Artificial Intelligence (AI) is opening up new possibilities in the healthcare sector. In this regard, researchers have developed an AI model capable of estimating a person’s biological age by analyzing images of their face. Furthermore, this technology could assist in assessing the risk of mortality among cancer patients.
The AI model was developed by researchers at Mass General Brigham in Boston, USA. According to the research, cues derived from standard facial images can be used to gain insights into a patient’s health status and potential clinical outcomes. It should be noted, however, that this technology is not a substitute for professional medical examinations or clinical decision-making.
What is the FAHR-Face AI model?
The researchers have named this new AI model “FAHR-Face.” It has been trained on a dataset comprising over 40 million facial images, with the objective of analyzing health-related indicators based on facial features.
The technology has two primary applications: first, estimating an individual’s biological age based on their facial image; and second, assessing indicators related to survival probabilities and mortality risk in cancer patients.
Biological age relates to the physical condition of an individual’s body, whereas chronological age reflects the number of years that have passed since birth—the two may differ. It is important to note, however, that a facial image alone cannot provide a definitive assessment of an individual’s overall health status.
Which facial areas provide these indicators?
According to the research, specific facial features play a significant role in the AI model’s analysis. In the context of assessing risk for cancer patients, cues derived from areas around the region beneath the eyes and the upper part of the nose were found to be particularly useful.
By analyzing these features, the model generates estimates regarding potential health outcomes for patients. In the future, such technology could assist physicians in better understanding a patient’s condition and provide additional insights to inform treatment-related decisions. How effective was the AI model for cancer patients?
According to the report, researchers evaluated the model using data from approximately 43,000 cancer patients. It was subsequently tested on a separate group of 3,820 patients.
In this test, the AI model categorized 122 patients into the highest-risk group. Among these, 52 percent passed away within 90 days. In contrast, 975 patients were placed in the lowest-risk group, with only 2 percent of them passing away during the same period.
These results indicate a significant difference in mortality rates between the high-risk and low-risk groups identified by the model. However, these findings do not imply that the AI can definitively predict an individual’s death; rather, it is a research-based tool designed to estimate risk.
Could AI also prove helpful in treatment?
According to researchers, analyzing facial images could be useful for assessing risk in cancer patients. This could provide doctors with additional support in monitoring patients and determining the need for further medical investigations.
Nevertheless, before this technology can be widely adopted in clinical practice, it is essential to evaluate its accuracy, its effectiveness across diverse populations, and its utility in real-world settings. Decisions regarding patient treatment should not be based solely on AI-generated estimates.
Conclusion: AI models like FAHR-Face offer new possibilities in the healthcare sector. While the technology—which uses selfies to estimate biological age and cancer-related risks—could prove valuable in the future, it should not be considered a substitute for routine medical check-ups or professional medical advice.
