mlsanityData to 14 Oct 2026

Foundation Artificial Intelligence Models for Health Recognition Using Face Photographs (FAHR-Face)

arXiv 2506.14909 ↗First posted 17 Jun 2025eess.IV cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Background: Facial appearance offers a noninvasive window into health. We built FAHR-Face, a foundation model trained on >40 million facial images and fine-tuned it for two distinct tasks: biological age estimation (FAHR-FaceAge) and survival risk prediction (FAHR-FaceSurvival). Methods: FAHR-FaceAge underwent a two-stage, age-balanced fine-tuning on 749,935 public images; FAHR-FaceSurvival was fine-tuned on 34,389 photos of cancer patients. Model robustness (cosmetic surgery, makeup, pose, lighting) and independence (saliency mapping) was tested extensively. Both models were clinically tested in two independent cancer patient datasets with survival analyzed by multivariable Cox models and adjusted for clinical prognostic factors. Findings: For age estimation, FAHR-FaceAge had the lowest mean absolute error of 5.1 years on public datasets, outperforming benchmark models and maintaining accuracy across the full human lifespan. In cancer patients, FAHR-FaceAge outperformed a prior facial age estimation model in survival prognostication. FAHR-FaceSurvival demonstrated robust prediction of mortality, and the highest-risk quartile had more than triple the mortality of the lowest (adjusted hazard ratio 3.22; P<0.001). These findings were validated in the independent cohort and both models showed generalizability across age, sex, race and cancer subgroups. The two algorithms provided distinct, complementary prognostic information; saliency mapping revealed each model relied on distinct facial regions. The combination of FAHR-FaceAge and FAHR-FaceSurvival improved prognostic accuracy. Interpretation: A single foundation model can generate inexpensive, scalable facial biomarkers that capture both biological ageing and disease-related mortality risk. The foundation model enabled effective training using relatively small clinical datasets.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 19 of its 75 references that are in the map. First posted on arXiv 17 Jun 2025.

Builds on

19 of 75 references.

References · 19

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  5. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  6. 2016Densely Connected Convolutional Networkscited by 6,720
  7. 2017Squeeze-and-Excitation Networkscited by 4,789
  8. 2017VGGFace2: A dataset for recognising faces across pose and agecited by 914
  9. 2017Age Progression/Regression by Conditional Adversarial Autoencodercited by 429
  10. 2022Masked Autoencoders As Spatiotemporal Learnerscited by 429
  11. 2015MADE: Masked Autoencoder for Distribution Estimationcited by 307
  12. 2021WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognitioncited by 223
7 more
  1. 2019RetinaFace: Single-stage Dense Face Localisation in the Wildcited by 165
  2. 2023MiVOLO: Multi-input Transformer for Age and Gender Estimationcited by 42
  3. 2022How to Understand Masked Autoencoderscited by 38
  4. 2023A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmarkcited by 19
  5. 2020Investigating Bias in Deep Face Analysis: The KANFace Dataset and Empirical Studycited by 18
  6. 2017Quantifying Facial Age by Posterior of Age Comparisonscited by 15
  7. 2020Understanding bias in facial recognition technologiescited by 15

Built on by

Not cited yet.