JOA - 2026-05-20 - Journal Article
A Novel 2.5-Dimensional Deep Learning Model for "Bone-on-Bone" Detection on Magnetic Resonance Imaging in Medial Unicompartmental Knee Arthroplasty Candidates.
Liu C, Ping H, Zhang Q, Wang W, Guo W, Huang C
Topics
Key Takeaway
A 2.5D deep learning model fused with radiomic and clinical features achieved micro-AUC 0.918 for detecting bone-on-bone medial compartment OA on MRI, matching senior surgeon accuracy and outperforming junior surgeons (P=0.027).
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study asked whether a 2.5D deep learning model integrating multi-sequence MRI, adjacent slices, and multiplanar views could automate detection of medial compartment bone-on-bone OA in mUKA candidates. The model was trained on 70% and tested on 30% of 191 patients, achieving micro-AUC 0.893 standalone and 0.918 when fused with radiomic and clinical features. The fused model matched senior surgeon classification accuracy and significantly outperformed junior surgeons (P=0.010), while reducing interobserver variability.
Key Limitation
Single-institution retrospective design with no external validation means model performance on MRI scanners, field strengths, or patient populations outside the training environment is unknown.
Original Abstract
BACKGROUND
Accurate identification of "bone-on-bone" (BoB) osteoarthritis is critical for patient selection for medial unicompartmental knee arthroplasty (mUKA); however, the assessment remains subjective. We aimed to develop and validate a deep learning model using multi-sequence knee magnetic resonance imaging (MRI) for automated detection of isolated medial compartment BoB osteoarthritis in UKA candidates, thereby reducing avoidable indication-related failures.
METHODS
We retrospectively collected preoperative knee MRI data from 191 patients (64 patients who had BoB osteoarthritis, 62 patients who had partial cartilage loss, and 65 patients who had normal cartilage) who underwent mUKA or knee arthroscopy between January 2022 and January 2025. We developed an automated pipeline comprising a cartilage-segmentation model followed by a multi-sequence MRI-based 2.5-dimensional (2.5D) deep-learning model integrating adjacent slices and multiplanar views. The dataset was divided into training (70%) and test (30%) sets. Model performance was evaluated on the test set and compared with models using different feature-construction strategies. Orthopaedic surgeons who had varying experience independently evaluated classification performance.
RESULTS
The 2.5D deep-learning model demonstrated superior diagnostic performance on the test set, with a micro area under the receiver operating characteristic curve (micro-AUC) of 0.893, compared with models based solely on clinical data (0.810) or radiomic features (0.851). Combining the deep learning model output with radiomic and clinical features further improved performance (micro-AUC = 0.918). The fused model's classification accuracy was comparable to that of senior orthopaedic surgeons and significantly higher than that of junior surgeons (P = 0.027 and 0.010, respectively). It also reduced interobserver variability and produced results substantially faster than manual interpretation.
CONCLUSION
A multi-sequence 2.5D deep learning model reliably detects medial compartment BoB osteoarthritis on MRI. Its objective, reproducible output may enhance diagnostic accuracy and consistency, providing a standardized reference for cartilage assessment and supporting preoperative clinical decision-making for UKA.