Deep Learning Predicts Total Knee Replacement from Magnetic Resonance Images

Abstract

Knee Osteoarthritis (OA) is a common musculoskeletal disorder in the United States. When diagnosed at early stages, lifestyle interventions such as exercise and weight loss can slow OA progression, but at later stages, only an invasive option is available: total knee replacement (TKR). Though a generally successful procedure, only 2/3 of patients who undergo the procedure report their knees feeling "normal" post-operation, and complications can arise that require revision. This necessitates a model to identify a population at higher risk of TKR, particularly at less advanced stages of OA, such that appropriate treatments can be implemented that slow OA progression and delay TKR. Here, we present a deep learning pipeline that leverages MRI images and clinical and demographic information to predict TKR with AUC 0.834 ± 0.036 (p < 0.05). Most notably, the pipeline predicts TKR with AUC 0.943 ± 0.057 (p < 0.05) for patients without OA. Furthermore, we develop occlusion maps for case-control pairs in test data and compare regions used by the model in both, thereby identifying TKR imaging biomarkers. As such, this work takes strides towards a pipeline with clinical utility, and the biomarkers identified further our understanding of OA progression and eventual TKR onset.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Pipeline predicting if patient will undergo TKR within 5 years from MRI/X-ray images and non-imaging variables. MRI and X-ray images are center-cropped and cropped to a region centered around the joint, respectively, and normalized. DenseNet-121 is pretrained to predict OA and fine-tuned to predict TKR. Image-based predictions and clinical information are fed to a logistic regression (LR) ensemble based on OA severity. Each ensemble, whose hyperparameters were optimized for Youden’s index in a hyperparameter search, averages predictions of LR models in its OA severity for final TKR prediction. Pipeline is subsequently analyzed through occlusion map analysis to identify imaging biomarkers of TKR.

Figure 2

ROC curves for X-ray and MRI architectures on test data. X-ray pipeline ROC curves are shown in (a), with AUCs as follows, p < 0.05: 0.848 ± 0.039 (image only), 0.868 ± 0.028 (non-imaging info. only), 0.890 ± 0.021 (integrated model). MRI pipeline ROC curves are shown in (b), with AUCs as follows, p < 0.05: 0.886 ± 0.020 (image only), 0.868 ± 0.028 (non-imaging info. only), 0.834 ± 0.036 (integrated model). Standard deviations used to calculate confidence intervals. ROC curves with AUCs within 1 standard deviation of the mean for each model type during bootstrapping are also shown on plots.

Figure 3

ROC curves for MRI and X-ray pipelines at selected OA classifications and pipeline versions in which MRI performance was significantly better than that of X-ray. MRI pipeline outperforms X-ray pipeline at no OA for both image-only and integrated models, as seen in (a,c). As shown in (b), integrated MRI pipeline also outperformed integrated X-ray pipeline for patients with severe OA, while (d) shows image-only MRI pipeline outperformed image-only X-ray pipeline across all OA stages. AUCs are displayed in the figure with p < 0.05. Standard deviations used to calculate confidence intervals. ROC curves with AUCs within 1 standard deviation of the mean for each pipeline version during bootstrapping are also shown on plots.

Figure 4

Slices of occlusion map of true positive detected by MRI pipeline, overlaid on corresponding slices of DESS MRI. Such maps were generated and analyzed for all 124 true positives and corresponding true negative controls of the integrated MRI pipeline.