Radiomics Grading of Hand Osteoarthritis Severity Using Standard Radiographs: Results from the DIGICOD Cohort.
Authors
Affiliations (13)
Affiliations (13)
- Université Paris Cité, PARCC UMRS 970, INSERM, Paris, France. Electronic address: [email protected].
- Université Paris Cité, PARCC UMRS 970, INSERM, Paris, France. Electronic address: [email protected].
- Université Paris Cité, PARCC UMRS 970, INSERM, Paris, France; Department of Neuroradiology, Hôpital Fondation Ophtalmologique Adolphe de Rothschild, Paris, France. Electronic address: [email protected].
- Université Paris Cité, AP-HP, Department of Radiology, Hôpital Saint Louis PARCC UMRS 970, INSERM, Paris, France. Electronic address: [email protected].
- Research Unit of Health Sciences and Technology, Faculty of Medicine, University of Oulu, Oulu, Finland; Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland. Electronic address: [email protected].
- Sorbonne Université, APHP, Department of Rheumatology, Hôpital Saint-Antoine, Paris, France. Electronic address: [email protected].
- Institute of Sports Imaging, French National Institute of Sports (INSEP), Paris, France; Sorbonne Université, APHP, Department of Radiology, Hôpital Saint-Antoine, Paris, France. Electronic address: [email protected].
- Sorbonne Université, APHP, Department of Radiology, Hôpital Saint-Antoine, Paris, France. Electronic address: [email protected].
- Sorbonne Université, APHP, Department of Rheumatology, Hôpital Saint-Antoine, Paris, France. Electronic address: [email protected].
- Université Paris Cité, AP-HP, Unité de Recherche Clinique / Centre d'Investigation Clinique 1418 - Épidémiologie Clinique, Hôpital Européen Georges-Pompidou, INSERM, Paris, France. Electronic address: [email protected].
- Université Paris Cité, AP-HP, Department of Radiology, Hôpital Lariboisière, Paris, France. Electronic address: [email protected].
- Sorbonne Université, APHP, Department of Rheumatology, Hôpital Saint-Antoine, Paris, France. Electronic address: [email protected].
- Université Paris Cité, AP-HP, Department of Radiology, Hôpital Européen Georges Pompidou, PARCC UMRS 970, INSERM, Paris, France. Electronic address: [email protected].
Abstract
This study evaluated whether radiomics analysis of standard hand radiographs could provide an automated and objective assessment of structural severity in hand osteoarthritis (HOA) compared to Kellgren-Lawrence (KL) scores, ranging from 0 (normal) to 4 (severe). We conducted a retrospective study using baseline data from the DIGICOD (DIGital Cohort Osteoarthritis Design) cohort, including patients with HOA. Standard posteroanterior radiographs were segmented semi-automatically using a U-Net model with manual correction. Radiomics features describing intensity, shape, and texture were extracted for each joint and reduced after filtering and correlation suppression. The cohort was split into training (80%) and test sets (20%) using stratified sampling. Random Forest classifiers were trained to detect structural involvement (KL ≥2), severe disease (KL 3-4), and predict multiclass KL grades (0-1, 2, 3, 4) at the joint level. 379 radiographs were analyzed. Detection of structural involvement (KL ≥2) achieved an AUC of 0.81 [95% CI: 0.79-0.83], with high sensitivity (89%) but low specificity (59%). Severe disease detection (KL 3-4) reached an AUC of 0.83 [95% CI: 0.81-0.85], with very good sensitivity (80%) and good specificity (70%). Multiclass KL prediction showed lower performance (macro-averaged AUC 0.76; accuracy 56%), reflecting challenges in distinguishing intermediate (KL 2-3) and severe (KL 4) grades. This is the first study applying radiomics to standard hand radiographs for automated and objective scoring of radiographic severity in HOA. The model showed good performance detecting structural damage, supporting radiomics as a potential tool to reduce reliance on subjective visual grading.