70/100

Moderate evidence

GainPublished recordHealthGood Space

Opportunistic Risk Assessment of Major Osteoporotic Fractures from Routine Chest Radiographs Using Deep Learning

Truvace record
TRV-2026-2981raw data (JSON)
Version
v1Published
First published
Evidence type
Controlled or validated comparison
Sources
1 sourcePeer-reviewed
sha256
c483e157568d…check it

Background Bone mineral density (BMD) and the Fracture Risk Assessment Tool (FRAX) incompletely capture fracture risk in the osteopenic range, where much of the fracture burden occurs. Routine chest radiographs are acquired across a broad range of patients and may contain information relevant to fracture risk, offering a potential opportunistic complement without additional imaging burden. Methods We developed a convolutional neural network survival model in 53,109 adults at a tertiary hospital and externally va…

The gain

In an external validation of 85,284 patients, a deep-learning model reading routine chest radiographs predicted major osteoporotic fracture with a C-index of 0.843, versus 0.779 for FRAX with bone mineral density.

What this doesn’t fix

All participants had been referred for densitometry, so absolute-risk and case-finding estimates require prospective evaluation in an unselected routine-radiography population.

Sources

1 on file
  1. NEJM AIdoi:10.1056/aioa2600590
    Peer-reviewed2026-11-16

Cite this record

v1

Truvace Impact Record TRV-2026-2981, v1: “Opportunistic Risk Assessment of Major Osteoporotic Fractures from Routine Chest Radiographs Using Deep Learning.” Truvace, 2026-11-17. /record/TRV-2026-2981 (accessed on the day you copy it). sha256 c483e157568d6b81…

Change history

1 version

Every change made to this record since it was first published, and why. Nothing has changed since it went up.

  1. Publishedv1c483e157568d…

    Certified into the record

Check this record hasn’t been altered

Check this record hasn’t been altered
How to check it yourself, without trusting this page

Download any version from /api/record/TRV-2026-2981 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.