In Brief
A new machine learning method developed by Insilico Medicine can predict one's age using data from blood tests, and it can determine important biomarkers that help track the effectiveness of aging therapies.

A bioinformatics company called Insilico Medicine has just announced Aging.AI, an online platform that guesses a person’s age using data from blood tests.

In the study to be published in Aging, the researchers designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth and structure to predict age using a basic blood test.

The Aging.AI logo.
The Aging.AI logo.

This ensemble was able to determine an individual’s age around 80% of the time. The study also determined the five most important markers for predicting human chronological age: albumin, glucose, alkaline phosphatase, urea, and erythrocytes.

Determining these biomarkers is important, since one of the major impediments in aging research is the absence of a set of biomarkers that may be measured to track the effectiveness of therapies.

Since neural networks often require large data sets to improve performance, the study used 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex.

The ensemble was then used to power Aging.AI, an online platform where any user may input data from their own blood test, and the DNN will guess age and sex based on the data.