Face2Bone explainable AI model predicts osteoporosis risk from facial images in proof of concept study
We created and validated the first explainable deep learning model for osteoporosis risk classification using facial images. Facial characteristics associated with bone loss represent changes to the skeleton that are expected with normal aging. This non-invasive technology allows for opportunistic screening and early intervention.
Authors
- Yang Song (ORCID: https://orcid.org/0000-0001-7438-6290)
- Liang Qing (ORCID: https://orcid.org/0000-0001-7644-5629)
- Zewen Shi
- Xianjun Chen (ORCID: https://orcid.org/0000-0003-2041-2638)
- Fang Yang (ORCID: https://orcid.org/0000-0002-8662-9998)
- J. Z. Zhao
- Qingjiang Pang
Institutions
- Ningbo University (CN)
- Ningbo No. 2 Hospital (CN)
- Wuhan Union Hospital (CN)
- Ningbo No.6 Hospital (CN)
- Hangzhou Medical College (CN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2025-11-20
- DOI
- https://doi.org/10.1038/s41598-025-20462-3
- Primary Topic
- Bone health and osteoporosis research
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Hwa Mei Hospital, University of Chinese Academy of Sciences