> ## Content Index
> Fetch the complete content index at: https://www.coios.me/llms.txt
> Use this file to discover other available public pages before exploring further.

# FAHR-Face, a face photograph foundation model, estimates biological age and survival
- URL: https://www.coios.me/fahr-face-a-face-photograph-foundation-model-estimates/
- Published: 2026-10-23T13:12:51.000Z
- Updated: 2026-10-24T06:06:05.000Z
- Description: A face photograph model fine-tuned on 34,363 patients with cancer separated mortality risk between quartiles.
- Author: Daniel Ryan
- Tags: Data, models and systems, AI in medicine, Slowing ageing itself, #peer-reviewed, #new, #item, #source-2026-10, #score-72

The study developed FAHR-Face, a foundation model pre-trained on over 40 million facial photographs. It was fine-tuned for biological age estimation using ten public image datasets. A separate fine-tuning for survival prediction used 34,363 photographs of patients with cancer, with two independent validation cohorts. For age estimation it had the lowest mean absolute error among the benchmark models tested across the lifespan. In patients with cancer, the highest-risk quartile had more than triple the mortality of the lowest after adjustment for clinical prognostic factors. Clinical testing was confined to patients with cancer, so general population performance is untested.

*Why it is interesting: Face photographs are cheap and widely captured, and here they predicted mortality independently of clinical prognostic factors. The work also separates facial ageing from disease-related mortality risk.*

Source

[The Lancet Digital Health, 19 October 2026](https://doi.org/10.1016/j.landig.2026.101056?ref=coios.me)

Paper

Foundation artificial intelligence models for health recognition by use of face photographs (FAHR-Face): a model development and validation study

Authors

Haugg F, Lee G, He J, Nürnberg L, Bontempi D, Bitterman DS et al.

DOI

10.1016/j.landig.2026.101056

Type

Peer-reviewed article

Design

Foundation model development and validation: pretraining on >40 million facial photographs, fine-tuning on ten public image datasets and 34,363 photographs of patients with cancer, tested in two independent cancer cohorts with multivariable Cox survival analysis

Verdict

New finding

Driver

[AI in medicine](https://www.coios.me/d-ai-medicine/)

Driver

[Slowing ageing itself](https://www.coios.me/d-ageing/)

Free full text

[Free to read at the publisher](https://doi.org/10.1016/j.landig.2026.101056?ref=coios.me)