Facial temperature patterns could help detect MASLD: Study

Machine learning tool analyzes thermal images to detect signs

Marisa Wexler, MS avatar

by Marisa Wexler, MS |

Share this article:

Share article via email
A person wearing a baseball cap speaks using a megaphone.

Metabolic dysfunction-associated steatotic liver disease (MASLD), a form of fatty liver disease, and other aging-related metabolic disorders may be detectable using computer-based analyses of the temperature of a person’s face, a study shows.

“We hope to apply thermal [temperature] facial imaging in clinical settings, as it holds significant potential for early disease diagnosis and intervention,” Jing-Dong Jackie Han, the study’s senior author and a professor at Peking University in Beijing, China, said in a press release.

The study, “Thermal facial image analyses reveal quantitative hallmarks of aging and metabolic diseases,” was published in Cell Metabolism.

MASLD, previously known as nonalcoholic fatty liver disease, is marked by the abnormal accumulation of fat in the liver that’s related to aging-exacerbated metabolic conditions, such as obesity or type 2 diabetes.

Liver fat buildup can lead to inflammation and scarring that damages the organ. But signs of fatty liver usually don’t become apparent until the disease has progressed, so there’s a need for earlier detection methods.

Recommended Reading
A person wearing a baseball cap speaks using a megaphone.

US government to launch 1st study of disease prevalence in MASLD

Body temperature decreases with age

Body temperature can say a lot about a person’s health and age. It’s been well established that as a person gets older, core body temperature tends to decrease. “High body temperature often implies high metabolic rates and can be induced by psychological and metabolic stress,” the researchers wrote.

Han and her colleagues explored how thermal imaging might be used to evaluate aging and detect metabolic conditions such as MASLD.

They conducted thermal imaging of the faces of more than 2,800 adults of Han Chinese descent, with ages ranging from 21 to 88 and balanced in terms of sex. The group included people with a range of metabolic disorders.

The team used machine learning, a form of artificial intelligence that uses algorithms to analyze data and make predictions, to interpret the data.

As body temperature is known to change with age, the researchers first explored how thermal imaging of the face changes with age. They found that the temperature of certain parts of the face, including the nose, cheeks, and eyebrows, showed clear temperature reductions with age.

“Our reconstructed average face profiles reveal previously unrecognized patterns of facial temperature change during aging,” the scientists wrote.

Using their findings, the researchers were able to calculate a thermal age score reflecting how facial heat signatures would be expected to change in an average person. They could then calculate the difference between the person’s predicted age based on thermal facial imaging and the person’s actual age.

The team found that metabolic diseases, including diabetes and high blood pressure, accelerated a person’s thermal age score, with people with these conditions showing significantly higher thermal ages than those without these diseases.

High blood pressure and MASLD also showed specific thermal facial patterns, which could accurately identify men with high blood pressure and both men and women with MASLD with about 80% accuracy.

Those findings imply that “there is a great potential for using [thermal face analysis] to detect these diseases,” the researchers wrote.

Recommended Reading
A person wearing a baseball cap speaks using a megaphone.

Global Fatty Liver Day to raise awareness of ‘concealed epidemic’

Exercise alters results

The researchers then recruited 30 volunteers for a pilot experiment to see if exercising could affect thermal age. Over the course of two weeks, 23 of the volunteers undertook a two-week exercise regime using a jump rope, while the other seven, used as controls, did not.

Those who completed the exercise program showed a significant decrease, by an average of five years, in the difference between their temperature-predicted ages and their actual ages, while the control group experienced a slight increase in this gap.

“Our current results indicate that thermal facial aging can be altered by a 2-week daily jump rope exercise,” the researchers wrote. They stressed that their experiment was limited by its short duration and small number of volunteers, and said there’s a need for additional studies to verify and expand on the findings.

Another notable limitation of the study, the team noted, was that it included only people of a single ethnic group. Most participants also lived in roughly the same area with similar weather patterns.

Further studies involving different ethnic groups and people living in different climates will be needed to assess whether thermal facial analysis can be used “as an aging and health assessment across diverse human populations,” the researchers concluded.

“Aging is a natural process,” said Han. “But our tool has the potential to promote healthy aging and help people live disease-free.”