Artificial Intelligence uses selfies to predict cancer survival.

Doctors often begin their examinations with the "eye test," a quick assessment of whether a patient appears older or younger than their age, which can influence medical decisions. This intuitive diagnosis could soon be improved with artificial intelligence (AI).

A deep learning algorithm was featured Thursday in The Lancet Digital Health: FaceAge that turns a simple headshot into a number that more accurately reflects a person's biological age , rather than the birth date on their medical record.

Trained with tens of thousands of photographs, he determined that cancer patients were, on average, five years biologically older than their healthy counterparts.

The study authors say this could help doctors decide who can safely tolerate harsh treatments and who would do better with milder ones.

“Our hypothesis is that FaceAge can be used as a biomarker in cancer care to measure a patient’s biological age and help doctors make these difficult decisions,” said Raymond Mak, a study co-author and oncologist at Mass Brigham Health, a Harvard-affiliated health system in Boston.

In the hypothetical case of two patients, one who is agile at 75 years old and has a biological age of 65, and the other who is more fragile, 60 years old but whose biological age is over 70, aggressive radiotherapy may be more appropriate for the former, but risky for the latter.

The same logic can be applied to decisions related to heart surgery, hip replacement, or palliative care.

There is growing evidence that humans age at different rates, depending on their genetics, stress, exercise, and habits, such as smoking or drinking alcohol.

While expensive genetic testing can reveal how DNA degrades over time, FaceAge promises to penetrate the body with just a selfie.

The model was trained on 58,851 portraits of presumably healthy adults over 60 years old, taken from public databases.

It was then tested on 6,196 patients receiving treatment in the United States and the Netherlands using photos taken before their radiation therapy. Patients with malignant tumors appeared on average 4.79 years older biologically than their chronological ages.

Among cancer patients, a higher FaceAge score predicted worse survival, even after adjusting for age, sex, and tumor type. The odds dropped dramatically for those whose biological age was over 85.

FaceAge identifies signs of aging differently than people generally do. For example, gray hair and baldness are less important than subtle changes in facial muscles.

Six doctors were asked to examine photographs of the faces of patients with terminal cancer and determine which of them would die within the next six months. With FaceAge data in hand, their predictions improved significantly.

The model also confirmed a now-famous internet meme, when he estimated the biological age of jovial American actor Paul Rudd to be 43 based on a photo taken when he was 50.

AI tools have come under scrutiny for not paying enough attention to non-white people.

Mak said preliminary tests revealed no significant racial bias in FaceAge's predictions; however, the group is training a second-generation model on 20,000 patients.

They also test how factors such as makeup, cosmetic surgery, and lighting variations can fool the system.

Ethical debates are evident: an AI that can read biological age from a selfie could be a boon for doctors, but also a temptation for life insurers or companies looking to measure risk.

Knowing that the body is biologically larger than previously thought can motivate positive changes in health or sow anxiety, another dilemma on the table.

The researchers plan to open a publicly accessible FaceAge portal where people can upload their portraits to participate in a research study to validate the algorithm. Commercial versions for physicians will follow, but only after further validation.

CT