had to guess the age. That is, it reaches an age that is quite close to the age taht your driver's shows. But what it really does is measure his biological age, which may be different from his real age, "adds Morten Scheibye-Knudsen.

The biological age can be different from the real age

The American company Insilico Medicine, specialized in AI, has been the one that has financed the project. "The special and impressive thing about this method is that it uses measurements obtained from standard blood samples, which are cheap to make. There are other ways to measure a person's age, but they are relatively expensive and require special equipment, "the doctor and professor at the University of Copenhagen explains in a statement.

"Right now we still don`t know at what exact age we will arrive, although we can create hypotheses and assumptions. What we do know is that age is the biggest risk factor when it comes to diseases such as cancer, dementia and cardiovascular disease. We would like to be better at curing these ailments, and Aging.IA can help us to do it, "says Scheibye-Knudsen.

The developed artificial intelligence seems, according to the researchers, to the best chess computer in the world. Aging.AI is based on deep neural networks, which is a kind of learning algorithm similar to the one used to develop the revolutionary AlphaZero, which has just defeated the best chess computer in the world without being taught anything other than the rules of the game.

Researchers have given the algorithm the age and blood sample responses of 130,000 people in Canada, South Korea and Eastern Europe. Each sample contains 21 parameters typically measured in blood samples such as cholesterol, markers of inflammation (one of the most reliable markers of cardiovascular risk), the hemoglobin count and albumin (a protein produced by the liver) in the blood.

Source: La Vanguardia

Under the framework of: Programa Operativo Cooperación Transfronteriza España-Portugal
Sponsors: Fundación General de la Universidad de Salamanca Fundación del Consejo Superior de Investigaciones Científicas Direção Geral da Saúde - Portugal Universidad del Algarve - Portugal