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

Diagnostics ◽  
2021 ◽  
Vol 11 (7) ◽  
pp. 1156
Author(s):  
Kang Hee Lee ◽  
Sang Tae Choi ◽  
Guen Young Lee ◽  
You Jung Ha ◽  
Sang-Il Choi

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.

2013 ◽  
Vol 72 (Suppl 3) ◽  
pp. A657.1-A657 ◽  
Author(s):  
L. Van Praet ◽  
L. Jans ◽  
F. Van den Bosch ◽  
P. Jacques ◽  
P. Carron ◽  
...  

Radiology ◽  
2019 ◽  
Vol 290 (1) ◽  
pp. 157-164 ◽  
Author(s):  
Haijun Wu ◽  
Guangfeng Zhang ◽  
Lei Shi ◽  
Xiuhui Li ◽  
Min Chen ◽  
...  

2018 ◽  
Vol 211 (6) ◽  
pp. 1306-1312 ◽  
Author(s):  
Christoph A. Agten ◽  
Veronika Zubler ◽  
Marco Zanetti ◽  
Christoph A. Binkert ◽  
Orpheus Kolokythas ◽  
...  

Author(s):  
Christoph Germann ◽  
Daniela Kroismayr ◽  
Florian Brunner ◽  
Christian W. A. Pfirrmann ◽  
Reto Sutter ◽  
...  

Abstract Objective To investigate long-term effects of pregnancy/childbirth on bone marrow edema (BME) and subchondral sclerosis of sacroiliac joints (SIJ) in comparison to MRI changes caused by spondyloarthritis (SpA) and assess the influence of birth method and number of children on SIJ-MRI changes. Materials and methods This is a retrospective cohort study with 349 women (mean age 47 ± 14 years) suffering low back pain. Four subgroups were formed based on SpA diagnosis and childbirth (CB) history. Two musculoskeletal radiologists scored the presence of BME and sclerosis on SIJ-MRI using the Berlin method. Further, an 11-point “global assessment score” representing the overall confidence of SpA diagnosis based on MRI was evaluated in addition to the ASAS (Assessment of Spondyloarthritis International Society) criterion of “positive MRI” for sacroiliitis. Results CB did not correlate with BME score (p = 0.38), whereas SpA diagnosis was associated with a higher BME score (r = 0.31, p < 0.001). Both CB (r = 0.21, p < 0.001) and SpA diagnosis (r = 0.33, p < 0.001) were correlated with a higher sclerosis score. CB was not associated with a higher confidence level in diagnosing SpA based on MRI (p = 0.07), whereas SpA diagnosis was associated with a higher score (r = 0.61, p < 0.001). Both CB (phi = 0.13, p = 0.02) and SpA diagnosis (phi = 0.23, p < 0.001) were significantly associated with a positive ASAS criterion for sacroiliitis. In non-SpA patients with CB, number of children (p = 0.001) was an independent predictor of sclerosis score, while birth method yielded no significant effect (p = 0.75). Conclusion Pregnancy/CB has no impact on long-term BME on SIJ, however, may cause long-term subchondral sclerosis—similar to SpA-associated sclerosis. Number of children is positively correlated with SIJ sclerosis. Birth method yields no effect on SIJ sclerosis.


2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 1839.1-1839
Author(s):  
F. Ladeb ◽  
D. Ben Nessib ◽  
M. Bouaziz ◽  
W. Hamdi ◽  
E. Labbene ◽  
...  

Background:In view of the limited accuracy of clinical evaluation to recognize sacroiliitis, several imaging techniques such as conventional radiographs, scintigraphy, ultrasound (US), computed tomography (CT) and magnetic resonance imaging (MRI) have been used to examine the sacroiliac joints (SIJ).Objectives:The aim of this study was to assess the performance of MRI for detecting sacroiliitis in early stages of spondyloarthritis (SpA).Methods:This cross-sectional prospective monocentric double-blind study included 57 patients consulting for symptoms suggestive of SpA during more than 3 months between February 2014 and February 2017. Patients with conventional radiograph showing a confirmed sacroiliitis (grade 3 or 4) were not included. After clinical examination and blood sampling, eligible patients underwent MRI of SIJ. MR images were interpreted by 2 experimented musculoskeletal radiologists blinded to clinical and laboratory data. Two professors in rheumatology blinded to radiologists’ conclusions, analyzed clinical data, laboratory tests, HLA typing, X-rays and MRI images and divided the patients into 2 groups: confirmed non radiographic SpA (nr-SpA) or no SpA. This classification was considered as the gold standard when analyzing the results.Results:Fifteen men and 42 women were enrolled. The mean age at inclusion was 39.75 ± 11 years [17-59]. The mean duration from the first symptom was 47 ± 39 months [6.6-180]. Forty-three patients were assessed as nr-SpA (75.4%) and 14 patients as no SpA (24.6%). Thirty-three percent of patients were HLA B27 positive. Totally 22 patients had sacroiliitis at MRI, all of them classified as confirmed nr-SpA. Among the nr-SpA group, MRI showed bone marrow edema (BME) in 34.9% of patients and erosions in 44.2% of patients. Among the patients in whom the diagnosis of SpA was excluded, MRI showed bone marrow edema (BME) in 7% of patients and erosions in 7% of patients. A statistically significant association was observed between the presence of sacroiliitis at MRI and rheumatologists’ diagnosis of SpA (p=0.001). The diagnostic value of MRI lesions is presented in the following table:Sensitivity (%)Specificity (%)Positive Predictive Value (%)Positive Predictive Value (%)BME34.992.993.731.7Erosions44.292.99535.1MRI conclusion: sacroiliitis51.210010040Conclusion:SIJ MRI had an excellent specificity for the diagnosis of SpA but a moderate sensitivity. Consequently, some patients in early stages of SpA might be missed by MRI. In addition, we found that diagnostic based solely on BME lacked sensitivity. Detection of erosions in addition to BME enhanced sensitivity (from 34.9% to 44.9%) without changing specificity. Indeed, many recent studies have pointed out the importance of considering structural lesions of SIJ in addition to inflammatory lesions [1, 2].References:[1]Weber U, Lambert RGW, Pedersen SJ, et al (2010) Assessment of structural lesions in sacroiliac joints enhances diagnostic utility of magnetic resonance imaging in early spondylarthritis. Arthritis Care Res 62:1763–1771.https://doi.org/10.1002/acr.20312[2]Weber U, Jurik AG, Lambert RGW, Maksymowych WP (2016) Imaging in Spondyloarthritis: Controversies in Recognition of Early Disease. Curr Rheumatol Rep 18:58.https://doi.org/10.1007/s11926-016-0607-7Disclosure of Interests:None declared


2019 ◽  
Vol 61 (4) ◽  
pp. 471-479
Author(s):  
Simon Krabbe ◽  
Mikkel Østergaard ◽  
Inge J Sørensen ◽  
Jakob Møller ◽  
Bente Jensen ◽  
...  

Background Sagittal magnetic resonance (MR) images are typically obtained with the same spatial resolution along the entire spine, but cervical vertebrae are smaller and may be harder to assess. Purpose To investigate if high-resolution (high-res) short tau inversion recovery (STIR) and T1-weighted turbo spin echo (T1W) MR imaging (MRI) sequences are superior to standard resolution for detecting inflammatory and structural lesions in the cervical spine of patients with axial spondyloarthritis. Material and Methods Images were obtained in 36 patients. Voxel sizes at high/standard resolution were 1.99/4.33 mm3 (STIR) and 0.89/3.71 mm3 (T1W). High-resolution and standard-resolution images were scored by two readers according to the Canada-Denmark (CANDEN) MRI spine scoring system. Results Higher bone marrow edema scores were obtained at high resolution versus standard resolution (mean 2.1 vs. 1.2, P = 0.040), whereas fat lesion scores (1.8 vs. 1.5, P = 0.27) and new bone formation scores (3.5 vs. 2.8, P = 0.21) were similar. High-resolution MRI did not classify more patients as positive for bone marrow edema, fat, or new bone formation in the cervical spine compared to standard resolution. Using lateral radiographs as reference standard, sensitivity for detecting anterior corner syndesmophytes with both high-resolution and standard-resolution MRI was low (range 7–22%) and sensitivity for detecting ankylosis was low to moderate (20–55%), while specificity was high (≥96%). Conclusion High-resolution MRI allowed identification of more inflammatory lesions in the cervical spine in patients with axial spondyloarthritis when compared to standard resolution, but it did not classify more patients as positive for bone marrow edema. The slightly increased sensitivity at high-resolution MRI seemed to be too modest to have any real clinical importance.


2019 ◽  
Vol 71 (11) ◽  
pp. 1942-1942
Author(s):  
Frank Verhoeven ◽  
Maxime Sondag ◽  
Clément Prati ◽  
Daniel Wendling

Sign in / Sign up

Export Citation Format

Share Document