Comparing image analysis approaches versus expert readers: the relation of knee radiograph features to knee pain

Predicting joint space changes in knee osteoarthritis over 6 years: a combined model of TransUNet and XGBoost

BACKGROUND

The progression of knee osteoarthritis is mainly characterized by the reduction in joint space width (JSW). The goal of this study was to build a knee joint space segmentation model through deep learning (DL) methods and develop a model for automatically measuring JSW. Furthermore, we predicted JSW changes in the sixth year based on regression models.

METHODS

The data for this study was sourced from the Osteoarthritis Initiative database. We filtered knee X-ray images from 1,947 participants and tested six neural networks for segmentation to build an automatic JSW measurement model. Subsequently, we combined the clinical data with the JSW measurement results to predict the sixth-year knee JSW using six different regression models.

RESULTS

The segmentation results showed that TransUNet performed the best, with an overall Dice coefficient of 0.889. The intraclass correlation coefficient (ICC) between manually measured and TransUNet's automatically measured JSW reached 0.927 (P<0.01). Among the regression models, eXtreme Gradient Boosting (XGBoost) demonstrated the best predictive performance, with a mean absolute error (MAE) of 0.48 and an ICC of 0.887 (P<0.01). To better align with clinical practice, we reduced the prediction model to utilize only 2 years of JSW images. The results showed that using the 0- and 12-month X-ray images still achieved high accuracy, with an MAE of 0.585 (P<0.05) and an ICC of 0.805 (P<0.01).

CONCLUSIONS

We developed a novel JSW measurement model that significantly improves accuracy compared to previous methods and identified the best prediction model by combining TransUNet and XGBoost. Additionally, in our built model, predicting the 72-month JSW using only 2 years of knee X-ray images and several clinical features achieved high accuracy.

Uncovering associations between data-driven learned qMRI biomarkers and chronic pain

Knee pain is the most common and debilitating symptom of knee osteoarthritis (OA). While there is a perceived association between OA imaging biomarkers and pain, there are weak or conflicting findings for this relationship. This study uses Deep Learning (DL) models to elucidate associations between bone shape, cartilage thickness and T₂ relaxation times extracted from Magnetic Resonance Images (MRI) and chronic knee pain. Class Activation Maps (Grad-CAM) applied on the trained chronic pain DL models are used to evaluate the locations of features associated with presence and absence of pain. For the cartilage thickness biomarker, the presence of features sensitive for pain presence were generally located in the medial side, while the features specific for pain absence were generally located in the anterior lateral side. This suggests that the association of cartilage thickness and pain varies, requiring a more personalized averaging strategy. We propose a novel DL-guided definition for cartilage thickness spatial averaging based on Grad-CAM weights. We showed a significant improvement modeling chronic knee pain with the inclusion of the novel biomarker definition: likelihood ratio test p-values of 7.01 × 10^–33 and 1.93 × 10^–14 for DL-guided cartilage thickness averaging for the femur and tibia, respectively, compared to the cartilage thickness compartment averaging.

Assessment of knee pain from MR imaging using a convolutional Siamese network

OBJECTIVES

It remains difficult to characterize the source of pain in knee joints either using radiographs or magnetic resonance imaging (MRI). We sought to determine if advanced machine learning methods such as deep neural networks could distinguish knees with pain from those without it and identify the structural features that are associated with knee pain.

METHODS

We constructed a convolutional Siamese network to associate MRI scans obtained on subjects from the Osteoarthritis Initiative (OAI) with frequent unilateral knee pain comparing the knee with frequent pain to the contralateral knee without pain. The Siamese network architecture enabled pairwise learning of information from two-dimensional (2D) sagittal intermediate-weighted turbo spin echo slices obtained from similar locations on both knees. Class activation mapping (CAM) was utilized to create saliency maps, which highlighted the regions most associated with knee pain. The MRI scans and the CAMs of each subject were reviewed by an expert radiologist to identify the presence of abnormalities within the model-predicted regions of high association.

RESULTS

Using 10-fold cross-validation, our model achieved an area under curve (AUC) value of 0.808. When individuals whose knee WOMAC pain scores were not discordant were excluded, model performance increased to 0.853. The radiologist review revealed that about 86% of the cases that were predicted correctly had effusion-synovitis within the regions that were most associated with pain.

CONCLUSIONS

This study demonstrates a proof of principle that deep learning can be applied to assess knee pain from MRI scans.

KEY POINTS

• Our article is the first to leverage a deep learning framework to associate MR images of the knee with knee pain. • We developed a convolutional Siamese network that had the ability to fuse information from multiple two-dimensional (2D) MRI slices from the knee with pain and the contralateral knee of the same individual without pain to predict unilateral knee pain. • Our model achieved an area under curve (AUC) value of 0.808. When individuals who had WOMAC pain scores that were not discordant for knees (pain discordance < 3) were excluded, model performance increased to 0.853.