Diagnostic Accuracy of Quantitative Multicontrast 5-Minute Knee MRI Using Prospective Artificial Intelligence Image Quality Enhancement

Time-efficient combined morphologic and quantitative joint MRI: an in situ study of standardized knee cartilage defects in human cadaveric specimens

BACKGROUND

Quantitative techniques such as T2 and T1ρ mapping allow evaluating the cartilage and meniscus. We evaluated multi-interleaved X-prepared turbo-spin echo with intuitive relaxometry (MIXTURE) sequences with turbo spin-echo (TSE) contrast and additional parameter maps versus reference TSE sequences in an in situ model of human cartilage defects.

METHODS

Standardized cartilage defects of 8, 5, and 3 mm in diameter were created in the lateral femora of ten human cadaveric knee specimens (81 ± 10 years old; nine males, one female). MIXTURE sequences providing proton density-weighted fat-saturated images and T2 maps or T1-weighted images and T1ρ maps as well as the corresponding two- and three-dimensional TSE reference sequences were acquired before and after defect creation (3-T scanner; knee coil). Defect delineability, bone texture, and cartilage relaxation times were quantified. Appropriate parametric or non-parametric tests were used.

RESULTS

Overall, defect delineability and texture features were not significantly different between the MIXTURE and reference sequences (p ≤ 0.47). After defect creation, relaxation times significantly increased in the central femur (T2pre = 51 ± 4 ms [mean ± standard deviation] versus T2post = 56 ± 4 ms; p = 0.002) and all regions combined (T1ρpre = 40 ± 4 ms versus T1ρpost = 43 ± 4 ms; p = 0.004).

CONCLUSIONS

MIXTURE permitted time-efficient simultaneous morphologic and quantitative joint assessment based on clinical image contrasts. While providing T2 or T1ρ maps in clinically feasible scan time, morphologic image features, i.e., cartilage defects and bone texture, were comparable between MIXTURE and reference sequences.

RELEVANCE STATEMENT

Equally time-efficient and versatile, the MIXTURE sequence platform combines morphologic imaging using familiar contrasts, excellent image correspondence versus corresponding reference sequences and quantitative mapping information, thereby increasing the diagnostic value beyond mere morphology.

KEY POINTS

• Combined morphologic and quantitative MIXTURE sequences are based on three-dimensional TSE contrasts. • MIXTURE sequences were studied in an in situ human cartilage defect model. • Morphologic image features, i.e., defect delineabilty and bone texture, were investigated. • Morphologic image features were similar between MIXTURE and reference sequences. • MIXTURE allowed time-efficient simultaneous morphologic and quantitative knee joint assessment.

Two for One-Combined Morphologic and Quantitative Knee Joint MRI Using a Versatile Turbo Spin-Echo Platform

Quantitative MRI techniques such as T2 and T1ρ mapping are beneficial in evaluating knee joint pathologies; however, long acquisition times limit their clinical adoption. MIXTURE (Multi-Interleaved X-prepared Turbo Spin-Echo with IntUitive RElaxometry) provides a versatile turbo spin-echo (TSE) platform for simultaneous morphologic and quantitative joint imaging. Two MIXTURE sequences were designed along clinical requirements: "MIX1", combining proton density (PD)-weighted fat-saturated (FS) images and T2 mapping (acquisition time: 4:59 min), and "MIX2", combining T1-weighted images and T1ρ mapping (6:38 min). MIXTURE sequences and their reference 2D and 3D TSE counterparts were acquired from ten human cadaveric knee joints at 3.0 T. Contrast, contrast-to-noise ratios, and coefficients of variation were comparatively evaluated using parametric tests. Clinical radiologists (n = 3) assessed diagnostic quality as a function of sequence and anatomic structure using five-point Likert scales and ordinal regression, with a significance level of α = 0.01. MIX1 and MIX2 had at least equal diagnostic quality compared to reference sequences of the same image weighting. Contrast, contrast-to-noise ratios, and coefficients of variation were largely similar for the PD-weighted FS and T1-weighted images. In clinically feasible scan times, MIXTURE sequences yield morphologic, TSE-based images of diagnostic quality and quantitative parameter maps with additional insights on soft tissue composition and ultrastructure.

[Current MR imaging of cartilage in the context of knee osteoarthritis (part 1) : Principles and sequences]

Magnetic resonance imaging (MRI) is the clinical method of choice for cartilage imaging in the context of degenerative and nondegenerative joint diseases. The MRI-based definitions of osteoarthritis rely on the detection of osteophytes, cartilage pathologies, bone marrow edema and meniscal lesions but currently a scientific consensus is lacking. In the clinical routine proton density-weighted, fat-suppressed 2D turbo spin echo sequences with echo times of 30-40 ms are predominantly used, which are sufficiently sensitive and specific for the assessment of cartilage. The additionally acquired T1-weighted sequences are primarily used for evaluating other intra-articular and periarticular structures. Diagnostically relevant artifacts include magic angle and chemical shift artifacts, which can lead to artificial signal enhancement in cartilage or incorrect representations of the subchondral lamina and its thickness. Although scientifically validated, high-resolution 3D gradient echo sequences (for cartilage segmentation) and compositional MR sequences (for quantification of physical tissue parameters) are currently reserved for scientific research questions. The future integration of artificial intelligence techniques in areas such as image reconstruction (to reduce scan times while maintaining image quality), image analysis (for automated identification of cartilage defects), and image postprocessing (for automated segmentation of cartilage in terms of volume and thickness) will significantly improve the diagnostic workflow and advance the field further.

Editorial Commentary: Evaluation for Cartilage Lesions on Magnetic Resonance Imaging Continues to Improve: Artificial Intelligence Applications May Result in Higher Sensitivity and Specificity

Accurate detection of cartilage lesions of the knee is required to offer patient-specific care and can alter surgical intervention options. To date, diagnostic arthroscopy remains the gold standard yet often requires the need for staged operative procedure for treatment. Magnetic resonance imaging (MRI) is the most accurate imaging modality with high specificity, yet even with recent advances, MRI has limited specificity. Newer scanners (3 T) and updated scanning sequences (3-dimensional MRI and quantitative MRI) are most sensitive in characterizing cartilage lesions of the knee, but these resources are not available to all users. Promising new avenues for patient-specific MRI scans along with the utilization of artificial intelligence will more accurately identify and quantify lesion size, location, and depth.

T2 Mapping of Patellar Cartilage After a Single First-Time Episode of Traumatic Lateral Patellar Dislocation

BACKGROUND

In most cases, lateral patellar dislocation (LPD) is accompanied by chondral injury and may initiate gradual degeneration of patellar cartilage, which might be detected with a T2 mapping, a well-established method for cartilage lesions assessment.

PURPOSE

To examine short-term consequences of single first-time LPD in teenagers by T2 mapping of the patellar-cartilage state.

STUDY TYPE

Prospective.

POPULATION

95 patients (mean age: 15.1 ± 2.3; male/female: 46/49) with first-time, complete, traumatic LPD and 51 healthy controls (mean age: 14.7 ± 2.2, male/female: 29/22).

FIELD STRENGTH/SEQUENCE

3.0 T; axial T2 mapping acquired using a 2D turbo spin-echo sequence.

ASSESSMENT

MRI examination was conducted 2-4 months after first LPD. T2 values were calculated in manually segmented cartilage area via averaging over three middle level slices in six cartilage regions: deep, intermediate, superficial layers, and medial lateral parts.

STATISTICAL TESTS

ANOVA analysis with Tukey's multiple comparison test, one-vs.-rest logistic regression analysis. The threshold of significance was set at P < 0.05.

RESULTS

In lateral patellar cartilage, a significant increase in T2 values was found in deep and intermediate layers in both patient groups with mild (deep: 34.7 vs. 31.3 msec, intermediate: 38.7 vs. 34.6 msec, effect size = 0.55) and severe (34.8 vs. 31.3 msec, 39.1 vs. 34.6 msec, 0.55) LPD consequences as compared to controls. In the medial facet, only severe cartilage damage showed significant prolongation of T2 times in the deep layer (34.3 vs. 30.7 msec, 0.55). No significant changes in T2 values were found in the lateral superficial layer (P = 0.99), whereas mild chondromalacia resulted in a significant decrease of T2 in the medial superficial layer (41.0 vs. 43.8 msec, 0.55).

DATA CONCLUSION

The study revealed substantial difference in T2 changes after LPD between medial and lateral areas of patellar cartilage.

EVIDENCE LEVEL

2 TECHNICAL EFFICACY STAGE: 2.

Quantitative Magnetic Resonance Imaging Had Greater Sensitivity in Diagnosing Chondral Lesions of the Knee: A Systematic Review and Meta-Analysis

PURPOSE

To investigate the accuracy and reliability of magnetic resonance imaging (MRI) in identifying and grading chondral lesions and explore the optimal imaging technique to image cartilage.

METHOD

A comprehensive search was conducted on Medline, Embase, and Cochrane Library. Eligible cohort studies published before August 2022 were included. The study reports used MRI to diagnose and grade cartilage lesions, with intraoperative findings as the reference standard. Summary estimates of diagnostic performance were obtained. The reliability of MRI interpretation was summarized. Subgroup analyses were performed based on assessed imaging techniques, field strength, and joint surface.

RESULTS

Forty-three trials and 3,706 patients were included in the systematic review. The overall area under curve for hierarchical summarized receiver operating characteristics was 0.91 (95% confidence interval [CI] 0.88-0.93). The pooled sensitivity for quantitative MRI, 3-dimensional MRI, and 2-dimensional MRI was 0.82 (95% CI 0.64-0.92), 0.79 (95% CI 0.74-0.83), and 0.63 (95% CI 0.51-0.73), respectively. The pooled sensitivity of 3 Tesla (3T), 1.5 Tesla (1.5T), and <1.5 Tesla MRI was 0.79 (95% CI 0.72-0.85), 0.67 (95% CI 0.60-0.74), and 0.55 (95% CI 0.39-0.71), respectively. There were differences in interobserver consistency across different studies.

CONCLUSIONS

In general, MRI had high specificity in discriminating normal cartilage, but its sensitivity for identifying chondral lesions is less optimal. Further analysis showed that quantitative MRI, 3D MRI, and 3T MRI demonstrate greater sensitivity compared with 2D MRI, 1.5T MRI, and <1.5 Tesla MRI.

LEVEL OF EVIDENCE

Level III, systematic review of Level II-III studies.

Accelerated Musculoskeletal Magnetic Resonance Imaging

With a substantial growth in the use of musculoskeletal MRI, there has been a growing need to improve MRI workflow, and faster imaging has been suggested as one of the solutions for a more efficient examination process. Consequently, there have been considerable advances in accelerated MRI scanning methods. This article aims to review the basic principles and applications of accelerated musculoskeletal MRI techniques including widely used conventional acceleration methods, more advanced deep learning-based techniques, and new approaches to reduce scan time. Specifically, conventional accelerated MRI techniques, including parallel imaging, compressed sensing, and simultaneous multislice imaging, and deep learning-based accelerated MRI techniques, including undersampled MR image reconstruction, super-resolution imaging, artifact correction, and generation of unacquired contrast images, are discussed. Finally, new approaches to reduce scan time, including synthetic MRI, novel sequences, and new coil setups and designs, are also reviewed. We believe that a deep understanding of these fast MRI techniques and proper use of combined acceleration methods will synergistically improve scan time and MRI workflow in daily practice. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 1.

Difference in quantitative MRI measurements of cartilage between Wiberg type III patella and stable patella based on a 3.0-T synthetic MRI sequence

Purpose

The purpose of this study was to investigate the difference between the quantitative MRI values of Wiberg type III and stable patellar cartilage, and to improve the accuracy of MRI quantification in early patellar cartilage damage.

Methods

The knee joints of 94 healthy volunteers were scanned by a GE Signa Pioneer 3.0-T synthetic MRI machine. According to the Wiberg classification, the patella was divided into types I-III. Types I-II made up the stable patella group, and type III made up the unstable patella group. Two radiologists independently measured patellar cartilage thickness and quantitative synthetic MRI values (T1, T2, PD) in both groups. Interobserver agreement for quantitative variables was assessed using the Bland-Altman method. A third radiologist assessed differences in measurements.

Results

The medial T2 and T1 value of Wiberg III patella did not show a normal distribution (all P > 0.05). Compared with the stable group, the Wiberg type III group had thinner cartilage of the medial surface of the patella (P < 0.05), lower cartilage T2 and PD values (P < 0.05), but a similar cartilage T1 value (P > 0.05). There was no significant difference in the cartilage thickness, T1, T2, or PD value of the lateral patella between the Wiberg type III and the stable group (P > 0.05).

Conclusion

There were certain differences in the cartilage thickness of the medial surface of the patella and the quantitative value of synthetic MRI in Wiberg type III patellas. Quantitative studies of patellar cartilage MRI measurements need to consider the influence of patellar morphology.

Physics-informed deep learning for T2-deblurred superresolution turbo spin echo MRI

PURPOSE

Deep learning superresolution (SR) is a promising approach to reduce MRI scan time without requiring custom sequences or iterative reconstruction. Previous deep learning SR approaches have generated low-resolution training images by simple k-space truncation, but this does not properly model in-plane turbo spin echo (TSE) MRI resolution degradation, which has variable T2 relaxation effects in different k-space regions. To fill this gap, we developed a T2 -deblurred deep learning SR method for the SR of 3D-TSE images.

METHODS

A SR generative adversarial network was trained using physically realistic resolution degradation (asymmetric T2 weighting of raw high-resolution k-space data). For comparison, we trained the same network structure on previous degradation models without TSE physics modeling. We tested all models for both retrospective and prospective SR with 3 × 3 acceleration factor (in the two phase-encoding directions) of genetically engineered mouse embryo model TSE-MR images.

RESULTS

The proposed method can produce high-quality 3 × 3 SR images for a typical 500-slice volume with 6-7 mouse embryos. Because 3 × 3 SR was performed, the image acquisition time can be reduced from 15 h to 1.7 h. Compared to previous SR methods without TSE modeling, the proposed method achieved the best quantitative imaging metrics for both retrospective and prospective evaluations and achieved the best imaging-quality expert scores for prospective evaluation.

CONCLUSION

The proposed T2 -deblurring method improved accuracy and image quality of deep learning-based SR of TSE MRI. This method has the potential to accelerate TSE image acquisition by a factor of up to 9.

[Formula: see text] Field inhomogeneity correction for qDESS [Formula: see text] mapping: application to rapid bilateral knee imaging

PURPOSE

[Formula: see text] mapping is a powerful tool for studying osteoarthritis (OA) changes and bilateral imaging may be useful in investigating the role of between-knee asymmetry in OA onset and progression. The quantitative double-echo in steady-state (qDESS) can provide fast simultaneous bilateral knee [Formula: see text] and high-resolution morphometry for cartilage and meniscus. The qDESS uses an analytical signal model to compute [Formula: see text] relaxometry maps, which require knowledge of the flip angle (FA). In the presence of [Formula: see text] inhomogeneities, inconsistencies between the nominal and actual FA can affect the accuracy of [Formula: see text] measurements. We propose a pixel-wise [Formula: see text] correction method for qDESS [Formula: see text] mapping exploiting an auxiliary [Formula: see text] map to compute the actual FA used in the model.

METHODS

The technique was validated in a phantom and in vivo with simultaneous bilateral knee imaging. [Formula: see text] measurements of femoral cartilage (FC) of both knees of six healthy participants were repeated longitudinally to investigate the association between [Formula: see text] variation and [Formula: see text].

RESULTS

The results showed that applying the [Formula: see text] correction mitigated [Formula: see text] variations that were driven by [Formula: see text] inhomogeneities. Specifically, [Formula: see text] left-right symmetry increased following the [Formula: see text] correction ([Formula: see text] = 0.74 > [Formula: see text] = 0.69). Without the [Formula: see text] correction, [Formula: see text] values showed a linear dependence with [Formula: see text]. The linear coefficient decreased using the [Formula: see text] correction (from 24.3 ± 1.6 ms to 4.1 ± 1.8) and the correlation was not statistically significant after the application of the Bonferroni correction (p value > 0.01).

CONCLUSION

The study showed that [Formula: see text] correction could mitigate variations driven by the sensitivity of the qDESS [Formula: see text] mapping method to [Formula: see text], therefore, increasing the sensitivity to detect real biological changes. The proposed method may improve the robustness of bilateral qDESS [Formula: see text] mapping, allowing for an accurate and more efficient evaluation of OA pathways and pathophysiology through longitudinal and cross-sectional studies.

Updates on Compositional MRI Mapping of the Cartilage: Emerging Techniques and Applications

Osteoarthritis (OA) is a widely occurring degenerative joint disease that is severely debilitating and causes significant socioeconomic burdens to society. Magnetic resonance imaging (MRI) is the preferred imaging modality for the morphological evaluation of cartilage due to its excellent soft tissue contrast and high spatial resolution. However, its utilization typically involves subjective qualitative assessment of cartilage. Compositional MRI, which refers to the quantitative characterization of cartilage using a variety of MRI methods, can provide important information regarding underlying compositional and ultrastructural changes that occur during early OA. Cartilage compositional MRI could serve as early imaging biomarkers for the objective evaluation of cartilage and help drive diagnostics, disease characterization, and response to novel therapies. This review will summarize current and ongoing state-of-the-art cartilage compositional MRI techniques and highlight emerging methods for cartilage compositional MRI including MR fingerprinting, compressed sensing, multiexponential relaxometry, improved and robust radio-frequency pulse sequences, and deep learning-based acquisition, reconstruction, and segmentation. The review will also briefly discuss the current challenges and future directions for adopting these emerging cartilage compositional MRI techniques for use in clinical practice and translational OA research studies. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 2.

Prediction of total knee replacement using deep learning analysis of knee MRI

Current methods for assessing knee osteoarthritis (OA) do not provide comprehensive information to make robust and accurate outcome predictions. Deep learning (DL) risk assessment models were developed to predict the progression of knee OA to total knee replacement (TKR) over a 108-month follow-up period using baseline knee MRI. Participants of our retrospective study consisted of 353 case-control pairs of subjects from the Osteoarthritis Initiative with and without TKR over a 108-month follow-up period matched according to age, sex, ethnicity, and body mass index. A traditional risk assessment model was created to predict TKR using baseline clinical risk factors. DL models were created to predict TKR using baseline knee radiographs and MRI. All DL models had significantly higher (p < 0.001) AUCs than the traditional model. The MRI and radiograph ensemble model and MRI ensemble model (where TKR risk predicted by several contrast-specific DL models were averaged to get the ensemble TKR risk prediction) had the highest AUCs of 0.90 (80% sensitivity and 85% specificity) and 0.89 (79% sensitivity and 86% specificity), respectively, which were significantly higher (p < 0.05) than the AUCs of the radiograph and multiple MRI models (where the DL models were trained to predict TKR risk using single contrast or 2 contrasts together as input). DL models using baseline MRI had a higher diagnostic performance for predicting TKR than a traditional model using baseline clinical risk factors and a DL model using baseline knee radiographs.

Improving Data-Efficiency and Robustness of Medical Imaging Segmentation Using Inpainting-Based Self-Supervised Learning

We systematically evaluate the training methodology and efficacy of two inpainting-based pretext tasks of context prediction and context restoration for medical image segmentation using self-supervised learning (SSL). Multiple versions of self-supervised U-Net models were trained to segment MRI and CT datasets, each using a different combination of design choices and pretext tasks to determine the effect of these design choices on segmentation performance. The optimal design choices were used to train SSL models that were then compared with baseline supervised models for computing clinically-relevant metrics in label-limited scenarios. We observed that SSL pretraining with context restoration using 32 × 32 patches and Poission-disc sampling, transferring only the pretrained encoder weights, and fine-tuning immediately with an initial learning rate of 1 × 10-3 provided the most benefit over supervised learning for MRI and CT tissue segmentation accuracy (p < 0.001). For both datasets and most label-limited scenarios, scaling the size of unlabeled pretraining data resulted in improved segmentation performance. SSL models pretrained with this amount of data outperformed baseline supervised models in the computation of clinically-relevant metrics, especially when the performance of supervised learning was low. Our results demonstrate that SSL pretraining using inpainting-based pretext tasks can help increase the robustness of models in label-limited scenarios and reduce worst-case errors that occur with supervised learning.

A method for measuring B0 field inhomogeneity using quantitative double-echo in steady-state

PURPOSE

To develop and validate a method forB 0 $$ {B}_0 $$ mapping for knee imaging using the quantitative Double-Echo in Steady-State (qDESS) exploiting the phase difference (Δ θ $$ \Delta \theta $$ ) between the two echoes acquired. Contrary to a two-gradient-echo (2-GRE) method,Δ θ $$ \Delta \theta $$ depends only on the first echo time.

METHODS

Bloch simulations were applied to investigate robustness to noise of the proposed methodology and all imaging studies were validated with phantoms and in vivo simultaneous bilateral knee acquisitions. Two phantoms and five healthy subjects were scanned using qDESS, water saturation shift referencing (WASSR), and multi-GRE sequences.Δ B 0 $$ \Delta {B}_0 $$ maps were calculated with the qDESS and the 2-GRE methods and compared against those obtained with WASSR. The comparison was quantitatively assessed exploiting pixel-wise difference maps, Bland-Altman (BA) analysis, and Lin's concordance coefficient (ρ c $$ {\rho}_c $$ ). For in vivo subjects, the comparison was assessed in cartilage using average values in six subregions.

RESULTS

The proposed method for measuringΔ B 0 $$ \Delta {B}_0 $$ inhomogeneities from a qDESS acquisition providedΔ B 0 $$ \Delta {B}_0 $$ maps that were in good agreement with those obtained using WASSR.Δ B 0 $$ \Delta {B}_0 $$ ρ c $$ {\rho}_c $$ values were≥ $$ \ge $$ 0.98 and 0.90 in phantoms and in vivo, respectively. The agreement between qDESS and WASSR was comparable to that of a 2-GRE method.

CONCLUSION

The proposed method may allow B0 correction for qDESST 2 $$ {T}_2 $$ mapping using an inherently co-registeredΔ B 0 $$ \Delta {B}_0 $$ map without requiring an additional B0 measurement sequence. More generally, the method may help shorten knee imaging protocols that require an auxiliaryΔ B 0 $$ \Delta {B}_0 $$ map by exploiting a qDESS acquisition that also providesT 2 $$ {T}_2 $$ measurements and high-quality morphological imaging.

Emerging Technology in Musculoskeletal MRI and CT

This article provides a focused overview of emerging technology in musculoskeletal MRI and CT. These technological advances have primarily focused on decreasing examination times, obtaining higher quality images, providing more convenient and economical imaging alternatives, and improving patient safety through lower radiation doses. New MRI acceleration methods using deep learning and novel reconstruction algorithms can reduce scanning times while maintaining high image quality. New synthetic techniques are now available that provide multiple tissue contrasts from a limited amount of MRI and CT data. Modern low-field-strength MRI scanners can provide a more convenient and economical imaging alternative in clinical practice, while clinical 7.0-T scanners have the potential to maximize image quality. Three-dimensional MRI curved planar reformation and cinematic rendering can provide improved methods for image representation. Photon-counting detector CT can provide lower radiation doses, higher spatial resolution, greater tissue contrast, and reduced noise in comparison with currently used energy-integrating detector CT scanners. Technological advances have also been made in challenging areas of musculoskeletal imaging, including MR neurography, imaging around metal, and dual-energy CT. While the preliminary results of these emerging technologies have been encouraging, whether they result in higher diagnostic performance requires further investigation.

Artificial Intelligence-Driven Ultra-Fast Superresolution MRI: 10-Fold Accelerated Musculoskeletal Turbo Spin Echo MRI Within Reach

ABSTRACT

Magnetic resonance imaging (MRI) is the keystone of modern musculoskeletal imaging; however, long pulse sequence acquisition times may restrict patient tolerability and access. Advances in MRI scanners, coil technology, and innovative pulse sequence acceleration methods enable 4-fold turbo spin echo pulse sequence acceleration in clinical practice; however, at this speed, conventional image reconstruction approaches the signal-to-noise limits of temporal, spatial, and contrast resolution. Novel deep learning image reconstruction methods can minimize signal-to-noise interdependencies to better advantage than conventional image reconstruction, leading to unparalleled gains in image speed and quality when combined with parallel imaging and simultaneous multislice acquisition. The enormous potential of deep learning-based image reconstruction promises to facilitate the 10-fold acceleration of the turbo spin echo pulse sequence, equating to a total acquisition time of 2-3 minutes for entire MRI examinations of joints without sacrificing spatial resolution or image quality. Current investigations aim for a better understanding of stability and failure modes of image reconstruction networks, validation of network reconstruction performance with external data sets, determination of diagnostic performances with independent reference standards, establishing generalizability to other centers, scanners, field strengths, coils, and anatomy, and building publicly available benchmark data sets to compare methods and foster innovation and collaboration between the clinical and image processing community. In this article, we review basic concepts of deep learning-based acquisition and image reconstruction techniques for accelerating and improving the quality of musculoskeletal MRI, commercially available and developing deep learning-based MRI solutions, superresolution, denoising, generative adversarial networks, and combined strategies for deep learning-driven ultra-fast superresolution musculoskeletal MRI. This article aims to equip radiologists and imaging scientists with the necessary practical knowledge and enthusiasm to meet this exciting new era of musculoskeletal MRI.

Artificial intelligence and machine learning in cancer imaging

An increasing array of tools is being developed using artificial intelligence (AI) and machine learning (ML) for cancer imaging. The development of an optimal tool requires multidisciplinary engagement to ensure that the appropriate use case is met, as well as to undertake robust development and testing prior to its adoption into healthcare systems. This multidisciplinary review highlights key developments in the field. We discuss the challenges and opportunities of AI and ML in cancer imaging; considerations for the development of algorithms into tools that can be widely used and disseminated; and the development of the ecosystem needed to promote growth of AI and ML in cancer imaging.

Deep learning-based quantitative estimation of lymphedema-induced fibrosis using three-dimensional computed tomography images

In lymphedema, proinflammatory cytokine-mediated progressive cascades always occur, leading to macroscopic fibrosis. However, no methods are practically available for measuring lymphedema-induced fibrosis before its deterioration. Technically, CT can visualize fibrosis in superficial and deep locations. For standardized measurement, verification of deep learning (DL)-based recognition was performed. A cross-sectional, observational cohort trial was conducted. After narrowing window width of the absorptive values in CT images, SegNet-based semantic segmentation model of every pixel into 5 classes (air, skin, muscle/water, fat, and fibrosis) was trained (65%), validated (15%), and tested (20%). Then, 4 indices were formulated and compared with the standardized circumference difference ratio (SCDR) and bioelectrical impedance (BEI) results. In total, 2138 CT images of 27 chronic unilateral lymphedema patients were analyzed. Regarding fibrosis segmentation, the mean boundary F1 score and accuracy were 0.868 and 0.776, respectively. Among 19 subindices of the 4 indices, 73.7% were correlated with the BEI (partial correlation coefficient: 0.420–0.875), and 13.2% were correlated with the SCDR (0.406–0.460). The mean subindex of Index 2 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {\frac{{P_{Fibrosis\, in\, Affected} - P_{Fibrosis\, in\, Unaffected} }}{{P_{Limb\, in\, Unaffected} }}} \right)$$\end{document}PFibrosisinAffected-PFibrosisinUnaffectedPLimbinUnaffected presented the highest correlation. DL has potential applications in CT image-based lymphedema-induced fibrosis recognition. The subtraction-type formula might be the most promising estimation method.

Detection Method of Athlete Joint Injury Based on Deep Learning Model

The research on accurate and intelligent segmentation of knee joint MRI images is of great significance to reduce the work intensity of clinical doctors and nurses. In order to solve the problem that knee joint MRI image segmentation model needs a large number of high-quality tagged images and excessive labeling workload, a semisupervised learning segmentation network model based on 3D scSE-UNet is proposed. The model adopts a self-training semisupervised learning framework and adds a cSE-block+ module on the basis of the 3D UNet model. This module can enhance the effective features of the feature image from two aspects of space and channel, while suppressing irrelevant features and preserving image edge information more completely. In order to solve the problem of rough edge of pseudolabel caused by model segmentation, a fully connected conditional random field is added to refine the edge of pseudolabel in the process of model training. The effectiveness of the model is verified by open source MRNet dataset and OAI dataset. The results show that the proposed model can achieve the segmentation effect of fully supervised learning through a small number of labeled images and effectively reduce the dependence of knee joint MRI image segmentation on expert labeling data.

Identification and diagnosis of meniscus tear by magnetic resonance imaging using a deep learning model

Objective

Meniscus tear is a common problem in sports trauma, and its imaging diagnosis mainly relies on MRI. To improve the diagnostic accuracy and efficiency, a deep learning model was employed in this study and the identification efficiency was evaluated.

Methods

Standard knee MRI images from 924 individual patients were used to complete the training, validation and testing processes. Mask regional convolutional neural network (R–CNN) was used to build the deep learning network structure, and ResNet50 was adopted to develop the backbone network. The deep learning model was trained and validated with a dataset containing 504 and 220 patients, respectively. Internal testing was performed based on a dataset of 200 patients, and 180 patients from 8 hospitals were regarded as an external dataset for model validation. Additionally, 40 patients who were diagnosed by the arthroscopic surgery were enrolled as the final test dataset.

Results

After training and validation, the deep learning model effectively recognized healthy and injured menisci. Average precision for the three types of menisci (healthy, torn and degenerated menisci) ranged from 68% to 80%. Diagnostic accuracy for healthy, torn and degenerated menisci was 87.50%, 86.96%, and 84.78%, respectively. Validation results from external dataset demonstrated that the accuracy of diagnosing torn and intact meniscus tear through 3.0T MRI images was higher than 80%, while the accuracy verified by arthroscopic surgery was 87.50%.

Conclusion

Mask R–CNN effectively identified and diagnosed meniscal injuries, especially for tears that occurred in different parts of the meniscus. The recognition ability was admirable, and the diagnostic accuracy could be further improved with increased training sample size. Therefore, this deep learning model showed great potential in diagnosing meniscus injuries.

Translational potential of this article

Deep learning model exerted unique effect in terms of reducing doctors’ workload and improving diagnostic accuracy. Injured and healthy menisci could be more accurately identified and classified based on training and learning datasets. This model could also distinguish torn from degenerated menisci, making it an effective tool for MRI-assisted diagnosis of meniscus injuries in clinical practice.

Upstream Machine Learning in Radiology

Machine learning (ML) and Artificial intelligence (AI) has the potential to dramatically improve radiology practice at multiple stages of the imaging pipeline. Most of the attention has been garnered by applications focused on improving the end of the pipeline: image interpretation. However, this article reviews how AI/ML can be applied to improve upstream components of the imaging pipeline, including exam modality selection, hardware design, exam protocol selection, data acquisition, image reconstruction, and image processing. A breadth of applications and their potential for impact is shown across multiple imaging modalities, including ultrasound, computed tomography, and MRI.

Image Quality and Diagnostic Performance of Accelerated Shoulder MRI With Deep Learning-Based Reconstruction

Background: Shoulder MRI using standard multiplanar sequences requires long scan times. Accelerated sequences have tradeoffs in noise and resolution. Deep learning-based reconstruction (DLR) may allow reduced scan time with preserved image quality. Objectives: To compare standard shoulder MRI sequences and accelerated sequences without and with DLR in terms of image quality and diagnostic performance. Methods: This retrospective study included 105 patients (45 men, 60 women; mean age 57.6±10.9 years) who underwent a total of 110 3-T shoulder MRI examinations. Examinations included standard sequences (scan time, 9 minutes 23 seconds) and accelerated sequences (3 minutes 5 seconds; 67% reduction), both including fast spin echo sequences in three planes. Standard sequences were reconstructed using the conventional pipeline; accelerated sequences were reconstructed using both conventional pipeline and a commercially available DLR pipeline. Two radiologists independently assessed three image sets (standard, accelerated without DLR, accelerated with DLR) for subjective image quality and artifacts using 4-point scales (4=highest quality), and identified pathologies of subscapularis tendon, supraspinatus-infraspinatus tendon, biceps brachii long head tendon, and glenoid labrum. Interobserver and inter-image set agreement for the evaluated pathologies was assessed using weighted kappa statistics. In 27 patients who underwent arthroscopy, diagnostic performance was calculated using arthroscopic findings as reference. Results: Mean subjective image quality for readers 1 and 2 was 10.6±1.2 and 10.5±1.4 for standard, 8.1±1.3 and 7.2±1.1 for accelerated without DLR, and 10.7±1.2 and 10.5±1.6 for accelerated with DLR. Mean artifact score for readers 1 and 2 was 9.3±1.2 and 10.0±1.0 for standard, 7.3±1.3 and 9.1±0.8 for accelerated without DLR, and 9.4±1.2 and 9.8±0.8 for accelerated with DLR. Interobserver agreement ranged from kappa=0.813-0.951 except for accelerated without DLR for SST-IST (κ=0.673). Inter-image set agreement ranged from kappa=0.809-0.957 except for reader 1 for SST-IST (κ=0.663-0.700). Accuracy, sensitivity, and specificity for tears of the four structures was not different (p>.05) among image sets. Conclusions: Accelerated sequences with DLR provide 67% scan time reduction with similar subjective image quality, artifacts, and diagnostic performance as standard sequences. Clinical impact: Accelerated sequences with DLR may provide an alternative to standard sequences for clinical shoulder MRI.

Low-count whole-body PET with deep learning in a multicenter and externally validated study

More widespread use of positron emission tomography (PET) imaging is limited by its high cost and radiation dose. Reductions in PET scan time or radiotracer dosage typically degrade diagnostic image quality (DIQ). Deep-learning-based reconstruction may improve DIQ, but such methods have not been clinically evaluated in a realistic multicenter, multivendor environment. In this study, we evaluated the performance and generalizability of a deep-learning-based image-quality enhancement algorithm applied to fourfold reduced-count whole-body PET in a realistic clinical oncologic imaging environment with multiple blinded readers, institutions, and scanner types. We demonstrate that the low-count-enhanced scans were noninferior to the standard scans in DIQ (p < 0.05) and overall diagnostic confidence (p < 0.001) independent of the underlying PET scanner used. Lesion detection for the low-count-enhanced scans had a high patient-level sensitivity of 0.94 (0.83-0.99) and specificity of 0.98 (0.95-0.99). Interscan kappa agreement of 0.85 was comparable to intrareader (0.88) and pairwise inter-reader agreements (maximum of 0.72). SUV quantification was comparable in the reference regions and lesions (lowest p-value=0.59) and had high correlation (lowest CCC = 0.94). Thus, we demonstrated that deep learning can be used to restore diagnostic image quality and maintain SUV accuracy for fourfold reduced-count PET scans, with interscan variations in lesion depiction, lower than intra- and interreader variations. This method generalized to an external validation set of clinical patients from multiple institutions and scanner types. Overall, this method may enable either dose or exam-duration reduction, increasing safety and lowering the cost of PET imaging.

Prospective Deployment of Deep Learning in MRI: A Framework for Important Considerations, Challenges, and Recommendations for Best Practices

Artificial intelligence algorithms based on principles of deep learning (DL) have made a large impact on the acquisition, reconstruction, and interpretation of MRI data. Despite the large number of retrospective studies using DL, there are fewer applications of DL in the clinic on a routine basis. To address this large translational gap, we review the recent publications to determine three major use cases that DL can have in MRI, namely, that of model-free image synthesis, model-based image reconstruction, and image or pixel-level classification. For each of these three areas, we provide a framework for important considerations that consist of appropriate model training paradigms, evaluation of model robustness, downstream clinical utility, opportunities for future advances, as well recommendations for best current practices. We draw inspiration for this framework from advances in computer vision in natural imaging as well as additional healthcare fields. We further emphasize the need for reproducibility of research studies through the sharing of datasets and software. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY STAGE: 2.

Imaging of Synovial Inflammation in Osteoarthritis, From the AJR Special Series on Inflammation

Synovitis, inflammation of the synovial membrane, is a common manifestation in osteoarthritis (OA) and is recognized to play a role in the complex pathophysiology of OA. Increased recognition of the importance of synovitis in the OA disease process and potential as a target for treatment has increased the need for non-invasive detection and characterization of synovitis using medical imaging. Numerous imaging methods can assess synovitis involvement in OA with varying sensitivity and specificity as well as complexity. This article reviews the role of contrast-enhanced MRI, conventional MRI, novel unenhanced MRI, gray-scale ultrasound (US), and power Doppler US in the assessment of synovitis in patients with OA. The role of imaging in disease evaluation as well as challenges in conventional imaging methods are discussed. We also provide an overview into the potential utility of emerging techniques for imaging of early inflammation and molecular inflammatory markers of synovitis, including quantitative MRI, superb microvascular imaging, and PET. The potential development of therapeutic treatments targeting inflammatory features, particularly in early OA, would greatly increase the importance of these imaging methods for clinical decision making and evaluation of therapeutic efficacy.

Non-contrast MRI of synovitis in the knee using quantitative DESS

OBJECTIVES

To determine whether synovitis graded by radiologists using hybrid quantitative double-echo in steady-state (qDESS) images can be utilized as a non-contrast approach to assess synovitis in the knee, compared against the reference standard of contrast-enhanced MRI (CE-MRI).

METHODS

Twenty-two knees (11 subjects) with moderate to severe osteoarthritis (OA) were scanned using CE-MRI, qDESS with a high diffusion weighting (qDESS_High), and qDESS with a low diffusion weighting (qDESS_Low). Four radiologists graded the overall impression of synovitis, their diagnostic confidence, and regional grading of synovitis severity at four sites (suprapatellar pouch, intercondylar notch, and medial and lateral peripatellar recesses) in the knee using a 4-point scale. Agreement between CE-MRI and qDESS, inter-rater agreement, and intra-rater agreement were assessed using a linearly weighted Gwet's AC2.

RESULTS

Good agreement was seen between CE-MRI and both qDESS_Low (AC2 = 0.74) and qDESS_High (AC2 = 0.66) for the overall impression of synovitis, but both qDESS sequences tended to underestimate the severity of synovitis compared to CE-MRI. Good inter-rater agreement was seen for both qDESS sequences (AC2 = 0.74 for qDESS_Low, AC2 = 0.64 for qDESS_High), and good intra-rater agreement was seen for both sequences as well (qDESS_Low AC2 = 0.78, qDESS_High AC2 = 0.80). Diagnostic confidence was moderate to high for qDESS_Low (mean = 2.36) and slightly less than moderate for qDESS_High (mean = 1.86), compared to mostly high confidence for CE-MRI (mean = 2.73).

CONCLUSIONS

qDESS shows potential as an alternative MRI technique for assessing the severity of synovitis without the use of a gadolinium-based contrast agent.

KEY POINTS

The use of the quantitative double-echo in steady-state (qDESS) sequence for synovitis assessment does not require the use of a gadolinium-based contrast agent. Preliminary results found that low diffusion-weighted qDESS (qDESS_Low) shows good agreement to contrast-enhanced MRI for characterization of the severity of synovitis, with a relative bias towards underestimation of severity. Preliminary results also found that qDESS_Low shows good inter- and intra-rater agreement for the depiction of synovitis, particularly for readers experienced with the sequence.

MRSaiFE: An AI-based Approach Towards the Real-Time Prediction of Specific Absorption Rate

The purpose of this study is to investigate feasibility of estimating the specific absorption rate (SAR) in MRI in real time. To this goal, SAR maps are predicted from 3T- and 7T-simulated magnetic resonance (MR) images in 10 realistic human body models via a convolutional neural network. Two-dimensional (2-D) U-Net architectures with varying contraction layers and different convolutional filters were designed to estimate the SAR distribution in realistic body models. Sim4Life (ZMT, Switzerland) was used to create simulated anatomical images and SAR maps at 3T and 7T imaging frequencies for Duke, Ella, Charlie, and Pregnant Women (at 3, 7, and 9 month gestational stages) body models. Mean squared error (MSE) was used as the cost function and the structural similarity index (SSIM) was reported. A 2-D U-Net with 4 contracting (and 4 expanding) layers and 64 convolutional filters at the initial stage showed the best compromise to estimate SAR distributions. Adam optimizer outperformed stochastic gradient descent (SGD) for all cases with an average SSIM of 90.5∓3.6 % and an average MSE of 0.7∓0.6% for head images at 7T, and an SSIM of >85.1∓6.2 % and an MSE of 0.4∓0.4% for 3T body imaging. Algorithms estimated the SAR maps for 224×224 slices under 30 ms. The proposed methodology shows promise to predict real-time SAR in clinical imaging settings without using extra mapping techniques or patient-specific calibrations.