Method for Diagnosing the Bone Marrow Edema of Sacroiliac Joint in Patients with Axial Spondyloarthritis Using Magnetic Resonance Image Analysis Based on Deep Learning

Abstract

Axial spondyloarthritis (axSpA) is a chronic inflammatory disease of the sacroiliac joints. In this study, we develop a method for detecting bone marrow edema by magnetic resonance (MR) imaging of the sacroiliac joints and a deep-learning network. A total of 815 MR images of the sacroiliac joints were obtained from 60 patients diagnosed with axSpA and 19 healthy subjects. Gadolinium-enhanced fat-suppressed T1-weighted oblique coronal images were used for deep learning. Active sacroiliitis was defined as bone marrow edema, and the following processes were performed: setting the region of interest (ROI) and normalizing it to a size suitable for input to a deep-learning network, determining bone marrow edema using a convolutional-neural-network-based deep-learning network for individual MR images, and determining sacroiliac arthritis in subject examinations based on the classification results of individual MR images. About 70% of the patients and normal subjects were randomly selected for the training dataset, and the remaining 30% formed the test dataset. This process was repeated five times to calculate the average classification rate of the five-fold sets. The gradient-weighted class activation mapping method was used to validate the classification results. In the performance analysis of the ResNet18-based classification network for individual MR images, use of the ROI showed excellent detection performance of bone marrow edema with 93.55 ± 2.19% accuracy, 92.87 ± 1.27% recall, and 94.69 ± 3.03% precision. The overall performance was additionally improved using a median filter to reflect the context information. Finally, active sacroiliitis was diagnosed in individual subjects with 96.06 ± 2.83% accuracy, 100% recall, and 94.84 ± 3.73% precision. This is a pilot study to diagnose bone marrow edema by deep learning based on MR images, and the results suggest that MR analysis using deep learning can be a useful complementary means for clinicians to diagnose bone marrow edema.

Keywords: axial spondyloarthritis; bone marrow edema; deep learning; magnetic resonance imaging; sacroiliitis.

Figure 1

Overall process of the proposed method for diagnosing sacroiliac arthritis. (a) Process of ROI generation; (b) ResNet based network for Bone marrow edema classification.

Figure 2

Example MR images of the SIJ region. (a) MR images including all areas surrounding the sacral and iliac bones; (b) annotated MR images of the sacral (green) and iliac (red) bones as well as bone marrow edema (yellow).

Figure 3

Data normalization. (a) Bounding boxes for the ROI (blue boxes) containing the left and right iliac bones (black boxes)); (b) ROI patch combining bounding boxes obtained from left and right iliac bone regions (blue boxes in (a)) with random noise.

Figure 4

Configuration of ResNet consisting of five convolutional stages, which contain the convolution blocks comprising $7\times 7$ and $3\times 3$ layers. The output layer consists of as many nodes as the number of classes to be obtained.

Figure 5

Results of validation experiment. (a) ResNet50 model; (b) ResNet18 model.

Figure 6

Results of automatically determining the presence of bone marrow edema using ResNet. (a) Results for axSpA subjects; (b) results for normal subjects; (c) results of applying the median filter to the results of (a); (d) results of applying the median filter to the results of (b).

Figure 7

ROC curves for bone marrow edema classifications from individual MR slices.

Figure 8

(a) Results of the class activation mapping for an example image with bone marrow edema; (b) Another example of an image with markers for bone marrow edema; (c) Gradient-based class activation map of (a); (d) Gradient-based class activation map of (b). The color represents regions with the greatest activation, and the degree of activation decreases in the order of orange, yellow, green, and blue.