Automatic knee osteoarthritis severity grading based on X-ray images using a hierarchical classification method
Authors' conclusion
The hierarchical classification method developed in the current study is a feasible approach to assess the severity of KOA.
Study context
- Population
- Patients with knee X-ray images assessed for knee osteoarthritis (KOA) severity (Kellgren-Lawrence (KL) grade 0–4)
- Participants
- 4,074 patients
- Outcome
- Accuracy of the hierarchical classification method for automatic KL grading of KOA severity
Methodological detailshierarchical classification method, four U-Net segmentation models, geometric, radiomic and combined models, strict decision strategy in testing cohort
- Hierarchical classification method
- Four U-Net segmentation models
- Geometric, radiomic and combined models
- Strict decision strategy in testing cohort
Performance
As reported in the abstract
Accuracy of hierarchical classification method, all X-ray images of testing cohort
Accuracy of the hierarchical classification method
65.98%
Further findings
Combined model accuracy by sub-task classification
- KL grade 0-2 vs KL grade 3-4
- 98.50%
- KL grade 3 vs KL grade 4
- 81.65%
- KL grade 0 vs KL grade 1-2
- 82.07%
- KL grade 1 vs KL grade 2
- 74.10%
- Combined models achieved the best performance in KL grading
U-Net segmentation, dice similarity coefficient
- Total joint space (TJS)
- 0.88
- Lateral joint space (LJS)
- 0.86
- Medial joint space (MJS)
- 0.88
- Osteophytes
- 0.64
Original publication
Automatic knee osteoarthritis severity grading based on X-ray images using a hierarchical classification method. Arthritis Res Ther. Published 18 November 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →