FaceAge biometric tool assesses how old you look, considers your chances of survival

New research published in medical journal The Lancet aims to develop and validate “a deep learning system to estimate biological age from easily obtainable and low-cost face photographs.”

The machine learning technology, dubbed FaceAge, is intended to be used by medical professionals to evaluate the physical appearance and biological age of patients from biometric data, and specifically to analyze signs of cancer.

In other words, it decides how old you look and what that says about your health – and how long you’re likely to survive.

While that may not sound like an appealing proposition to many (and, indeed, may sound like a potential ethical nightmare to some), FaceAge can reportedly predict whether or not cancer patients are likely to survive treatment, enabling doctors and caregivers to optimize palliative end-of-life care for terminal cases.

“In these patients, clinical prediction models can help to improve physicians’ decision making as to whether or not to administer treatment, as well as the appropriate treatment intensity, both of which are largely a function of a physician’s impression of overall prognosis, performance status, and frailty.”

The research team, based out of Mass General Brigham in Boston reports that FaceAge was trained on biometric data from 58,851 presumed healthy individuals aged 60 years or older: 56 304 individuals from the IMDb-Wiki dataset (training) and 2547 from the UTKFace dataset (initial validation). Clinical validation compared data from 6196 patients with cancer diagnoses from two institutions in the Netherlands and the U.S. FaceAge estimates in these cancer cohorts were compared with a non-cancerous reference cohort of 535 individuals.

The biometric diagnostics tool appears to work well enough. “FaceAge showed significant independent prognostic performance in various cancer types and stages,” the research says. “We found that, on average, patients with cancer look approximately 5 years older than their chronological age and have a statistically higher FaceAge compared with clinical cohorts of patients without cancer who are treated for conditions that are benign or precancerous.”

“Our results suggest that a deep learning model can estimate biological age from face photographs and thereby enhance survival prediction in patients with cancer.”

Although doctors already evaluate whether people with cancer should undergo treatment, the introduction of biometrics algorithms into the decision-making process raises troubling questions. The authors recognize the need for further research using larger cohorts before their idea can be used in clinical settings. Yet there is little reason to doubt their claim that “FaceAge could be used to translate a patient’s visual appearance into objective, quantitative, and clinically valuable measures.”

The risk is that objectivity can be asserted for ideological ends to bring the technology out of the hospital. As the specter of eugenics pops its nasty head into current political discussions about autism and public health, there is potential danger in a technology that makes it easier for anyone claiming authority to decide who should live or die.

As is the case with all so-called AI, while algorithms may come with good intentions, they are unlikely, in the end, to be applied exclusively by the benevolent.

Article Topics

age estimation  |  age verification  |  biometrics  |  face biometrics  |  FaceAge