Related Experiment Video
Updated: Apr 10, 2026
Automated Joint Space Detection Improves Bone Segmentation Accuracy
Published on: November 28, 2025
Multicenter deep learning for multi-abnormality screening on hip radiographs: development, external validation, and
Shenghao Xu1, Chaohui Guo2, Qibo Xu3
1Department of Orthopedics, The Second Hospital of Jilin University, Changchun, Jilin, China; Joint International Research Laboratory of Ageing Active Strategy and Bionic Health in Northeast Asia of the Ministry of Education, Jilin University, Changchun, Jilin, China.
Introduction:
Hip abnormalities are a major cause of pain and functional impairment, yet missed diagnoses remain common due to high clinical workloads and interobserver variability. Existing deep learning (DL) models predominantly target single pathologies, limiting their utility for comprehensive screening.
Objectives:
To develop and validate a DL model for simultaneous screening of multiple hip conditions using a large, multicenter dataset of pelvic radiographs.
Methods:
A ResNet-50-based model was trained and internally validated using data from two hospitals (25,908 hips) with enhancements (convolutional block attention module, generalized mean pooling, and class-balanced focal loss) for classifying eight categories: normal, hip osteoarthritis, osteonecrosis of the femoral head, femoral neck fracture (FNF), intertrochanteric fracture (ITF), developmental dysplasia of the hip, total hip arthroplasty, and metallic internal fixation. External validation was performed on 4,600 hips from a third hospital. Model performance was compared with that of six orthopedic surgeons of varying seniority, and a two-phase reader study assessed diagnostic performance with and without model assistance; non-inferiority was prespecified at a 5% margin for accuracy and macro-F1.
Results:
On the internal test set, the model achieved 93.93% accuracy (95% confidence interval [CI]: 93.26-94.56), macro-AUC 0.99, and macro-F1 90.66% (95% CI: 88.83-92.14). On the external test set, accuracy was 90.11% (95% CI: 89.28-90.98), with macro-AUC 0.99 and macro-F1 87.29% (95% CI: 86.02-88.56). Sensitivity was high for acute injuries (FNF: 95.61% [95% CI: 92.82-98.17]; ITF: 95.31% [95% CI: 90.36-98.94]). The model outperformed residents and attendings (P < 0.001) and was non-inferior to deputy chiefs. Assistance improved surgeon accuracy (gains: 3.93%-11.33%) and inter-rater agreement (κ: from 0.52-0.88 to 0.69-0.88).
Conclusions:
We developed and externally validated a DL model for automated screening of multiple common hip abnormalities. The system is well-suited for triage workflows by flagging high-risk cases for expedited review. Additionally, it may function as a supportive second reader in high-volume or resource-limited settings.