Face photo-based age acceleration predicts all-cause mortality and differs among occupations
Kiraly, B.; Fejes, I.; Kerepesi, C.
Show abstract
While scientists argue what aging is and what drives aging, it is widely accepted that our face changes drastically with age and that mortality increases in late life. We hypothesize that people of the same age can be biologically older than others and that the human face may reflect accelerated molecular aging. To test this hypothesis we examine the associations of face photo-based age acceleration with mortality and lifestyle. For this purpose, we trained and tested artificial intelligence models on 442,110 photos of famous people. We found that face photo-based age predicts all-cause mortality for middle-aged and older individuals meaning that those age faster based on their face photo die sooner. We also found that, based on face photos, sport is the slowest aging occupation among famous people consistently to previous findings showing the benefits of exercise to epigenetic aging. Overall, we demonstrate that the face photo-base age model approaches biological age in some extent and provides a low-cost and fast complementary measurement for personalized medicine, as well as aging and rejuvenation studies. The model is available for demonstration and academic research purposes at https://photoage.sztaki.hu/.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- AltumAge: A Pan-Tissue DNA-Methylation Epigenetic Clock Based on Deep Learning 94%
- scAgeClock: a single-cell transcriptome based human aging clock model using gated multi-head attention neural networks 93%
- Multinational evaluation of anthropometric age (AnthropoAge) as a measure of biological age in the USA, England, Mexico, Costa Rica, and China: a population-based longitudinal study 90%
Similar papers in this journal
- Deep longitudinal phenotyping of wearable sensor data reveals independent markers of longevity, stress, and resilience 93%
- Healthspan pathway maps in C. elegans and humans highlight transcription, prolifera-tion/biosynthesis and lipids 92%
- A tandem segmentation-classification approach for the localization of morphological predictors of C. elegans lifespan and motility 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.