Deep Learning for Bone Mineral Density and T-Score Prediction from Chest X-rays: A Multicenter Study
Problem
Osteoporosis diagnosis and treatment are inadequate worldwide; the study predicts bone mineral density and T-score from routine chest X-rays.
Why it matters
Osteoporosis is increasingly common, and chest X-rays are cheap and widely available unlike DXA scans.
What the AI did
A deep learning model ensembling chest X-rays, age and sex was trained across six hospitals to predict BMD by regression and T-score by multiclass classification.
Results
Correlation coefficients for BMD prediction: hip = 0.75, lumbar spine = 0.63 (abstract truncated before further metrics).
Impact
Multicenter validation suggests chest X-rays could support opportunistic bone-density assessment.
Record
| Venue | Biomedicines |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Yoichi Sato, Norio Yamamoto, Naoya Inagaki, Yusuke Iesaki, Takamune Asamoto, Tomohiro Suzuki, Shunsuke Takahara |
| Citations | 84 (OpenAlex, at import time) |
| References | 43 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/biomedicines10092323 |
| Open access | gold |
| Keywords | deep learning, chest X-ray, bone mineral density, osteoporosis, T-score, multicenter |
Importance rating
significant — thousands to millions affected
Abstract states osteoporosis is increasing worldwide with inadequate diagnosis/treatment; cheap chest X-ray screening could extend bone-density assessment broadly.