A Progressive Risk Formulation for Enhanced Deep Learning based Total Knee Replacement Prediction in Knee Osteoarthritis

2024-06-14

AI Summary

This study addresses temporal risk prediction for total knee replacement (TKR) in patients with knee osteoarthritis (KOA). We propose a deep learning framework explicitly embedding clinical disease progression priors. Methodologically, we design a dual-model progressive risk modeling architecture that accepts single-image inputs and enforces temporal monotonicity—i.e., non-decreasing TKR risk over time—in multi-timepoint imaging scenarios. Leveraging multicenter X-ray and MRI data from the OAI and MOST cohorts, we introduce a monotonicity-regularized loss function. Our key contribution is the first explicit integration of clinically observed disease progression patterns into deep learning training, departing from conventional per-scan independent modeling. Experiments demonstrate significant improvements: on the OAI test set, our model achieves an AUROC of 0.87 (+0.08) and AUPRC of 0.47 (+0.13) for 1-year TKR prediction; on MOST, it attains an AUROC of 0.77 (+0.06), outperforming all baselines.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Deep Neural Architectures and Foundation Models
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We developed deep learning models for predicting Total Knee Replacement (TKR) need within various time horizons in knee osteoarthritis patients, with a novel capability: the models can perform TKR prediction using a single scan, and furthermore when a previous scan is available, they leverage a progressive risk formulation to improve their predictions. Unlike conventional approaches that treat each scan of a patient independently, our method incorporates a constraint based on disease's progressive nature, ensuring that predicted TKR risk either increases or remains stable over time when multiple scans of a knee are available. This was achieved by enforcing a progressive risk formulation constraint during training with patients who have more than one available scan in the studies. Knee radiographs and MRIs from the Osteoarthritis Initiative (OAI) and Multicenter Osteoarthritis Study (MOST) were used in this work and deep learning models were trained to predict TKR within 1, 2, and 4-year time periods. The proposed approach, utilizing a dual-model risk constraint architecture, demonstrated superior performance compared to baseline - conventional models trained with standard binary cross entropy loss. It achieved an AUROC of 0.87 and AUPRC of 0.47 for 1-year TKR prediction on the OAI radiograph test set, considerably improving over the baseline AUROC of 0.79 and AUPRC of 0.34. For the MOST radiograph test set, the proposed approach achieved an AUROC of 0.77 and AUPRC of 0.25 for 1-year predictions, outperforming the baseline AUROC of 0.71 and AUPRC of 0.19. Similar trends were observed in the MRI testsets

Problem

• Predicting Total Knee Replacement need in osteoarthritis patients using deep learning.
• Leveraging progressive risk formulation with single or multiple scans for improved predictions.
• Ensuring predicted risk increases or remains stable over time with multiple scans.

Innovation

• Deep learning models predict TKR with single scan
• Progressive risk formulation improves multi-scan predictions
• Dual-model risk constraint enhances prediction accuracy