As the most frequent cause of physical disability, musculoskeletal diseases such as arthritis and osteoporosis have a great social and economical impact. Quantitative magnetic resonance imaging (MRI) biomarkers are important tools that allow clinicians to better characterize, monitor, and even predict musculoskeletal disease progression. Post-processing pipelines often include image segmentation. Manually identifying the border of the region of interest (ROI) is a difficult and time-consuming task. Manual segmentation is also affected by inter- and intrauser variability, thus limiting standardization. Fully automatic or semi-automatic methods that minimize the user interaction are highly desirable. Unfortunately, an ultimate, highly reliable and extensively evaluated solution for joint and musculoskeletal tissue segmentation has not yet been proposed, and many clinical studies still adopt fully manual procedures. Moreover, the clinical translation of several promising quantitative MRI techniques is highly affected by the lack of an established, fast, and accurate segmentation method. The goal of this review is to present some of the techniques proposed in recent literature that have been adopted in clinical studies for joint and musculoskeletal tissue analyses in arthritis patients. The most widely used MRI sequences and image processing algorithms employed to accomplish segmentation challenges will be discussed in this paper.

Purpose To analyze how automatic segmentation translates in accuracy and precision to morphology and relaxometry compared with manual segmentation and increases the speed and accuracy of the work flow that uses quantitative magnetic resonance (MR) imaging to study knee degenerative diseases such as osteoarthritis (OA). Materials and Methods This retrospective study involved the analysis of 638 MR imaging volumes from two data cohorts acquired at 3.0 T: (a) spoiled gradient-recalled acquisition in the steady state T1_ρ-weighted images and (b) three-dimensional (3D) double-echo steady-state (DESS) images. A deep learning model based on the U-Net convolutional network architecture was developed to perform automatic segmentation. Cartilage and meniscus compartments were manually segmented by skilled technicians and radiologists for comparison. Performance of the automatic segmentation was evaluated on Dice coefficient overlap with the manual segmentation, as well as by the automatic segmentations' ability to quantify, in a longitudinally repeatable way, relaxometry and morphology. Results The models produced strong Dice coefficients, particularly for 3D-DESS images, ranging between 0.770 and 0.878 in the cartilage compartments to 0.809 and 0.753 for the lateral meniscus and medial meniscus, respectively. The models averaged 5 seconds to generate the automatic segmentations. Average correlations between manual and automatic quantification of T1_ρ and T2 values were 0.8233 and 0.8603, respectively, and 0.9349 and 0.9384 for volume and thickness, respectively. Longitudinal precision of the automatic method was comparable with that of the manual one. Conclusion U-Net demonstrates efficacy and precision in quickly generating accurate segmentations that can be used to extract relaxation times and morphologic characterization and values that can be used in the monitoring and diagnosis of OA. ^© RSNA, 2018 Online supplemental material is available for this article.

Purpose

To develop an automated pipeline based on convolutional neural networks to segment lumbar intervertebral discs and characterize their biochemical composition using voxel-based relaxometry, and establish local associations with clinical measures of disability, muscle changes, and other symptoms of lower back pain.

Methods

This work proposes a new methodology using MRI (n = 31, across the spectrum of disc degeneration) that combines deep learning-based segmentation, atlas-based registration, and statistical parametric mapping for voxel-based analysis of T_1ρ and T₂ relaxation time maps to characterize disc degeneration and its associated disability.

Results

Across degenerative grades, the segmentation algorithm produced accurate, high-confidence segmentations of the lumbar discs in two independent data sets. Manually and automatically extracted mean disc T_1ρ and T₂ relaxation times were in high agreement for all discs with minimal bias. On a voxel-by-voxel basis, imaging-based degenerative grades were strongly negatively correlated with T_1ρ and T₂ , particularly in the nucleus. Stratifying patients by disability grades revealed significant differences in the relaxation maps between minimal/moderate versus severe disability: The average T_1ρ relaxation maps from the minimal/moderate disability group showed clear annulus nucleus distinction with a visible midline, whereas the severe disability group had lower average T_1ρ values with a homogeneous distribution.

Conclusion

This work presented a scalable pipeline for fast, automated assessment of disc relaxation times, and voxel-based relaxometry that overcomes limitations of current region of interest-based analysis methods and may enable greater insights and associations between disc degeneration, disability, and lower back pain.

Purpose

To develop and compare with the classical region of interest (ROI)-based approach a fully automatic, local, and unbiased way of studying the knee T1ρ relaxation time by creating an atlas and using voxel-based relaxometry (VBR) in osteoarthritis (OA) and anterior cruciate ligament (ACL) subjects.

Materials and methods

In this study 110 subjects from two cohorts: 1) Mild OA 40 patients with mild-OA Kellgren-Lawrence (KL) ≤ 2 and 15 controls KL ≤ 1; 2) ACL cohort (a model for early OA): 40 ACL-injured patients imaged prior to ACL reconstruction and 1-year postsurgery and 15 controls are analyzed. All the subjects were acquired at 3T with a protocol that includes: 3D-FSE (CUBE) and 3D-T1ρ . A nonrigid registration technique was applied to align all the images on a single template. This allows for performing VBR to assess local statistical differences of T1ρ values using z-score analysis. VBR results were compared with those obtained with classical ROI-based technique.

Results

ROI-based results from atlas-based segmentation were consistent with classical ROI-based method (coefficient of variation [CV] = 3.83%). Voxel-based group analysis revealed local patterns that were overlooked by the ROI-based approach; eg, VBR showed posterior lateral femur and posterior lateral tibia significant T1ρ elevations in ACL-injured patients (sample mean z-score=9.7 and 10.3). Those elevations were overlooked by the classical ROI-based approach (sample mean z-score=1.87 and -1.73) CONCLUSION: VBR is a feasible and accurate tool for the local evaluation of the biochemical composition of knee articular cartilage. VBR is capable of detecting specific local patterns on T1ρ maps in OA and ACL subjects.

BACKGROUND: Although T1ρ and T2 have emerged as early indicators for hip osteoarthritis (OA), there is little information regarding longitudinal changes across the cartilage in the early stages of this disease. PURPOSE: To characterize the variability in 2-year hip cartilage T1ρ and T2 changes and investigate associations between these patterns of change and common indicators of hip OA. STUDY TYPE: Prospective. POPULATION: A total of 25 women (age: 51.9 ± 16.3 years old; BMI: 22.6 ± 2.0 kg/m2 ) and 17 men (age: 55.8 ± 14.9 years old; body mass index (BMI): 24.4 ± 3.8 kg/m2 ) who were healthy or with early-to-moderate hip OA. FIELD STRENGTH/SEQUENCE: A 3 T MRI (GE), 3D combined T1ρ /T2 magnetization-prepared angle-modulated partitioned k-space spoiled gradient echo snapshots. ASSESSMENT: Principal component (PC) analysis of Z-score difference maps of 2-year changes in hip cartilage T1ρ and T2 relaxation times, participant hip disability and osteoarthritis outcome scores (HOOS) and functional tests at 2-year follow-up. STATISTICAL TESTS: Shapiro-Wilk test, unpaired t-tests, Kruskal Wallis tests, Pearson or Spearman (ρ) correlations. Significance was set at P < 0.05. RESULTS: Women (-6.40 ± 14.48) had significantly lower T1ρ PC1 scores than men (10.05 ± 26.15). T1ρ PC4 was significantly correlated with HOOSsport , HOOSsymptoms , HOOSpain , HOOSadl , and HOOSqol at 2-year follow-up (ρ: [0.36, 0.50]). T1ρ PC2 and PC4 were significantly correlated with 30-second chair test (ρ = -0.39 and ρ = 0.24, respectively) and side plank (ρ = -0.32 and ρ = 0.21). T1ρ and T2 PC2 were significantly correlated with 40 m walk test (ρ = 0.34 and ρ = 0.31) and 30-second chair rise test (ρ = -0.39 and ρ = -0.32). DATA CONCLUSION: Men exhibited accelerated T1ρ increases across the femoral cartilage compared to women, suggesting sex should be considered when evaluating early hip OA. Participants with poorer HOOS and function exhibited greater T1ρ and T2 increases in superior and anterior femoral cartilage and greater T1ρ increases in the anterior femoral cartilage. These patterns of short-term relaxometry increases could indicate hip OA progression. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 3.

MRI T₂ mapping sequences quantitatively assess tissue health and depict early degenerative changes in musculoskeletal (MSK) tissues like cartilage and intervertebral discs (IVDs) but require long acquisition times. In MSK imaging, small features in cartilage and IVDs are crucial for diagnoses and must be preserved when reconstructing accelerated data. To these ends, we propose region of interest-specific postprocessing of accelerated acquisitions: a recurrent UNet deep learning architecture that provides T₂ maps in knee cartilage, hip cartilage, and lumbar spine IVDs from accelerated T₂-prepared snapshot gradient-echo acquisitions, optimizing for cartilage and IVD performance with a multi-component loss function that most heavily penalizes errors in those regions. Quantification errors in knee and hip cartilage were under 10% and 9% from acceleration factors R = 2 through 10, respectively, with bias for both under 3 ms for most of R = 2 through 12. In IVDs, mean quantification errors were under 12% from R = 2 through 6. A Gray Level Co-Occurrence Matrix-based scheme showed knee and hip pipelines outperformed state-of-the-art models, retaining smooth textures for most R and sharper ones through moderate R. Our methodology yields robust T₂ maps while offering new approaches for optimizing and evaluating reconstruction algorithms to facilitate better preservation of small, clinically relevant features.

Purpose

To study the local distribution of hip cartilage T_1ρ and T₂ relaxation times and their association with changes in patient reported outcome measures (PROMs) using a fully automatic, local, and unbiased method in subjects with and without hip osteoarthritis (OA).

Materials and methods

The 3 Tesla MRI studies of the hip were obtained for 37 healthy controls and 16 subjects with radiographic hip OA. The imaging protocol included a three-dimensional (3D) SPGR sequence and a combined 3D T_1ρ and T₂ sequence. Quantitative cartilage analysis was compared between a traditional region of interest (ROI)-based method and a fully automatic voxel-based relaxometry (VBR) method. Additionally, VBR was used to assess local T_1ρ and T₂ differences between subjects with and without OA, and to evaluate the association between T_1ρ and T₂ and 18-month changes PROMs.

Results

Results for the two methods were consistent in the acetabular (R = 0.79; coefficients of variation [CV] = 2.9%) and femoral cartilage (R = 0.90; CV = 2.6%). VBR revealed local patterns of T_1ρ and T₂ elevation in OA subjects, particularly in the posterosuperior acetabular cartilage (T_1ρ : P = 0.02; T₂ : P = 0.038). Overall, higher T_1ρ and T₂ values at baseline, particularly in the anterosuperior acetabular cartilage (T_1ρ : Rho = -0.42; P = 0.002; T₂ : Rho = -0.44; P = 0.002), were associated with worsening PROMS at 18-month follow-up.

Conclusion

VBR is an accurate and robust method for quantitative MRI analysis in hip cartilage. VBR showed the capability to detect local variations in T_1ρ and T₂ values in subjects with and without osteoarthritis, and voxel based correlations demonstrated a regional dependence between baseline T_1ρ and T₂ values and changes in PROMs.

Level of evidence

1 J. MAGN. RESON. IMAGING 2017;45:1523-1533.

BACKGROUND: The purpose of this study was to develop a deep learning approach to automatically segment the scapular bone on magnetic resonance imaging (MRI) images and to compare the accuracy of these three-dimensional (3D) models with that of 3D computed tomography (CT). METHODS: Fifty-five patients with high-resolution 3D fat-saturated T2 MRI were retrospectively identified. The underlying pathology included rotator cuff tendinopathy and tears, shoulder instability, and impingement. Two experienced musculoskeletal researchers manually segmented the scapular bone. Five cross-validation training and validation splits were generated to independently train two-dimensional (2D) and 3D models using a convolutional neural network approach. Model performance was evaluated using the Dice similarity coefficient (DSC). All models with DSC > 0.70 were ensembled and used for the test set, which consisted of four patients with matching high-resolution MRI and CT scans. Clinically relevant glenoid measurements, including glenoid height, width, and retroversion, were calculated for two of the patients. Paired t-tests and Wilcoxon signed-rank tests were used to compare the DSC of the models. RESULTS: The 2D and 3D models achieved a best DSC of 0.86 and 0.82, respectively, with no significant difference observed. Augmentation of imaging data significantly improved 3D but not 2D model performance. In comparing clinical measurements of 3D MRI and CT, there was a mean difference ranging from 1.29 mm to 3.46 mm and 0.05° to 7.47°. CONCLUSION: We have presented a fully automatic, deep learning-based strategy for extracting scapular shape from a high-resolution MRI scan. Further developments of this technology have the potential to allow for surgeons to obtain all clinically relevant information from MRI scans and reduce the need for multiple imaging studies for patients with shoulder pathology.

Background

Bone-cartilage interactions have been implicated in causing osteoarthritis (OA).

Purpose

To use [¹⁸ F]-NaF PET-MRI to 1) develop automatic image processing code in MatLab to create a model of bone-cartilage interactions and 2) find associations of bone-cartilage interactions with known manifestations of OA.

Study type

Prospective study aimed to evaluate a data analysis method.

Population

Twenty-nine patients with knee pain or joint stiffness.

Field strength/sequence

3T MRI (GE), 3D CUBE FSE, 3D combined T₁ ρ/T₂ MAPSS, [18F]-sodium fluoride, SIGNA TOF (OSEM).

Assessment

Correlation between MRI (cartilage) and PET (bone) quantitative parameters, bone-cartilage interactions model described by modes of variation as derived by principal component analysis (PCA), WORMS scoring on cartilage lesions, bone marrow abnormalities, subchondral cysts.

Statistical tests

Linear regression, Pearson correlation.

Results

Mode 1 was a positive predictor of the bone abnormality score (P = 0.0003, P = 0.001, P = 0.0007) and the cartilage lesion score (P = 0.03, P = 0.01, P = 0.02) in the femur, tibia, and patella, respectively. For the cartilage lesion scores, mode 5 was the most important positive predictor in the femur (P = 3.9E-06), and mode 2 were predictors, significant negative predictor in the tibia (P = 0.007). In the patella, mode 1 was a significant positive predictor of the bone abnormality score (P = 0.0007).

Data conclusion

By successfully building an automatic code to create a bone-cartilage interface, we were able to observe dynamic relationships between biochemical changes in the cartilage accompanied with bone remodeling, extended to the whole knee joint instead of simple colocalized observations, shedding light on the interactions that occur between bone and cartilage in OA. Evidence Level: 3 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2020;52:1462-1474.