R.Du Hyun RoOrthopedics & Research
Text
Paper Library
AIKOAKLexternal validation

2026 · Journal of Arthroplasty

Can OA grades be classified consistently on knee X-rays from other hospitals?

Binary OA detection (KL ≥2) had AUC 0.94 (95% CI 0.92–0.96), sensitivity 96.6%, specificity 90.5% and accuracy 94.2%.

Results explained with study numbers

Across 2,546 knees, binary OA detection (KL ≥2) had sensitivity 96.6%, specificity 90.5% and accuracy 94.2%. These differ from grade-specific classification: KL 2 sensitivity was 77.2%.Enlarge Figure

Across 2,546 knees, binary OA detection (KL ≥2) had sensitivity 96.6%, specificity 90.5% and accuracy 94.2%. These differ from grade-specific classification: KL 2 sensitivity was 77.2%.

Journal of Arthroplasty (2026). DOI: 10.1016/j.arth.2026.01.078 · Original explanatory chart based on reported data

Evidence & Original SourceImage use terms

The research question

Can OA grades be classified consistently on knee X-rays from other hospitals?

Participants and Methods

A model trained on OAI/MOST was applied to Rosenberg images at two Korean university hospitals. Consensus of two orthopaedic surgeons and one radiologist provided the reference. KL 0 and 1 were merged. This retrospective multicentre external validation analysed 2,546 knees in 1,273 patients.

Main Findings

Binary OA detection (KL ≥2) had AUC 0.94 (95% CI 0.92–0.96), sensitivity 96.6%, specificity 90.5% and accuracy 94.2%. For borderline KL 2, AUC was 0.87, sensitivity 77.2% and specificity 95.3%. Grades were not equally easy to distinguish.

Implications for clinical and surgical practice

This provides clinical validation evidence for CONNEVO KOA. Consistent assessment of borderline radiographic grades supports clinicians who also consider pain, function and examination findings.

Limitations

This external validation compared standalone AI with consensus grading. It was not an intervention trial measuring changes in AI-assisted clinicians' performance or patient outcomes. KL 0 and 1 were merged; data came from two Korean university hospitals using a specific imaging view. Du Hyun Ro was not a coauthor.

Let's put it briefly.

Which clinical question does this address?

This provides clinical validation evidence for CONNEVO KOA. Consistent assessment of borderline radiographic grades supports clinicians who also consider pain, function and examination findings.

Keywords

Abstract · Original text

Byung Sun Choi, Seong Hwan Hong, Hae Jung Lee, Seong Hwan Kim. Performance of an Artificial Intelligence-Based Software for Automated Kellgren-Lawrence Grading of Knee Osteoarthritis: A Multicenter Cohort Study. Journal of Arthroplasty. 2026. doi:10.1016/j.arth.2026.01.078.

Excerpt from the original abstract

The Kellgren-Lawrence (KL) grading system is the standard for assessing knee osteoarthritis (OA) severity. However, it is limited by major observer variability.…
Read abstract on PubMedView original

The original paper.

Performance of an Artificial Intelligence-Based Software for Automated Kellgren-Lawrence Grading of Knee Osteoarthritis: A Multicenter Cohort Study

Byung Sun Choi, Seong Hwan Hong, Hae Jung Lee, Seong Hwan Kim
Journal of Arthroplasty · 2026

DOI · 10.1016/j.arth.2026.01.078PubMedBased data
Academic Discussion