Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2013
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-20T10:43:10.996475Z
status: published
lens: g_space
sector: health
headline: Multicentric development and validation of a fully automated artificial intelligence system for planning medial open wedge high tibial osteotomy on weight-bearing anterior-posterior long leg radiographs: The ISAKOS John Joyce Award 2025
dek: Purpose: Preoperative planning, currently performed by orthopaedic surgeons (OS), is essential for successful osteotomy, but is labour-intensive and error-prone. As such, there is significant potential in the use of automated artificial intelligence (AI) algorithms to augment the abilities of OS in managing lower extremity pathologies in a high-volume task that demands high precision and reliability. The aim was to develop a deep learning (DL) system capable of fully autonomous planning a medial open wedge high…
gain_title: A fully automated deep learning system planned medial open wedge high tibial osteotomy in about 25 seconds versus about 149 seconds for expert surgeons, with higher reliability and comparable accuracy.
problem_title: (none)
trace_subject: (none)
gain_reading: A fully automated deep learning system planned medial open wedge high tibial osteotomy in about 25 seconds versus about 149 seconds for expert surgeons, with higher reliability and comparable accuracy.
gain_evidence: reliability comparable to that of expert OS while significantly reducing processing time and improving measurement consistency
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built a deep learning system with twelve expert networks that identifies landmarks on weight-bearing long leg radiographs and then plans a medial open wedge high tibial osteotomy without human input. They trained it on retrospective data from patients who had osteotomy and tested it on held-out and external datasets.

The result matters because manual planning is described as labour-intensive and error-prone, so automation could speed up and standardise a precision-dependent task. The evidence comes from a retrospective comparison and shows the tool as an assistant to surgeons, not a replacement or proof of better patient outcomes.
limitation: The study was retrospective (Level III), so it shows agreement with an expert reference standard rather than prospective clinical benefit.
tag: Evidence-backed gain
key_points: The study enrolled 594 retrospective patients, split into training, validation and hold-out test sets, plus an external set of 100. | Surgeon planning accuracy was compared against an expert-derived reference using FDA-approved planning software and three expert orthopaedic surgeons. | For gap planning, the model reached 98.1% clinically acceptable accuracy (%-based) and 96.2% (°-based). | Model reliability reached ICC 0.99 for osteotomy gap planning, versus ICC 0.9 to 0.97 for human raters. | Processing time for alignment analysis fell significantly with the model.
rundown: The model was tested on two ways of planning the osteotomy: the desired postoperative intersection of the mechanical axis with the tibial plateau, and the desired femorotibial axis. Testing used data from two institutions.

For alignment parameters, model accuracy ranged from 0.22° to 1.04°, comparable to human accuracy of 0.13° to 1.06°. Clinically acceptable accuracy was 91% to 100% for the model and 72% to 100% for humans.

Osteotomy gap errors were similar for model and humans, about 0.44 to 0.53 mm for the model and 0.43 to 0.52 mm for humans in most measures. The authors conclude that such systems may help accelerate and standardise planning, and classify the work as Level III retrospective comparative evidence.
sources:
- peer_reviewed | Knee Surgery, Sports Traumatology, Arthroscopy | https://doi.org/10.1002/ksa.70635 | 2026-10-19
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

9509024ce1949188846abdad21c1d4284df2033ddd8971ee7b6aaaeebf2684c1

previous

0000000000000000000000000000000000000000000000000000000000000000