A simple selfie can have surprising clues about overall health and life expectancy, especially for people with cancer.
Researchers at the Mass General Brigham Foundation in Boston have developed artificial intelligence models that they say can predict a cancer patient's biological age and chances of survival, using just one picture of his face.
The study, which was published the day before yesterday, Monday, in the journal "The Lancet Digital Health", indicates that facial analysis may become an additional tool that doctors can use when assessing a patient's chances of survival, and when determining treatment methods that may achieve the best results for cancer patients, according to what was reported by the American "CNN" network.
“I was really surprised when I discovered that we could detect indicators of death risk from just a facial image,” said Dr. Raymond Mack, study author, radiation oncologist at MassGeneral Brigham Cancer Institute, and a faculty member in the institution’s Artificial Intelligence in Medicine program.
The AI models were trained on more than 40 million images of faces, and they evaluate different aspects of the face to estimate the cancer patient's general health and biological age. Unlike chronological age, which measures the number of years a person has lived, biological age reflects how the body ages at both the cellular and physical levels.
“The age estimation algorithm looks at the area around the temples, the forehead area, above the eyes, and then what we call the nasolabial crease,” Mack said, explaining the parts of the face that AI models examine to estimate biological age. It is the line that extends from the nose to the corner of the mouth.
“In contrast, the death risk prediction algorithm looks at a completely different part of the face, examining the bridge of the nose and the area under the eyes,” Mack added.
Hippocrates... and artificial intelligence
Surprisingly, the same areas of the face that the AI model analyzes to predict biological age and chances of survival are the parts that the Greek physician and philosopher Hippocrates referred to in the book “Chapters on Disease Prediction,” written around 400 BC, when he described “a sharp nose, sunken eyes, and shrunken temples” as signs of muscle wasting and a high risk of death. “The thought that our algorithms might replicate thousands of years of Hippocratic wisdom was amazing to us,” Mack said.
The AI algorithms were not familiar with the ancient texts of Hippocrates. “Our algorithms are trained exclusively on images of faces, and then on whether the person has died or not, or on their age,” Mack said. He added: “It does not take into account texts, knowledge or wisdom accumulated on the Internet, so it was surprising that Hippocrates’ ideas resonated across thousands of years in our algorithms.”
Mack and his colleagues first developed a powerful AI model called Foundation Artificial Intelligence for Health Recognition-Face (FAHR-Face), which was trained on more than 40 million images of faces representing a variety of ages and ethnic backgrounds.
The model was then fine-tuned and used as a basis for building two other models capable of carrying out specific tasks. The first is the “FAHR-FaceAge” model, which estimates a person’s biological age based on a single photo, and the second is the “FAHR-FaceSurvival” model, which predicts a cancer patient’s chances of survival, by combining facial features and other clinical information.
Using the FAHR-FaceSurvival model, the researchers found that of the 122 cancer patients the model classified as being at high risk of death, more than half died; That is, 52.5 percent, within 90 days of the evaluation, consistent with the model’s expectations. In contrast, the percentage of those who died within 90 days did not exceed about 2 percent out of 975 patients in the group that the model classified as the lowest-risk category.
When evaluating the FAHR-FaceAge model, researchers found that 21 percent, out of 1,060 patients whose biological age exceeded their chronological age by more than 20 years, died within 90 days. In contrast, about 6 percent died, out of 4,195 patients whose biological age was at least ten years younger than their chronological age, during the same period.
The researchers found that cosmetic changes to the face, such as surgery, cosmetic injections such as Botox, or the use of cosmetics, did not affect the performance of the artificial intelligence models.
The study authors stress that the AI models they developed are not intended to replace doctors or existing clinical assessments. Instead, they describe the technology as an emerging tool that may help doctors in the future assess a patient's health status or make decisions about treatment options.
The Lebanese MTV television screen is showing the series “Bahou El Sitt Lamia”, which combines social drama and comedy. Its events take place in a small hotel in Beirut run by “Sitt Lamia”, whose role is played by Rola Shamieh, along with a group of actors, including Joe Sader, Naim Halawi, and Nour Ghandour.
Joe Sader's participation in this work represents his first experience starring in a comedy series, as he previously played many roles in various dramatic works, in which he moved between romance, suspense, and excitement.
In an interview with Asharq Al-Awsat, he said that this experience was new to him, and that he hesitated to engage in it before the producer of the work, Marwan Haddad, convinced him that it suited him completely. He adds: “Haddad has a prior vision of the characters in the series he produces, and he has a sensitive sense that allows him to draw a picture of the actor who succeeds in performing a particular role. When I sensed his insistence, I yielded to the idea and saw in it a new challenge that I would face in my acting career.”
Sader points out that his role in the series is very similar to his real personality. He explains: “(Shadi) is similar to me in his love of sports and fitness, as well as his romantic style in his romantic relationships. Those around me also know very well how much I love funny jokes and comedic situations. All of these qualities are present in the personality, which encouraged me even more to engage in the experience.”
Regarding how he prepared for a role that he had never performed before, he says: “The series is not a comedy par excellence, as it also has social aspects. Therefore, I mixed the two schools, and the presence of Rula Shamiya and Naeem Halawi helped me a lot. They are masters of satirical comedy, and their success in this field attests to them. “My generation was influenced by their performance.”
Does he consider himself to have passed this exam? He replies: “I think so, as I see, through comments on social media, as well as those I meet, that I have accomplished the task properly.”
Followers of the work, which is shown daily at seven in the evening except on weekends, will notice the abundance of spontaneous and improvisational situations that give it a special character. Sader explains: “This impression is correct, because Rula and Naeem do not always adhere to the written text, but rather practice a kind of spontaneous improvisational acting. This was a real challenge for me, as I was required to keep up with them. We often laughed genuinely because of a situation we faced, which gave the work smoothness and spontaneity.”
As for whether he is ready to repeat the experience in a comedy series that brings him together with the two heroes themselves, he replies: “I will not hesitate to do that, especially with Rola and Naeem. I grew up on their works, and was impressed by their funny performances and wit. But what we lack today is the production of works of this kind, as it has become rare to see them on our local screens.”
Sader expresses his admiration for a number of comedians in Lebanon, and puts Fadi Raidy at the forefront, along with George Khabbaz, Adel Karam, and Abbas Shaheen, in addition to the heroes of the series “SLCHI.”
He believes that he is an actor different from others of his generation, as he has charted a special path for himself. He adds: “Since my beginnings, I have refused to fall into the trap of imitation, and I have always been keen to present my own style of performance, lest I be compared to another actor. There are acting professors, but I chose to follow a path that resembles me.”
As for whether he will move from drama to comedy, like well-known names who became famous in the comedy field before taking drama as a fixed space in their career, he says: “I think that the actor should diversify his work and not limit his performance to a specific dramatic genre. Comedy, as everyone knows, is the most difficult in the world of drama. I will not limit myself to a specific type of series, and I will always choose what convinces me and makes me feel comfortable.”
Joe Sader turns to his family first to identify errors that may appear in his performance. He says: “In (Bahi El-Sitt Lamia) I was afraid of falling into exaggeration in my comedic performance, but there is always someone who alerts me to right and wrong, and foremost among them are my parents. I listened to their comments carefully, and Rula and Naeem were constantly guiding me so that I would not make these types of mistakes.”
He believes that the dramatic arena is experiencing difficult conditions today due to the war in Lebanon and the wars taking place in the region. This was reflected in local productions and television stations, with a noticeable decline in commercial advertising revenues. He adds: “If it were not for a few producers who still trust the Lebanese actor and director, we would not have been able to continue the journey.”
After “Bahi El-Sitt Lamia,” Joe Sader is expected to appear in a new series called “The Drunkenness of Love,” which was filmed a long time ago and is expected to see the light soon. He says: “Mohamed Al-Ahmad and Haya Maraachli participate in the film, and I embody a different character from what I presented previously. “I play the role of a police investigator who follows up on incidents and crimes related to a museum theft.”
As for the possibility of his participation in Turkish works dubbed in the Lebanese dialect, which have been popular recently, he says: “I do not care much about this type of participation, but rather I prefer the view in which the viewer senses my presence with sound and image.”
