An innovative AI-based approach to bone mineral density assessment in psoriasis patients using routine chest X-rays.

Authors

Machiko Kamara, Hirohito Hirata, Tadatsugu Morimoto, Atsushi Kawaguchi, Yoichi Sato, Kazunari Sugita

Published in

European journal of dermatology : EJD. Volume 36. Issue 4. Pages 324-331. Aug 01, 2026.

Abstract

Psoriasis is a chronic immune-mediated inflammatory disease that affects both the skin and musculoskeletal system. Recent evidence suggests an elevated risk of low bone mineral density (BMD) in patients with psoriasis. However, BMD assessment is rarely performed in routine clinical practice due to limited access to dual-energy X-ray absorptiometry (DXA). To evaluate BMD in patients with psoriasis using an artificial intelligence (AI)-based tool applied to routine chest radiographs, and to investigate associations with disease duration, disease severity (measured by Psoriasis Area and Severity Index [PASI]), and disease subtype. We retrospectively analysed 132 Japanese patients with psoriasis who underwent chest radiographs prior to initiating biologic therapy between 2010 and 2023. BMD was estimated using an original AI model for predicting BMD on chest radiographs. Patients were categorized as having normal BMD, osteopenia, or osteoporosis based on T-scores. Clinical associations were assessed using statistical methods, including ROC curve analysis to determine cut-off values for disease duration and PASI scores. A higher prevalence of AI-estimated osteoporosis was observed among middle-aged male patients compared to the general population. Longer disease duration (≥seven years, as determined by ROC analysis) was significantly associated with lower AI-estimated BMD (p=0.0032). Although patients with PASI scores ≥10.4 and those with the pustular subtype exhibited trends toward lower BMD, these differences did not reach statistical significance. Our findings indicate a substantial reduction in BMD among male psoriasis patients in their 40s and 50s.

PMID:
42702021
Bibliographic data and abstract were imported from PubMed on 07 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.