aceAccuracy of general-purpose large language models (LLMS) in interpreting adolescent idiopathic scoliosis radiographs: a comparative study with experienced spine surgeons
Purpose: This study evaluates the accuracy of general-purpose, vision-capable large language models (LLMs) in interpreting adolescent idiopathic scoliosis (AIS) radiographs and compares their outputs with an expert surgeon consensus reference standard. Methods: This retrospective comparative diagnostic accuracy study analyzed 128 anonymized AIS radiographic case sets from 128 different patients, including standing coronal and sagittal radiographs and lateral bending films. ChatGPT GPT-5.2 Instant, Gemini 3 Flash…
General-purpose vision LLMs estimated scoliosis Cobb angles within 5 degrees of spine surgeon consensus in most cases, with ChatGPT agreeing in 94.5% of the 128 cases.
The same models showed substantially lower exact agreement on Lenke classification, modifiers, and fusion-related landmarks, so they are not reliable for comprehensive radiograph interpretation or surgical planning.
Only specific versions of three general-purpose models were tested in January 2026, and no formal inferential ranking between the models was performed.
Evidence
- Peer-reviewedEuropean Spine Journal2026-10-16
How should this claim be treated?
Truvace Impact Record TRV-2026-1679, v1: “Accuracy of general-purpose large language models (LLMS) in interpreting adolescent idiopathic scoliosis radiographs: a comparative study with experienced spine surgeons.” Truvace, 2026-10-17. /record/TRV-2026-1679 (accessed at citation time). sha256 8eaab90a26692bb5…
Calibration history
Every change to this record since certification, in the open. None yet — the reading has held since it entered the record.
Certified into the record
How to verify without trusting this page
Fetch the canonical text of any version from /api/record/TRV-2026-1679 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.