Bone marrow edema detection for diagnostic support of axial spondyloarthritis using MRI.

PURPOSE: This study proposes a process for detecting slices with bone marrow edema (BME), a typical finding of axSpA, using MRI scans as the input. This process does not require manual input of ROIs and provides the results of the judgment of the presence or absence of BME on a slice and the location of edema as the rationale for the judgment. METHODS: First, the signal intensity of the MRI scans of the sacroiliac joint was normalized to reduce the variation in signal values between scans. Next, slices containing synovial joints were extracted using a slice selection network. Finally, the BME slice detection network determines the presence or absence of the BME in each slice and outputs the location of the BME. RESULTS: The proposed method was applied to 86 MRI scans collected from 15 hospitals in Japan. The results showed that the average absolute error of the slice selection process was 1.49 slices for the misalignment between the upper and lower slices of the synovial joint range. The accuracy, sensitivity, and specificity of the BME slice detection network were 0.905, 0.532, and 0.974, respectively. CONCLUSION: This paper proposes a process to detect the slice with BME and its location as the rationale of the judgment from an MRI scan and shows its effectiveness using 86 MRI scans. In the future, we plan to develop a process for detecting other findings such as bone erosion from MR scans, followed by the development of a diagnostic support system.

Bone marrow lesions in knee osteoarthritis assessed by dynamic contrast-enhanced MRI with histopathological correlations.

BACKGROUND: Bone marrow lesions (BMLs) in knee osteoarthritis (OA) have been assessed histopathologically and by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI); however, a direct comparison of the results has not been reported. PURPOSE: To evaluate and compare the findings by DCE-MRI and histopathology of subchondral BMLs in knee OA. MATERIAL AND METHODS: In total, 19 patients with medial tibiofemoral knee OA undergoing total knee arthroplasty were analyzed. Preoperative MRI, including a DCE sequence, was performed, and bone biopsies were obtained from the resected specimens corresponding to BML areas. The contrast enhancement by DCE-MRI was analyzed using semi-quantitative (area under the curve [AUC]), peak enhancement [PE]), and quantitative (Ktrans, Kep) methods. Enhancement in the medial OA compartment was compared with similar areas in a normal lateral compartment, and the DCE characteristics of BMLs were correlated with semi-quantitatively graded histopathological features. RESULTS: AUC and PE were significantly higher in medial tibial and femoral BMLs compared with the values in the lateral condyles; Ktrans and Kep were only significantly higher in the tibial plateau. In the tibia, AUC and PE were significantly correlated with the grade of vascular proliferation, and PE also with the degree of marrow fibrosis. There was no significant correlation between AUC/PE and histopathological findings in the femur and no correlation between quantitative DCE parameters and histopathological findings. CONCLUSION: BML characteristics by semi-quantitative DCE in the form of AUC and PE may be used as parameters for the degree of histopathological vascularization in the bone marrow whereas quantitative DCE data were less conclusive.

Bone marrow lesion and 5-year incident joint surgery in patients with knee osteoarthritis: a retrospective cohort study.

BACKGROUND: It is beneficial for society to discover the risk factors associated with surgery and to carry out some early interventions for patients with these risk factors. Few studies specifically explored the relationship between bone marrow lesions (BMLs) and long-term incident joint surgery. OBJECTIVE: To investigate the association between BML severity observed in knee osteoarthritis (OA) patients' first MRI examination and incident knee surgery within 5 years. Additionally, to assess the predictive value of BMLs for the incident knee surgery. DESIGN: Retrospective cohort study. METHODS: We identified patients diagnosed with knee OA and treated at our institution between January 2015 and January 2018, and retrieved their baseline clinical data and first MRI examination films from the information system. Next, we proceeded to determine the Max BML grades, BML burden grades and Presence BML grades for the medial, lateral, patellofemoral, and total compartments, respectively. Multi-variable logistic regression models examined the association of the BML grades with 5-year incident knee surgery. Positive and negative predictive values (PPVs and NPVs) were determined for BML grades referring to 5-year incident knee surgery. RESULTS: Totally, 1011 participants (knees) were found eligible to form the study population. Within the 5 years, surgery was performed on 74 knees. Max BML grade 2 and grade 3 of medial, patellofemoral and total compartments were strongly and significantly associated with incident surgery. None of the BML grades from lateral compartment was associated with incident surgery. The PPV was low and NPV was high for BMLs. CONCLUSIONS: BMLs found in the first MRI examination were associated with 5-year incident joint surgery, except for those allocated in lateral compartments. The high NPVs imply that patients without BMLs have a low risk of requiring surgery within 5 years.

W-DRAG: A joint framework of WGAN with data random augmentation optimized for generative networks for bone marrow edema detection in dual energy CT.

Dual-energy computed tomography (CT) is an excellent substitute for identifying bone marrow edema in magnetic resonance imaging. However, it is rarely used in practice owing to its low contrast. To overcome this problem, we constructed a framework based on deep learning techniques to screen for diseases using axial bone images and to identify the local positions of bone lesions. To address the limited availability of labeled samples, we developed a new generative adversarial network (GAN) that extends expressions beyond conventional augmentation (CA) methods based on geometric transformations. We theoretically and experimentally determined that combining the concepts of data augmentation optimized for GAN training (DAG) and Wasserstein GAN yields a considerably stable generation of synthetic images and effectively aligns their distribution with that of real images, thereby achieving a high degree of similarity. The classification model was trained using real and synthetic samples. Consequently, the GAN technique used in the diagnostic test had an improved F1 score of approximately 7.8% compared with CA. The final F1 score was 80.24%, and the recall and precision were 84.3% and 88.7%, respectively. The results obtained using the augmented samples outperformed those obtained using pure real samples without augmentation. In addition, we adopted explainable AI techniques that leverage a class activation map (CAM) and principal component analysis to facilitate visual analysis of the network's results. The framework was designed to suggest an attention map and scattering plot to visually explain the disease predictions of the network.

Bone marrow edema of the knee: a narrative review.

Bone marrow edema (BME) is a frequent MRI finding in patients with knee pain. According to the etiology, BME of the knee can be classified into three main categories: ischemic, mechanic, and reactive. The diagnosis may be difficult, because of the specificity of symptoms and the poor radiographic findings. MRI is the gold standard, showing an area of altered signal of the bone with an high signal intensity on fat-suppressed, T2 weighted images, usually in combination with an intermediate or low signal intensity on T1 weighted images. Bone marrow edema tends to be self-limiting and, in most cases, resolves without any consequences in a varying amount of time. However, since it may evolve to complete joint destruction, early diagnosis and correct treatment are crucial to prevent the articular degeneration. Conservative therapy is the first step, with no weight-bearing for 3 to 6 weeks on the affected side, in combination with the administration of anti-inflammatory drugs or painkillers to manage symptoms. In non-responding forms and more advanced stages, minimally invasive preservative surgery can provide significant results, with subchondroplasty and core decompression being the two main procedures available. Knee arthroplasty, both total (TKA) or unicompartmental (UKA), is the only effective option when the degradation of cartilage is diffuse and in patients with subchondral bone collapse.

Efficacy of fat quantification methods used in MRI to distinguish between normal, benign, and malignant bone marrow pathologies in children.

BACKGROUND: Fat quantification methods in magnetic resonance imaging (MRI) have been studied to differentiate bone marrow pathologies in adult patients; however, scarce literature is available in pediatric patients. PURPOSE: To evaluate the efficacy of the T1 signal intensity value (T1-SIV), out-of-phase/in-phase signal ratio (OP/IP SR), and fat fraction (FF) to differentiate between normal, benign, and malignant pathological processes. MATERIAL AND METHODS: A total of 48 pediatric patients with lumbar and pelvic MRI were classified into three groups according to bone marrow pathology (group 1, normal; group 2, benign pathology/reconversion; group 3, malignant). The efficacy of T1-SIV, OP/IP SR, and FF values in differentiating these pathologies was evaluated using Kruskal-Wallis or analysis of variance and followed by Bonferroni or Dunn-Bonferroni tests. Cutoff values for malignant infiltration were defined using ROC analysis. RESULTS: Although these values were significantly different in all three groups (P = 0.001-0.008), this difference was not sufficient to discriminate between all groups. Subgroup analyses showed significant differences in T1-SIV between groups 1-3, in OP/IP SR between groups 1-3, 2-3, and 1-2, in FF between groups 1-2 and 1-3 in various regions (P = 0.001-0.049). Cutoff values had a sensitivity and specificity of 90%-100% for OP/IP SR and FF. CONCLUSION: T1-SIV, OP/IP SR, and FF may potentially distinguish normal from pathological bone marrow. OP/IP SR and FF values detected malignant infiltration with high sensitivity and specificity in this study. However, only OP/IP SR may significantly differentiate benign and malignant bone marrow pathologies which needs to be confirmed in the future study with a larger patient population.

Pictorial review: challenges in distinguishing bilateral metaphyseal marrow abnormalities on magnetic resonance imaging.

The paediatric metaphysis is afflicted by a wide range of pathological processes as it is the most metabolically active and well-vascularised part of the developing skeleton. This review focuses on metaphyseal marrow signal change detected with magnetic resonance imaging, which is most often occult on radiographs. When bilateral, these imaging appearances frequently present a diagnostic quandary. This review assists the radiologist to confidently dismiss physiological signal change and confidently work through the differential diagnosis. This is achieved by illustrating a practical method of classifying signal change into four categories: physiological red marrow, red marrow reconversion, marrow infiltration, and oedema-like marrow signal intensity. In doing so, various pathological entities are reviewed along with imaging pearls and next-step investigations.

Prediction of the activity of early ankylosing spondylitis using radiomics texture analysis on short tau inversion recovery (STIR).

OBJECTIVES: The study aimed to explore the value of texture analysis of radiomics based on the short tau inversion recovery (STIR) sequence to evaluate the activity of bone marrow oedema of sacroiliac joints in early AS. METHODS: 43 patients with early AS whose data were randomly divided into the training cohort (n=116) and verification cohort (n=56) according to the ratio of 7:3. The optimal feature subsets were obtained by Mann-Whitney U-test, the minimum-Redundancy Maximum-Relevancy (mRMR), and then least absolute shrinkage and selection operator (LASSO) using these texture feature parameters, which were used to construct the final prediction model and obtained the Radscore. The ROC curve was performed to evaluate the performance of the model. The Spearman correlation test was used to analyse the correlation of various indicators. RESULTS: In the training cohort, to differentiate early AS sacroiliac joint bone marrow oedema between the active and stable groups, the AUCs of the Radscore, SPARCC and ADC were 0.81, 0.91, 0.78, respectively. In the validation cohort, the AUCs were 0.87, 0.89, 0.85. In the two cohorts, there were no significant differences in AUCs between values of the Radscore and SPARCC, ADC (p>0.05). There was a significant difference in AUC between SPARCC and ADC in the training cohort (p<0.05), with no statistical significance in the validation cohort (p>0.05). The correlations were all low between the Radscore values and the values of ESR, CRP, tI, ASDAS-ESR and ASDAS-CRP (p<0.05). CONCLUSIONS: Radiomics analysis based on STIR texture analysis has a good prediction for the evaluation of bone marrow oedema activity of sacroiliac joints in AS. It can be a new non-invasive and objective evaluation method for AS activity.

[The detecting value of virtual non-calcium technique of dual-energy CT for bone marrow edema around nontraumatic osteonecrosis of the femoral head].

Objective: To evaluate the value of virtual non-calcium (VNCa) technique of dual-energy CT (DECT) for detecting bone marrow edema (BME) around nontraumatic osteonecrosis of the femoral head (ONFH) using MRI as reference standard. Methods: Nontraumatic ONFH patients were prospectively studied in the Fourth Medical Center of Chinese PLA General Hospital from October 2022 to May 2023, and their MRI and DECT images were analyzed. The diagnostic efficiency of the subjective assessment of BME around ONFH by two radiologists in VNCa color-coded images were calculated using the MRI results as the reference standard. The BME ranges were compared between VNCa images and MRI. Traditional CT values and VNCa CT values were compared between normal bone marrow and BME. The receiver operator characteristic (ROC) curve was established based on the statistically different CT values, and the area under the curve (AUC) was calculated to find the threshold to distinguish normal bone marrow from BME and evaluate the diagnostic efficacy. Results: Thirty patients with ONFH were included, including 24 males and 6 females, aged (39±12) years. There were 18 bilateral hips and 12 unilateral hips, with a total of 48 hips, 34 hips of which showed BME on MRI. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of subjective detection of BME on VNCa color coded maps by two physicians were 97.1% (33/34) and 97.1% (33/34), 92.9% (13/14) and 71.4% (10/14), 97.1% (33/34) and 89.2% (33/37), 92.9% (13/14) and 90.9% (10/11), 95.8% (46/48) and 89.6% (43/48), respectively, with no statistical difference (all P>0.05).There was no statistical difference between VNCa color-coded images and MRI in the BME range (P=1.160). The traditional CT values measured by the two radiologists were in good agreement with VNCa CT values, with intraclass correlation coefficient (ICC) of 0.948 (95%CI: 0.908-0.971) and 0.982 (95%CI: 0.969-0.990), respectively. The traditional CT value of normal bone marrow was (400.7±82.8) HU, and that of BME was (443.7±65.7) HU, with no statistical difference (P=0.062). The VNCa CT value of normal bone marrow was (-103.1±27.8) HU, and that of BME was (-32.9±25.7) HU, with statistical difference (P<0.001). The AUC of distinguishing normal bone marrow from BME based on VNCa CT value was 0.958 (95%CI: 0.857-0.995). The best cut-off value was -74.5 HU, and when the VNCa CT value was higher than -74.5 HU, the sensitivity, specificity, PPV, NPV and accuracy of diagnosing BME were 97.1%, 92.9%, 97.1%, 92.9% and 95.8 %, respectively. Conclusion: The VNCa technique of DECT has high efficiency in detecting BME around ONFH, and can accurately demonstrate the range of BME.

Differentiation of bone metastases from benign red marrow depositions: utilizing qualitative and quantitative analysis of conventional T1-weighted imaging and fat-suppressed T2-weighted imaging.

OBJECTIVES: To distinguish bone metastases (BMs) from benign red marrow depositions (BRMs) by qualitative and quantitative analyses of T1-weighted imaging and fat-suppressed T2-weighted imaging (T2 FS). METHODS: For 75 lesions including 38 BMs and 37 BRMs, two radiologists independently evaluated magnetic resonance images by qualitative (signal intensity [SI] of lesions compared to that of normal muscle [NM] or normal bone marrow [NBM]) and quantitative (parameters of the region of interests in the lesions, including T1 ratio [T1 SI ratio of lesion and NM], T2FMu ratio [T2 FS SI ratio of lesion and NM], and T2FMa ratio [T2 FS SI ratio of lesion and NBM]) analyses. RESULTS: Hyperintensity relative to NM or NBM on T2 FS was more frequent in BMs than in BRMs (100% vs 59.5%-78.4%, respectively; P ≤ 0.001) but also was present in more than half of BRMs. All quantitative parameters showed a significant difference between BMs and BRMs (T1 ratio, 1.075 vs 1.227 [P = 0.002]; T2FMu ratio, 2.094 vs 1.282 [P < 0.001]; T2FMa ratio, 3.232 vs 1.810 [P < 0.001]). The receiver operating characteristics areas under the curves of T2FMu and T2FMa ratios were clinically useful (0.781 and 0.841, respectively) and did not demonstrate statistically significant differences. CONCLUSIONS: The quantitative analysis of T2 FS facilitates distinguishing between BMs and BRMs, regardless of whether the reference was NM or NBM. ADVANCES IN KNOWLEDGE: Quantitative parameters derived from T2 FS facilitate differentiation of BMs BRMs without additional scans. The role of NBM as an internal standard for T2 FS to differentiate between BMs and BRMs is similar to that of NM.

Association between sacroiliac joint forms and subchondral changes in patients with Crohn's disease.

AIM: To assess the relationship between anatomical variants of sacroiliac joint (SIJ) and subchondral changes detected in magnetic resonance enterography (MRE) in patients with Crohn's disease (CD). METHODS: This was a retrospective study of 60 CD patients, who were divided into two groups: with (n = 16) and without SIJ (n = 44) involvement, depending on the presence of inflammatory (bone marrow edema) and structural changes (sclerosis and erosions) in MRE. Anatomical variants of SIJ were assessed in CT of the abdomen and/or pelvis, distinguishing typical form with convex iliac surface and atypical forms. Univariate and multivariate analyses were performed to reveal an association between joint changes and forms. RESULTS: Our study included 60 patients (38 males; mean age 38.72 years ± 13.33). Patients with SIJ changes were older (p = .044). No significant differences in CD localization and behavior were found. The most common SIJ lesions were structural changes (in 75% of patients); the main atypical form was the iliosacral complex. The univariate and multivariate analyses showed a significant association of atypical forms with total subchondral changes (odds ratio [OR]: 3.429, 95% confidence interval [CI] 1.043-11.268; p = .042; OR: 5.066, 95% CI: 1.273-20.167; p = .021, respectively), and with structural changes (OR: 4.185, 95% CI: 1.155-15.160; p = .029; OR: 5.986, 95% CI: 1.293-27.700; p = .022, respectively). CONCLUSION: Atypical forms of SIJ are a risk factor for the occurrence of structural joint changes in CD patients. An association between bone marrow edema and atypical forms was not found.

Factors associated with trajectories of bone marrow lesions over 4 years: data from the Osteoarthritis Initiative.

OBJECTIVE: To identify bone marrow lesion (BML) trajectories over 4 years and their demographic and structural predictors in middle-aged and older adults with or at increased risk of knee osteoarthritis (OA). METHODS: A total of 614 participants (mean age 61 years, 62% female) from the Osteoarthritis Initiative cohort (OAI) were included. BMLs in 15 anatomical locations of the knee were measured annually from baseline to 4 years using the Magnetic Resonance Imaging Osteoarthritis Knee Score (MOAKS) method. BML trajectories were determined using latent class mixed models (LCMMs). Multinomial logistic regression was used to examine baseline characteristics that predicted BML trajectories. RESULTS: Three distinct BML trajectories were identified: "Mild-stable BMLs" (25.9%), "Moderate-stable BMLs" (66.4%), and "Rapid-rise BMLs" (7.7%). Compared to the "Mild-stable BMLs" trajectory, current smokers were more likely to be in the "Moderate-stable BMLs" (odds ratio [OR] 2.089, P < 0.001) and "Rapid-rise" (OR 2.462, P < 0.001) trajectories. Moreover, female sex and meniscal tears were associated with an increased risk of being in the "Rapid-rise BMLs" trajectory (OR 2.023 to 2.504, P < 0.05). Participants who had higher education levels and drank more alcohol were more likely to be in the "Rapid-rise BMLs" trajectory (OR 1.624 to 3.178, P < 0.05) and less likely to be in the "Moderate-stable BMLs" trajectory (OR 0.668 to 0.674, P < 0.05). CONCLUSIONS: During the 4-year follow-up, most participants had relatively stable BMLs, few had enlarged BMLs, and no trajectory of decreased BMLs was identified. Sociodemographic factors, lifestyle, and knee structural pathology play roles in predicting distinct BML trajectories.

Characterization of bone marrow lesions in axial spondyloarthritis using quantitative T1 mapping MRI.

OBJECTIVE: Conventional magnetic resonance imaging (MRI) uses T1-weighted and short-tau inversion recovery (STIR) sequences to characterize bone marrow in axial spondyloarthritis. However, quantification is restricted to estimating the extent of lesions because signal intensities are highly variable both within individuals and across patients and MRI scanners. This study evaluates the performance of quantitative T1 mapping for distinguishing different types of bone marrow lesions of the sacroiliac joints. MATERIALS AND METHODS: In this prospective study, 62 patients underwent computed tomography (CT) and MRI of the sacroiliac joints including T1, STIR, and T1 mapping. Bone marrow lesions were characterized by three readers and assigned to one of four groups: sclerosis, osteitis, fat lesions, and mixed marrow lesions. Relaxation times on T1 maps were compared using generalized estimating equations and receiver operating characteristics (ROC) analysis. RESULTS: A total of 119 lesions were selected (sclerosis: 38, osteitis: 27, fat lesions: 40; mixed lesions: 14). T1 maps showed highly significant differences between the lesions with the lowest values for sclerosis (1516±220 ms), followed by osteitis (1909±75 ms), and fat lesions (2391±200 ms); p<0.001. T1 mapping differentiated lesions with areas under the ROC curve of 99% (sclerosis vs. osteitis) and 100% (other comparisons). CONCLUSION: T1 mapping allows accurate characterization of sclerosis, osteitis, and fat lesions at the sacroiliac joint but only for homogeneous, non-mixed lesions. Thus, further sequence development is needed before implementation in clinical routine.

Association of anatomical variants of the sacroiliac joint with bone marrow edema in patients with axial spondyloarthritis.

OBJECTIVE: To determine the prevalence of sacroiliac joint variants in patients with axial spondyloarthritis (axSpA) using MRI-based synthetic CT images and to evaluate their relationships with the presence of bone marrow edema, as this may potentially complicate diagnosing active sacroiliitis on MRI in patients with suspected axSpA. METHODS: 172 patients were retrospectively included. All patients underwent MRI because of clinical suspicion of sacroiliitis. The diagnosis of axSpA was made by a tertiary hospital rheumatologist. Two readers independently determined the presence of bone marrow edema and the presence of one or more of the nine known sacroiliac joint (SIJ) variants. RESULTS: SIJ variants were common in axSpA patients (82.9%) and the non-SpA group (85.4%); there were no significant differences in prevalence. Bone marrow edema was frequently found in axSpA (86.8%) and non-SpA patients (34%). AxSpA patients with SIJ variants (except for accessory joint) demonstrated 4 to 10 times higher odds for bone marrow edema, however not statistically significant. The more variants were present in this group, the higher the chance of bone marrow edema. However, some multicollinearity cannot be excluded, since bone marrow edema is very frequent in the axSpA group by definition. CONCLUSION: SIJ variants are common in axSpA and non-SpA patients. SIJ variants were associated with higher prevalence of bone marrow edema in axSpA patients, potentially due to altered biomechanics, except for accessory joint which may act as a stabilizer.

Diagnostic accuracy of dual-energy CT for bone marrow edema in patients with acute knee injury: a systematic review and meta-analysis.

BACKGROUND: Knee injuries are prevalent, and early diagnosis is crucial for guiding clinical therapy. MRI is the diagnostic gold standard for bone marrow edema (BME) in patients with acute knee injuries, yet there are still limitations. Dual-energy CT, a possible viable replacement, is being explored (DECT). METHODS: We systematically retrieved studies from EMBASE, Scopus, PUBMED, and the Cochrane Library and collected gray literatures. In accordance with the PRISMA-DTA standards, a systematic review was conducted between the study's initiation and July 31, 2021, utilizing an MRI reference standard and at least 10 adult patients with acute knee injuries to evaluate the diagnostic effectiveness of DECT for diagnosing BME. Two reviewers collected the study's details independently. For the meta-analysis, a bivariate mixed-effects regression model was utilized, and subgroup analysis was employed to determine the sources of variability. RESULTS: The research included nine studies that examined 290 individuals between the ages of 23 and 53 with acute knee injuries who had DECT and MRI. Overall, the sensitivity, specificity, and AUC of the BME were 85% (95% confidence interval [CI]: 77-90%), 96% (95% CI: 93-97%), and 0.97 (95% CI: 0.95-0.98), respectively. To account for the assumed diversity of research, there were no statistically significant differences between the comparison groups in terms of specificity and sensitivity. CONCLUSION: DECT is a viable alternative to MRI for individuals with acute knee injuries when MRI is inappropriate or unavailable.

Foot and Ankle Bone Marrow Edema Assessment in Long Distance Runners.

BACKGROUND: The clinical significance of foot and ankle bone marrow edema (BME) is poorly understood. METHODS: Magnetic resonance imaging (MRI) and the visual analog scale (VAS) pain scores were collected for 17 runners who participated in the Twin Cities Marathon, immediately postmarathon and at a 6-week follow-up. Bone marrow edema lesions were classified using lesion frequency, anatomical location, and grading scale change to calculate a BME score for each affected bone. Spearman rank correlation coefficient test was used to identify a possible correlation between VAS and postmarathon BME. A paired Student t test was used to detect differences between total mileage ran 6 weeks postmarathon in participants with or without BME. RESULTS: After completing the marathon, 8 BME lesions were identified in 5 participants (29.4%; 5/17), 3 were men, and 2 were women, with a mean age of 33.8 years (range: 24-52), and BMI of 22.9 ± 4. All lesions were resolved on 6-week follow-up imaging. VAS pain scores did not correlate with postmarathon BME. A significant difference in total miles logged over 6 weeks postmarathon could not be found among participants with and without BME. CONCLUSION: Foot and ankle BME changes identified by MRI were not correlated to clinical symptoms and may resolve with self-directed activity in less time than other areas of the lower extremity. LEVEL OF EVIDENCE: Level II, Prospective Cohort Study.

Machine Learning Pipeline for Predicting Bone Marrow Edema Along the Sacroiliac Joints on Magnetic Resonance Imaging.

OBJECTIVE: We aimed to develop and validate a fully automated machine learning (ML) algorithm that predicts bone marrow edema (BME) on a quadrant level in sacroiliac (SI) joint magnetic resonance imaging (MRI). METHODS: A computer vision workflow automatically locates the SI joints, segments regions of interest (ilium and sacrum), performs objective quadrant extraction, and predicts presence of BME, suggestive of inflammatory lesions, on a quadrant level in semicoronal slices of T1/T2-weighted MRI scans. Ground truth was determined by consensus among human readers. The inflammation classifier was trained using a ResNet18 backbone and five-fold cross-validated on scans of patients with spondyloarthritis (SpA) (n = 279), postpartum individuals (n = 71), and healthy subjects (n = 114). Independent SpA patient MRI scans (n = 243) served as test data set. Patient-level predictions were derived from aggregating quadrant-level predictions, ie, at least one positive quadrant. RESULTS: The algorithm automatically detects the SI joints with a precision of 98.4% and segments ilium/sacrum with an intersection over union of 85.6% and 67.9%, respectively. The inflammation classifier performed well in cross-validation: area under the curve (AUC) 94.5%, balanced accuracy (B-ACC) 80.5%, and F1 score 64.1%. In the test data set, AUC was 88.2%, B-ACC 72.1%, and F1 score 50.8%. On a patient level, the model achieved a B-ACC of 81.6% and 81.4% in the cross-validation and test data set, respectively. CONCLUSION: We propose a fully automated ML pipeline that enables objective and standardized evaluation of BME along the SI joints on MRI. This method has the potential to screen large numbers of patients with (suspected) SpA and is a step closer towards artificial intelligence-assisted diagnosis and follow-up.

Bone marrow edema in the sacroiliac joints is associated with the development of structural lesions at the same anatomical location over time in patients with axial spondyloarthritis.

OBJECTIVE: To assess whether the presence of bone marrow edema (BME) leads to the development of structural lesions at the same anatomical location of the sacroiliac joints (SIJ), and to investigate the association between BME patterns over time and structural lesions in patients with early axial spondyloarthritis (axSpA). METHODS: Patients with axSpA from the DESIR cohort with ≥2 consecutive magnetic resonance imaging (MRI)-SIJ were assessed at baseline, 2 and 5 years. MRI-SIJ images were divided into 8 quadrants. The association between BME and subsequent structural lesions (sclerosis, erosions, fatty lesions, and ankylosis) on MRI in the same quadrant was tested longitudinally. Additionally, patients were grouped according to the pattern of BME evolution across quadrants over time (no BME, sporadic, fluctuating, and persistent). The association between these patterns and 5-year imaging outcomes (eg: ≥5 erosions and/or fatty lesions on MRI-SIJ) was tested. RESULTS: In total, 196 patients were included. BME in each quadrant was associated with sclerosis (OR:1.9 (95%CI: 1.1;3.4)), erosions (1.9 (1.5;2.5)) and fatty lesions (1.9 (1.4;2.6)). Ankylosis was uncommon. There was a gradient between increased level of inflammation and subsequent damage: compared to the 'no BME' pattern, the sporadic (OR (95% CI): 2.1 (1.0;4.5)), fluctuating (OR:5.6(2.2;14.4)) and persistent (OR:7.5(2.8;19.6)) patterns were associated with higher structural damage on MRI-SIJ at 5-years. CONCLUSIONS: In early axSpA, inflammation on MRI-SIJ leads to damage at the quadrant level. The higher the exposure to inflammation across quadrants in the SIJs over time the higher the likelihood of subsequent structural damage, suggesting a cumulative effect.

[Study on the relationship between severe knee osteoarthritis and bone marrow edema].

OBJECTIVE: To investigate the relationship between bone marrow edema and pathological changes, symptoms and signs of severe knee osteoarthritis. METHODS: From January 2020 to March 2021, 160 patients with severe knee osteoarthritis who underwrent MRI of the knee at the Department of Bone and Joint, Wangjing Hospital, China Academy of Chinese Medical Sciences were included. Eighty patients with bone marrow edema were selected as the case group, including 12 males and 68 females, aged from 51 to 80 years old with an average of (66.58±8.10) years old, the duration of disease 5 to 40 months with an average of (15.61±9.25) months. Eighty patients without bone marrow edema were selected as the control group, including 15 males and 65 females, aged from 50 to 80 years old with an average of (67.82±8.05) years old, the duration of disease 6 to 37 months with an average of (15.75±8.18) months, BMI was (28.26±3.13) kg·m-2 ranged from 21.39 to 34.46 kg·m-2. The degree of bone marrow edema was evaluated by knee whole oragan magnetic resonance imaging score (WORMS). The degree of knee osteoarthritis was evaluated by Kellgren- Lawrence(K-L) grade and Western Ontario and McMaster University Osteoarthritis Index (WOMAC). The degree of joint pain was evaluated by visual analogue scale(VAS) and WOMAC pain score, the joint signs were evaluated by tenderness, percussion pain, joint swelling and joint range of motion. To explore the relationship between bone marrow edema and knee osteoarthritis, the prevalence of bone marrow edema and K-L grade were compared between the two groups. Furthermore the WORMS score and WOMAC index, pain-related score, and sign-related score correlation coefficient were analyzed to further explore the relationship between bone marrow edema and knee osteoarthritis index, joint pain symptoms and signs. RESULTS: There was 68.75% (55/80) of the patients in the case group were in K-L grade Ⅳ, and 52.5% (42/80) in the control group, indicating a higher proportion of patients with grade Ⅳ in the case group than the control group (χ2=4.425, P<0.05). In the case group, there was a strong correlation between bone marrow edema WORMS score and knee osteoarthritis WOMAC index. (r=0.873>0.8, P<0.001), a moderate correlation between WORMS score and VAS score and WOMAC pain score(r=0.752, 0.650>0.5, P<0.001), a moderate correlation between WORMS score and percussion pain score (r=0.784>0.5, P<0.001), and a weak correlation between WORMS score and VAS and tenderness score, joint swelling score and joint range of motion score (r=0.194, 0.259, 0.296<0.3, P<0.001). CONCLUSION: Our study suggests that severe knee osteoarthritis is associated with an increased risk of bone marrow edema. Bone marrow edema can also lead to knee osteoarthritis joint pain, with percussion pain being a positive sign, but tenderness, joint swelling and limitation of activity are not significantly related to bone marrow edema.