Magnetic resonance imaging of the knee: An overview and update of conventional and state of the art imaging

Ex vivo evaluation of the soft tissue components of the equine stifle using 3 Tesla magnetic resonance imaging under flexion, extension, and loading

High-field MRI of the equine stifle provides high-resolution information about soft tissues that is useful in the diagnosis of stifle lameness. The aim of this prospective anatomic study was to describe the appearance, position, size, and shape of the equine femorotibial ligaments, meniscal ligaments, and menisci using 3 Tesla MRI under extended, extended-loaded, and flexed conditions. Additionally, histologic examination of the collateral and cruciate ligaments (CLs) of a single stifle was performed to compare with MRI images. In extension, mild variations in MRI signal intensity were apparent in the CLs, and the cranial had two distinct longitudinal regions indicating two ligament bundles. Flexion had minor effects on CL signal intensity and altered the tibial angles of attachment. Histology indicated that both CLs were comprised of two fiber bundles. The collateral ligaments were the same low-signal intensity. The medial collateral ligament had a smaller cross-sectional area than the lateral, and flexion increased the length of the medial collateral ligament and the cross-sectional area of the lateral. Low loads in extension did not affect the MRI appearance of stifle soft tissues. Flexion of the stifle impacted cruciate ligament insertion angles and the size and shape of collateral ligaments. This study provides support for the use of MRI to understand the anatomy and function of stifle ligaments.

Variations in knee cartilage thickness: Fully automatic three-dimensional analysis of MRIs from five manufacturers

BACKGROUND

Measurements of knee cartilage thickness derived from MR images are attractive biomarkers for osteoarthritis research. Although some cross-sectional multivendor studies exist, none have employed fully automatic three-dimensional MRI analysis. Our objective was to evaluate the variations in knee cartilage thickness measurements obtained using automated methods and MRI instruments from five different vendors.

METHODS

The subjects were 10 healthy volunteers aged 22-60 years. MRI models with 3 Tesla strength from five different companies were used. Cartilage thickness was quantified fully automatically for seven regions. We hypothesized that "the MRI model influences cartilage thickness measurements." Inter-measurement error, defined as the absolute difference between the targeted and median thicknesses determined by the five MRI models, was analyzed using histograms. The factors generating the largest inter-measurement error were also examined.

RESULTS

No exceptional trends attributable to a specific instrument model were observed, and the p-value from the Kruskal-Wallis test exceeded 0.05 in all seven regions. Therefore, the study hypothesis was rejected. Of the 350 measurements, the inter-measurement error was ≤0.05 mm in 53 %, ≤0.10 mm in 75 %, and ≤0.20 mm in 95 %. Analysis of the medial tibial cartilage, which had the largest inter-measurement error, revealed mis-extraction of synovial fluid as cartilage.

CONCLUSIONS

The choice of MRI model did not influence cartilage thickness measurements. Overall, 95 % of the inter-measurement errors were within 0.20 mm. The greatest error resulted from mis-extracting synovial fluid as cartilage.

Diagnosing popliteofibular ligament injuries in anterior cruciate ligament-injured knees: A prospective magnetic resonance imaging study investigating the inter- and intraobserver reliability of identification of the popliteofibular ligament

Purpose

The aim of our study was to investigate the intra- and interobserver reliability for the identification of the popliteofibular ligament (PFL) in magnetic resonance imaging (MRI) scans in patients with an anterior cruciate ligament (ACL) injury and ascertain the prevalence of PFL tears in ACL-injured knees without clinically high-grade posterolateral corner injury.

Methods

MRI readings were performed retrospectively by two surgeons on 84 patients who underwent ACL reconstruction in our department. The presence of the PFL on both sagittal and coronal images as well as the presence of PFL tears was noted. Readings were repeated 6 weeks later for one observer. The κ value was calculated to determine the intra- and interobserver reliability for identification of the PFL and the prevalence of PFL tears was ascertained.

Results

The PFL was visualized in 90.5%-91.7% of MRI scans. The intra- and interobserver reliability of visualizing the PFL on MRI had an κ value of 0.63 and 0.66 (substantially reliable), respectively. The intraobserver reliability for identification of PFL tears had an κ value of 0.26 (fair reliability). We found a 4.8% prevalence of PFL tears in ACL-injured knees.

Conclusions

There is substantially reliable intra- and interobserver reliability for the identification of the PFL on MRI scans but only fair reliability for the identification of PFL tears. A 4.8% prevalence of PFL tears in ACL-injured knees without clinically confirmed high-grade posterolateral corner injury can be observed in our series.

Level of Evidence

Level IV.

Evaluating the Diagnostic Performance of MRI for Identification of Meniscal Ramp Lesions in ACL-Deficient Knees: A Systematic Review and Meta-Analysis

BACKGROUND

Despite vigorous efforts to delineate the efficacy of magnetic resonance imaging (MRI) for the diagnosis of meniscal ramp lesions, there is still a great deal of uncertainty regarding its diagnostic performance. Therefore, we conducted a systematic review and meta-analysis to investigate the diagnostic performance of MRI for detecting ramp lesions in anterior cruciate ligament (ACL)-deficient knees.

METHODS

We performed a systematic search of MEDLINE via PubMed, Scopus, Web of Science, and Embase and included all articles, published before October 20, 2022, comparing the accuracy of MRI with that of arthroscopy as the gold standard for diagnosis of ramp lesions. We performed statistical analysis using Stata and Meta-DiSc software. Quality assessment of the included studies was performed using the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2) tool.

RESULTS

This meta-analysis evaluated 21 diagnostic performance comparisons from 19 original research articles (2,149 patients). The pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and area under the summary receiver operating characteristic (SROC) curve for diagnosing a ramp lesion were 0.70 (95% confidence interval [Cl], 0.66 to 0.73), 0.88 (95% Cl, 0.86 to 0.89), 6.49 (95% Cl, 4.12 to 10.24), 0.36 (95% Cl, 0.28 to 0.46), 24.33 (95% Cl, 12.81 to 46.19), and 0.88, respectively. Meta-regression using different variables yielded the same results.

CONCLUSIONS

MRI exhibited a DOR of 24.33 and moderate sensitivity, specificity, and accuracy for diagnosing ramp lesions in ACL-deficient knees. However, arthroscopy using a standard anterolateral portal with intercondylar viewing is recommended to confirm a diagnosis of a ramp lesion.

LEVEL OF EVIDENCE

Diagnostic Level III . See Instructions for Authors for a complete description of levels of evidence.

High-resolution magnetic resonance vessel wall imaging in ischemic stroke and carotid artery atherosclerotic stenosis: A review

Ischemic stroke is a significant burden on human health worldwide. Carotid Atherosclerosis stenosis plays an important role in the comprehensive assessment and prevention of ischemic stroke patients. High-resolution vessel wall magnetic resonance imaging has emerged as a successful technique for assessing carotid atherosclerosis stenosis. This advanced imaging modality has shown promise in effectively displaying a wide range of characteristics associated with the condition, leading to a comprehensive evaluation. High-resolution vessel wall magnetic resonance imaging not only enables a comprehensive evaluation of the instability of carotid atherosclerosis stenosis plaques but also provides valuable information for understanding the pathogenesis and predicting the prognosis of ischemic stroke patients. The purpose of this article is to review the application of high-resolution magnetic resonance imaging in ischemic stroke and carotid atherosclerotic stenosis.

Three-Dimensional Orientation of the Native Anterior Cruciate Ligament in Magnetic Resonance Imaging

The aim of this study was to describe the three-dimensional orientation of the native anterior cruciate ligament (ACL) in magnetic resonance imaging (MRI) by calculating the angles of inclination in relation to the axial plane which is given for the knee joint line in a group of healthy individuals. These could help to establish guidelines that may be used for the surgical positioning of bone tunnels during ACL reconstruction. A total of 290 MRI scans of patients with integrity of the ACL were evaluated; three observers identified the coordinates of the femoral and tibial insertion sites, then it is defined the vector and evaluated its angles with respect to axial axis and calculated the angles with trigonometric equations. The data were analyzed according to the age, sex, side, BMI, and height of the patients, and the interobserver reliability was calculated. The patient demographics were as follows: age average: 45 years old, BMI average: 27.1, 54% right knees, and 60% female. The average angle for all the measurements was 76.95 degrees (SD ± 6.8 degrees) in the sagittal plane, 81.65 degrees (SD ± 7.79 degrees) in the coronal plane and 33.17 degrees (SD ± 4.98 degrees) in the axial plane. No statistically significant differences were found between the categorical variables mentioned; moderate to substantial interobserver reliability strength was found with an average kappa of 0.791 for all measurements. The three-dimensional orientation of the native ACL in a group of healthy individuals was established. The findings can be helpful for performing anatomical reconstructions of the ACL in injured patients using as reference the average calculated angles, or measure of the contralateral non-injured knee for surgical planning; these results serve as a basis for the design of a technique that optimizes the three-dimensional position of the ACL when it undergoes reconstruction in the trend toward greater precision for better functional results. Biomechanical and clinical-surgical studies are required to further evaluate our results.

Evaluation of the knee joint with ultrasound and magnetic resonance imaging

The knee joint relies on a combination of deep and superficial structures for stability and function. Both ultrasound and high-resolution magnetic resonance imaging are extremely useful in evaluating these structures and associated pathology. This article reviews a combination of critical anatomic structures, joint abnormalities, and pathologic conditions at the knee joint, while highlighting the merits, limitations, and pitfalls of the two imaging modalities. A clear appreciation of each method paired with its relative strengths will aid in expediting diagnosis and appropriate treatment for a wide range of knee joint conditions.

Cruciate and Collateral Ligaments: 2-Dimensional and 3-Dimensional MR Imaging-Aid to Knee Preservation Surgery

Being a major load-bearing and dynamic functional joint of the body, the knee joint is prone to injuries to the menisci and ligaments. Injury to one or more of these structures leads to accelerated or premature osteoarthritis and chronic debilitation. The estimated cost of these injuries in terms of diagnosis, treatment, rehabilitation, and person-hours lost amounts to over a billion dollars annually. Understanding the relevant anatomy of knee cruciate and collateral ligaments, their biomechanical functional role, and imaging considerations are essential to assist multi-disciplinary communications and timely patient treatments for improved management. MRI is considered the gold standard over other imaging modalities like X-ray, ultrasound (US), and Computed Tomography (CT) for diagnosing and assessing such injuries. High-resolution 2D and 3D MRI are considerably helpful in planning for knee preservation surgery, and this article reviews the role of such imaging as a valuable aid to knee preservation surgery in pre-and post-operative states. Relevant arthroscopy correlations have been illustrated.

Acute Repair of Meniscus Root Tear Partially Restores Joint Displacements as Measured With Magnetic Resonance Images and Loading in a Cadaveric Porcine Knee

The meniscus serves important load-bearing functions and protects the underlying articular cartilage. Unfortunately, meniscus tears are common and impair the ability of the meniscus to distribute loads, increasing the risk of developing osteoarthritis. Therefore, surgical repair of the meniscus is a frequently performed procedure; however, repair does not always prevent osteoarthritis. This is hypothesized to be due to altered joint loading post-injury and repair, where the functional deficit of the meniscus prevents it from performing its role of distributing forces. The objective of this study was to quantify joint kinematics in an intact joint, after a meniscus root tear, and after suture repair in cadaveric porcine knees, a frequently used in vivo model. We utilized an magnetic resonance images-compatible loading device and novel use of a T1 vibe sequence to measure meniscus and femur displacements under physiological axial loads. We found that anterior root tear led to large meniscus displacements under physiological axial loading and that suture anchor repair reduced these displacements but did not fully restore intact joint kinematics. After tear and repair, the anterior region of the meniscus moved posteriorly and medially as it was forced out of the joint space under loading, while the posterior region had small displacements as the posterior attachment acted as a hinge about which the meniscus pivoted in the axial plane. Methods from this study can be applied to assess altered joint kinematics following human knee injuries and evaluate repair strategies aimed to restore joint kinematics.

Deep Learning Convolutional Neural Network Reconstruction and Radial k-Space Acquisition MR Technique for Enhanced Detection of Retropatellar Cartilage Lesions of the Knee Joint

OBJECTIVES

To assess diagnostic performance of standard radial k-space (PROPELLER) MRI sequences and compare with accelerated acquisitions combined with a deep learning-based convolutional neural network (DL-CNN) reconstruction for evaluation of the knee joint.

METHODS

Thirty-five patients undergoing MR imaging of the knee at 1.5 T were prospectively included. Two readers evaluated image quality and diagnostic confidence of standard and DL-CNN accelerated PROPELLER MR sequences using a four-point Likert scale. Pathological findings of bone, cartilage, cruciate and collateral ligaments, menisci, and joint space were analyzed. Inter-reader agreement (IRA) for image quality and diagnostic confidence was assessed using intraclass coefficients (ICC). Cohen's Kappa method was used for evaluation of IRA and consensus between sequences in assessing different structures. In addition, image quality was quantitatively evaluated by signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) measurements.

RESULTS

Mean acquisition time of standard vs. DL-CNN sequences was 10 min 3 s vs. 4 min 45 s. DL-CNN sequences showed significantly superior image quality and diagnostic confidence compared to standard MR sequences. There was moderate and good IRA for assessment of image quality in standard and DL-CNN sequences with ICC of 0.524 and 0.830, respectively. Pathological findings of the knee joint could be equally well detected in both sequences (κ-value of 0.8). Retropatellar cartilage could be significantly better assessed on DL-CNN sequences. SNR and CNR was significantly higher for DL-CNN sequences (both p < 0.05).

CONCLUSIONS

In MR imaging of the knee, DL-CNN sequences showed significantly higher image quality and diagnostic confidence compared to standard PROPELLER sequences, while reducing acquisition time substantially. Both sequences perform comparably in the detection of knee-joint pathologies, while DL-CNN sequences are superior for evaluation of retropatellar cartilage lesions.

Commercially Available Deep-learning-reconstruction of MR Imaging of the Knee at 1.5T Has Higher Image Quality Than Conventionally-reconstructed Imaging at 3T: A Normal Volunteer Study

PURPOSE

This study aimed to evaluate whether the image quality of 1.5T magnetic resonance imaging (MRI) of the knee is equal to or higher than that of 3T MRI by applying deep learning reconstruction (DLR).

METHODS

Proton density-weighted images of the right knee of 27 healthy volunteers were obtained by 3T and 1.5T MRI scanners using similar imaging parameters (21 for high resolution image and 6 for normal resolution image). Commercially available DLR was applied to the 1.5T images to obtain 1.5T/DLR images. The 3T and 1.5T/DLR images were compared subjectively for visibility of structures, image noise, artifacts, and overall diagnostic acceptability and objectively. One-way ANOVA and Friedman tests were used for the statistical analyses.

RESULTS

For the high resolution images, all of the anatomical structures, except for bone, were depicted significantly better on the 1.5T/DLR compared with 3T images. Image noise scored statistically lower and overall diagnostic acceptability scored higher on the 1.5T/DLR images. The contrast between lateral meniscus and articular cartilage of the 1.5T/DLR images was significantly higher (5.89 ± 1.30 vs. 4.34 ± 0.87, P < 0.001), and also the contrast between medial meniscus and articular cartilage of the 1.5T/DLR images was significantly higher (5.12 ± 0.93 vs. 3.87 ± 0.56, P < 0.001). Similar image quality improvement by DLR was observed for the normal resolution images.

CONCLUSION

The 1.5T/DLR images can achieve less noise, more precise visualization of the meniscus and ligaments, and higher overall image quality compared with the 3T images acquired using a similar protocol.

Preoperative diagnosis of knee cartilage, meniscal, and ligament injuries by magnetic resonance imaging

PURPOSE

The purpose of the study was to report on the current accuracy measures specific to 1.5-Tesla MRI of the knee in the patient population prone to injuries of the anterior cruciate ligament (ACL), the menisci, and the articular cartilage.

METHODS

We accrued patients between January 2018 through August 2021 who underwent a preoperative MRI and were diagnosed with an articular cartilage injury either due to unevenness of articular cartilage in T2-weighted sequences or due to the irregularity of subchondral bone in T1-weighted sequences. All patients were treated arthroscopically. Sensitivity, specificity, and accuracy were calculated for the detection of ACL, meniscus, and cartilage injuries. A P-value of < 0.05 represented statistical significance.

RESULTS

One-hundred and forty-seven cases which included 150 knee joints were enrolled in this study. The mean age at the time of surgery was 42.9 years-old. The sensitivity in the diagnosis of ACL injuries was significantly greater than that in the diagnosis of cartilage injuries (P = 0.0083). The ratios of the equality of operative indication in 6 recipient sites were found to be between 90.0% and 96.0%. The diagnostic critical point was within a 1 cm in diameter.

CONCLUSION

The diagnostic sensitivity in cartilage injuries was significantly lower than ones of ACL and meniscal injuries. The ratios of the equality of operative indication was determined to be between 90.0% and 96.0%, if we consider the unevenness of articular cartilage or the irregularity of subchondral bone.

LEVEL OF EVIDENCE

Level III, Prospective diagnostic cohort study.

Particularities on Anatomy and Normal Postsurgical Appearances of the Knee

Detailed knowledge of anatomy helps to understand pathologic processes. This article focuses on the anatomy and functionality of the knee, with emphasis on recently studied concepts and anatomic features that have an association with the development of pathology. The most common anatomic variants posing a challenge for diagnosis and other common findings in asymptomatic patients are reviewed. Good understanding of the different surgical procedures helps in providing as much information as possible to guarantee a positive outcome, improving prognosis. We review what are the commonly expected postsurgical appearances and the most common postsurgical complications.

Increased signal of the fibular collateral ligament of the knee on MRI, clinically significant?

INTRODUCTION

The purpose of this study was to determine the clinical significance of signal hyperintensity in the proximal fibular collateral ligament (FCL) on coronal proton density (PD) fat-saturated (FS) MRI of the knee, a common finding. This study is unique in that it characterizes the FCL of a comprehensive, large cohort of both symptomatic and asymptomatic patients, which to our knowledge represents the first study with such broad inclusion criteria.

METHODS

A large case series was performed analyzing MRI of the knee of 250 patients from July 2021 through September 2021 and retrospectively reviewed. All studies were performed on 3-Tesla MRI scanners with a dedicated knee coil and in accordance with standard institutional knee MRI protocol. Signal in the proximal fibular collateral ligament was assessed on coronal PDFS and axial T2-weighted FS images. Increased signal was classified as none, mild, moderate, or severe. A corresponding chart review of clinic notes was performed to determine the presence or absence of lateral knee pain. An FCL sprain or injury was considered present if the medical record described tenderness on palpation of the lateral knee, positive finding against resistance to the leg (varus stress test) or reverse pivot shift, or any clinical suspicion for lateral complex sprain or posterolateral corner injury.

RESULTS

The majority (74%) of knee MRIs demonstrated the presence of increased signal in the proximal fibular collateral ligament on coronal PD FS images. <5% of these patients had associated clinical findings of fibular collateral ligament and/or lateral supporting structure injury.

DISCUSSION

Although increased signal in the proximal FCL of the knee is a common finding on coronal PDFS images, the majority are not associated with clinical symptoms. Thus, this increased signal is likely not a pathological finding in the absence of clinical findings of fibular collateral ligament sprain/injury. Our study emphasizes the importance of clinical correlation in identifying increased signal in the proximal FCL as pathologic.

Positional MR imaging of normal and injured knees

OBJECTIVES

This study uses a practical positional MRI protocol to evaluate tibiofemoral translation and rotation in normal and injured knees.

METHODS

Following ethics approval, positional knee MRI of both knees was performed at 35° flexion, extension, and hyperextension in 34 normal subjects (mean age 31.1 ± 10 years) and 51 knee injury patients (mean age 36.4 ± 11.5 years, ACL tear n = 23, non-ACL injury n = 28). At each position, tibiofemoral translation and rotation were measured.

RESULTS

Normal knees showed 8.1 ± 3.3° external tibial rotation (i.e., compatible with physiological screw home mechanism) in hyperextension. The unaffected knee of ACL tear patients showed increased tibial anterior translation laterally (p = 0.005) and decreased external rotation (p = 0.002) in hyperextension compared to normal knees. ACL-tear knees had increased tibial anterior translation laterally (p < 0.001) and decreased external rotation (p < 0.001) compared to normal knees. Applying normal thresholds, fifteen (65%) of 23 ACL knees had excessive tibial anterior translation laterally while 17 (74%) had limited external rotation. None (0%) of 28 non-ACL-injured knees had excessive tibial anterior translation laterally while 13 (46%) had limited external rotation. Multidirectional malalignment was much more common in ACL-tear knees.

CONCLUSIONS

Positional MRI shows (a) physiological tibiofemoral movement in normal knees, (b) aberrant tibiofemoral alignment in the unaffected knee of ACL tear patients, and (c) a high frequency of abnormal tibiofemoral malalignment in injured knees which was more frequent, more pronounced, more multidirectional, and of a different pattern in ACL-tear knees than non-ACL-injured knees.

KEY POINTS

• Positional MRI shows physiological tibiofemoral translation and rotation in normal knees. • Positional MRI shows a different pattern of tibiofemoral alignment in the unaffected knee of ACL tear patients compared to normal control knees. • Positional MRI shows a high prevalence of abnormal tibiofemoral alignment in injured knees, which is more frequent and pronounced in ACL-tear knees than in ACL-intact injured knees.

Knee Cartilage Defect Assessment by Graph Representation and Surface Convolution

Knee osteoarthritis (OA) is the most common osteoarthritis and a leading cause of disability. Cartilage defects are regarded as major manifestations of knee OA, which are visible by magnetic resonance imaging (MRI). Thus early detection and assessment for knee cartilage defects are important for protecting patients from knee OA. In this way, many attempts have been made on knee cartilage defect assessment by applying convolutional neural networks (CNNs) to knee MRI. However, the physiologic characteristics of the cartilage may hinder such efforts: the cartilage is a thin curved layer, implying that only a small portion of voxels in knee MRI can contribute to the cartilage defect assessment; heterogeneous scanning protocols further challenge the feasibility of the CNNs in clinical practice; the CNN-based knee cartilage evaluation results lack interpretability. To address these challenges, we model the cartilages structure and appearance from knee MRI into a graph representation, which is capable of handling highly diverse clinical data. Then, guided by the cartilage graph representation, we design a non-Euclidean deep learning network with the self-attention mechanism, to extract cartilage features in the local and global, and to derive the final assessment with a visualized result. Our comprehensive experiments show that the proposed method yields superior performance in knee cartilage defect assessment, plus its convenient 3D visualization for interpretability.

Application of deep learning-based image reconstruction in MR imaging of the shoulder joint to improve image quality and reduce scan time

OBJECTIVES

To compare the image quality and diagnostic performance of conventional motion-corrected periodically rotated overlapping parallel line with enhanced reconstruction (PROPELLER) MRI sequences with post-processed PROPELLER MRI sequences using deep learning-based (DL) reconstructions.

METHODS

In this prospective study of 30 patients, conventional (19 min 18 s) and accelerated MRI sequences (7 min 16 s) using the PROPELLER technique were acquired. Accelerated sequences were post-processed using DL. The image quality and diagnostic confidence were qualitatively assessed by 2 readers using a 5-point Likert scale. Analysis of the pathological findings of cartilage, rotator cuff tendons and muscles, glenoid labrum and subacromial bursa was performed. Inter-reader agreement was calculated using Cohen's kappa statistic. Quantitative evaluation of image quality was measured using the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR).

RESULTS

Mean image quality and diagnostic confidence in evaluation of all shoulder structures were higher in DL sequences (p value = 0.01). Inter-reader agreement ranged between kappa values of 0.155 (assessment of the bursa) and 0.947 (assessment of the rotator cuff muscles). In 17 cases, thickening of the subacromial bursa of more than 2 mm was only visible in DL sequences. The pathologies of the other structures could be properly evaluated by conventional and DL sequences. Mean SNR (p value = 0.01) and CNR (p value = 0.02) were significantly higher for DL sequences.

CONCLUSIONS

The accelerated PROPELLER sequences with DL post-processing showed superior image quality and higher diagnostic confidence compared to the conventional PROPELLER sequences. Subacromial bursa can be thoroughly assessed in DL sequences, while the other structures of the shoulder joint can be assessed in conventional and DL sequences with a good agreement between sequences.

KEY POINTS

• MRI of the shoulder requires long scan times and can be hampered by motion artifacts. • Deep learning-based convolutional neural networks are used to reduce image noise and scan time while maintaining optimal image quality. The radial k-space acquisition technique (PROPELLER) can reduce the scan time and has potential to reduce motion artifacts. • DL sequences show a higher diagnostic confidence than conventional sequences and therefore are preferred for assessment of the subacromial bursa, while conventional and DL sequences show comparable performance in the evaluation of the shoulder joint.

Prone Lachman with Ultrasound: A Literature Review and Description of the Technique

ABSTRACT

Magnetic resonance imaging is the current gold standard imaging modality for diagnosing anterior cruciate ligament (ACL) tears. However, ultrasound has grown in popularity for detecting ACL injuries because of its low cost, portability, and dynamic assessment capabilities. Recent studies demonstrate high sensitivity and specificity for diagnosing isolated ACL tears via ultrasound, but tremendous heterogeneity remains for optimal technique including patient positioning, transducer placement, and dynamic versus static ultrasound usage. As ultrasound becomes ubiquitous in clinics, training rooms, and on the sidelines, identifying objective and sensitive measurements to appropriately screen athletes for significant knee injuries is imperative. This article aims to review the current role of diagnostic ultrasound in ACL injuries and propose a standardized version of the Prone Lachman with Ultrasound test, which is an objective, reliable, and easily reproducible technique to evaluate ACL competency. Developing a standardized protocol will expand the use of point-of-care ultrasound, which may reduce cost and improve efficiency in care.

Periodical assessment of four horns of knee meniscus using MR T2 mapping imaging in volunteers before and after amateur marathons

To observe the changes and recovery of T2 values of menisci in amateur marathon participants at different times, and to examine the effect of marathon exercise on meniscal microstructure. Twelve healthy marathon volunteers were recruited continuously, including 5 males and 7 females, with mean (± SD) age of 27.5 ± 5.2 years. The body mass indices (BMIs) ranged from 17.6 to 27.2 kg/m², with a mean of 21.9 ± 2.5 kg/m². The 24 knee joints were scanned using a 3 T MR scanner at 1 week before the event, and at 12 h and 2 months after the event. T2 values of the anterior horn of the medial meniscus (MMAH), posterior horn of the medial meniscus (MMPH), anterior horn of the lateral meniscus (LMAH), and posterior horn of the lateral meniscus (LMPH) were measured by drawing the regions of interest (ROIs) on the T2 map images. Wilcoxon sign rank test was used to compare the T2 values between 1 week before and 12 h after the event, and between 1 week before and 2 months after the event in each anatomical region, respectively. The T2 values of the menisci at 12 h after the event were significantly higher (P < 0.05) than those at 1 week before the event. No statistically significant differences in the T2 values of the menisci were found between 2 months after and 1 week before the event (P > 0.05). The T2 values of MMAH, MMPH, LMAH, and LMPH showed a trend of "increasing first and then decreasing" over time, suggesting that the T2 values may reflect meniscal microstructure in amateur marathon runner.

Comparative Analysis of Backbone Networks for Deep Knee MRI Classification Models

This paper focuses on different types of backbone networks for machine learning architectures which perform classification of knee Magnetic Resonance Imaging (MRI) images. This paper aims to compare different types of feature extraction networks for the same classification task, in terms of accuracy and performance. Multiple variations of machine learning models were trained based on the MRNet architecture, choosing AlexNet, ResNet, VGG-11, VGG-16, and Efficientnet as the backbone. The models were evaluated on the MRNet validation dataset, computing Area Under the Receiver Operating Characteristics Curve (ROC-AUC), accuracy, f1 score, and Cohen’s Kappa as evaluation metrics. The MRNet-VGG16 model variant shows the best results for Anterior Cruciate Ligament (ACL) tear detection. For general abnormality detection, MRNet-VGG16 is dominated by MRNet-Resnet in confidence between 0.5 and 0.75 and by MRNet-VGG11 for confidence more than 0.8. Due to the non-uniform nature of backbone network performance on different MRI planes, it is advisable to use an LR ensemble of: VGG16 on a coronal plane for all classification tasks; on an axial plane for abnormality and ACL tear detection; Alexnet on a sagittal plane for abnormality detection, and an axial plane for meniscal tear detection; and VGG11 on a sagittal plane for ACL tear detection. The results also indicate that the Cohen’s Kappa metric is valuable in model evaluation for the MRNet dataset, as it provides deeper insights on classification decisions.

Knee Ligament Sprains: Diagnosing Anterior Cruciate Ligament Injuries by Patient Interview. Development and Evaluation of the Anterior Cruciate Ligament Injury Score (ACLIS)

BACKGROUND

Knee ligament sprains are a common reason for emergency-room visits. Initially, the often difficult physical examination provides limited information, creating a risk of missing cruciate-ligament injuries, which can result in substantial functional impairments. No simple tool is available to emergency and primary-care physicians for decisions regarding specialist referral of patients with knee ligament sprains. An easy to use clinical score for the emergency setting would help identify patients at high risk of anterior cruciate ligament (ACL) tears after knee ligament sprains. The primary objective of this study, in two separate cohorts with acute knee injuries, was to develop, then validate a score for assessing the probability of ACL tear and, therefore, the need for specialist referral.

HYPOTHESIS

A score based on patient-interview information with a cut-off associated to good sensitivity and positive predictive value (PPV) for ACL tears can be developed.

MATERIAL AND METHODS

A literature review identified seven items to be used in the score: pivoting and contact activity at the time of injury, perceived cracking sound, sensation of dislocation, joint effusion, suggestive mechanism, inability to resume the activity, and immediate sensation of instability upon walking. To select the most relevant items, we recruited a development cohort of 228 patients (127 males and 101 females) with a mean age of 32±9 years who were seen for knee injuries between November 2017 and November 2018 at three healthcare institutions; 183 (80%) had ACL tears. The score was then tested in a validation cohort of 121 patients (79 males and 42 females) with a mean age of 28±2.5 years seen at two healthcare institutions between November 2019 and November 2020; 81 (67%) had ACL tears. In all patients, the diagnosis of ACL tear was confirmed by a specialist examination and magnetic resonance imaging.

RESULTS

Four items proved both sensitive and specific for ACL injury and were combined into the score: an immediate sensation of knee instability, an inability to resume the sports activity, a sensation of dislocation, and injury during a pivoting-contact activity. Patient report of two or more of these four criteria had 96% sensitivity and 66% specificity for ACL tear, with a PPV of 91% and an NPV of 83%. Results were similar in the validation cohort, confirming that a cut-off of at least two of the four items strongly suggested an ACL tear, with 94% sensitivity, 56% specificity, a PPV of 82% and an NPV of 82%.

CONCLUSION

The ACLIS score performs well for the emergency-room diagnosis of ACL tear, with 95% sensitivity, 62% specificity, an 88% PPV, and an 82% NPV. Patients with ACLIS scores of 2 or more probably require specialist referral with or without magnetic resonance imaging. The ACLIS score could be used routinely in emergency departments to decrease the proportion of patients with undiagnosed ACL tears.

LEVEL OF EVIDENCE

III, prospective case-control study of a diagnostic score.

Ex vivo comparison of 3 Tesla magnetic resonance imaging and multidetector computed tomography arthrography to identify artificial soft tissue lesions in equine stifles

OBJECTIVE

To determine the diagnostic performance of computed tomographic arthrography (CTA) and 3 Tesla magnetic resonance imaging (MRI) for detecting artificial meniscal, meniscotibial ligament (MTL) lesions and cruciate ligament (CL) lesions in horses.

STUDY DESIGN

Ex vivo controlled laboratory study.

ANIMALS

Nineteen stifles from adult horses.

METHODS

Stablike defects (n = 84) (16 mm long, 10 mm deep) were created in the menisci (n = 35), CLs (n = 24), and MTLs (n = 25) via arthroscopy prior to MRI and CTA (80 mL contrast at 85 mg/mL per joint). Two radiologists, unaware of the lesions, reached a consensus regarding the presence of lesions, based on 2 reviews of each study. Sensitivity and specificity of MRI and CTA were determined using arthroscopy as a reference and compared with McNemar's tests.

RESULTS

The sensitivity and specificity of MRI (41% and 86% respectively) and CTA (32% and 90% respectively) did not differ (P = .65). The sensitivity (MRI: 24%‐50%; CTA:19%‐40%) and specificity (MRI: 75%‐92%; CTA 75%‐100%) of imaging modalities did not differ when detecting lesions of the menisci, MTLs, and CLs (P = .1‐1.0). The highest sensitivities were achieved when MTLs were evaluated with MRI (50%) and CLs with both modalities (40%).

CONCLUSIONS

The diagnostic performance of CTA was comparable with that of MRI, with a low to moderate sensitivity and high specificity.

CLINICAL SIGNIFICANCE

Computed tomographic arthrography should be considered as an adjunct to diagnose CL injuries. This is important for equine clinicians, as the CL cannot be visualized adequately using basic imaging techniques preoperatively.

Automatic detection of attachment sites for knee ligaments and tendons on CT images

PURPOSE

The diseases and injuries of the knee joint are the most common orthopedic disorders. Personalized knee models can be helpful in the process of early intervention and lasting treatment techniques development. Fully automatic reconstruction of knee joint anatomical structures from medical images (CT, MRI, ultrasound) remains a challenge. For this reason, most of state-of-the-art knee joint models contain simplifications such as representation of muscles and ligaments as line segments connecting two points which replace attachment areas. The paper presents algorithms for automatic detection of such points on knee CT images.

METHODS

This paper presents three approaches to automatic detection of ligaments and tendons attachment sites on the patients CT images: qualitative anatomical descriptions, analysis of bones curvature, and quantitative anatomical descriptions. Combinations of these approaches result in new automatic detection algorithms. Each algorithm exploits anatomical peculiarities of each attachment site, e.g., bone curvature and number of other attachments in a neighborhood of the site.

RESULTS

The experimental dataset consisted of 26 anonymized CT sequences containing right and left knee joints in different resolutions. The proposed algorithms take into account bone surface curvatures and spatial differences in locations of medial and lateral parts of both knees. The algorithms for detection of quadriceps femoris, popliteus, biceps femoris tendons, and lateral collateral and medial collateral ligaments attachment sites are provided, as well as examples of their application. Two algorithms are validated by comparison with known statistics of ligaments lengths and also using ground truth annotations for anatomical landmarks approved by clinical experts.

CONCLUSIONS

The algorithms simplify generation of patient-specific knee joint models demanded in personalized biomechanical models. The algorithms in the current implementation have two important limitations. First, the correctness of the produced results depends on the bones segmentation quality. Second, the presented algorithms detect a point of the attachment site, which is not necessarily its center. Therefore, manual correction of the attachment site location may be required for attachments with relatively large area.

Unnecessary magnetic resonance imaging of the knee: How much is it really costing the NHS?

Aims and objectives

The aim of this study was to evaluate the indications for patients undergoing magnetic resonance imaging (MRI) of the knee prior to referral to an orthopaedic specialist and to ascertain whether these scans altered initial management.

Materials and method

A retrospective review of all referrals received by a single specialist knee surgeon over a 1-year period was performed. Patient demographics, relevant history, examination findings and past surgical procedures were documented. Patients having undergone Magnetic resonance imaging (MRI) prior to referral were identified and indications for the scans recorded. These were reviewed against The National health services (NHS) guidelines for Primary Care Physicians to identify if the imaging performed was appropriate in each case.

Results

A total of two sixty-one (261) patients were referred between 1^st July 2018 and 30^th June 2019. Eight seven out of two hundred and sixty-one patients (87/261) patients underwent knee MRI prior to referral. The average patient age was 53 years with male predominance (52 verses 35 females). Twenty-one out of eight seven patients under review (24%) underwent appropriate imaging prior to referral as per guidelines. However, only thirteen percent of patients underwent plain radiograph of knee before their scan. In cases where magnetic resonance imaging was not indicated, patients waited an average of twelve weeks between their scan and for a referral to be sent to a knee surgeon.

Conclusion

Seventy six percent of patients referred to orthopaedics had inappropriate Magnetic resonance imaging arranged by their primary care physician. For a single consultant's referrals over 1 year these unnecessary MRI (magnetic resonance imaging) of knee cost National Health Services (NHS) £13,200. Closer adherence to the guidelines by primary care physicians will result in a financial saving, better patient experience and a more effective use of resources.

•

Plain radiograph should be the first line of investigation for diagnosis of osteoarthritis.

•

A good history covering nature, site of pain, locking and history of trauma should be considered.

•

In patients with clear history of trauma and that are young then MRI should be considered or urgent referral to orthopedic specialty.

•

Incidence of degenerative tear increases with age. The first line of management for degenerative tear is conservative treatment in terms of physiotherapy and pain relief.

•Plain radiograph particularly the weight bearing views can give more information in terms of joint space narrowing compared to MRI which is a static investigation.

A Deep Learning System for Synthetic Knee Magnetic Resonance Imaging: Is Artificial Intelligence-Based Fat-Suppressed Imaging Feasible?

MATERIALS AND METHODS

This single-center study was approved by the institutional review board. Artificial intelligence-based FS MRI scans were created from non-FS images using a deep learning system with a modified convolutional neural network-based U-Net that used a training set of 25,920 images and validation set of 16,416 images. Three musculoskeletal radiologists reviewed 88 knee MR studies in 2 sessions, the original (proton density [PD] + FSPD) and the synthetic (PD + AFSMRI). Readers recorded AFSMRI quality (diagnostic/nondiagnostic) and the presence or absence of meniscal, ligament, and tendon tears; cartilage defects; and bone marrow abnormalities. Contrast-to-noise rate measurements were made among subcutaneous fat, fluid, bone marrow, cartilage, and muscle. The original MRI sequences were used as the reference standard to determine the diagnostic performance of AFSMRI (combined with the original PD sequence). This is a fully balanced study design, where all readers read all images the same number of times, which allowed the determination of the interchangeability of the original and synthetic protocols. Descriptive statistics, intermethod agreement, interobserver concordance, and interchangeability tests were applied. A P value less than 0.01 was considered statistically significant for the likelihood ratio testing, and P value less than 0.05 for all other statistical analyses.

RESULTS

Artificial intelligence-based FS MRI quality was rated as diagnostic (98.9% [87/88] to 100% [88/88], all readers). Diagnostic performance (sensitivity/specificity) of the synthetic protocol was high, for tears of the menisci (91% [71/78], 86% [84/98]), cruciate ligaments (92% [12/13], 98% [160/163]), collateral ligaments (80% [16/20], 100% [156/156]), and tendons (90% [9/10], 100% [166/166]). For cartilage defects and bone marrow abnormalities, the synthetic protocol offered an overall sensitivity/specificity of 77% (170/221)/93% (287/307) and 76% (95/125)/90% (443/491), respectively. Intermethod agreement ranged from moderate to substantial for almost all evaluated structures (menisci, cruciate ligaments, collateral ligaments, and bone marrow abnormalities). No significant difference was observed between methods for all structural abnormalities by all readers (P > 0.05), except for cartilage assessment. Interobserver agreement ranged from moderate to substantial for almost all evaluated structures. Original and synthetic protocols were interchangeable for the diagnosis of all evaluated structures. There was no significant difference for the common exact match proportions for all combinations (P > 0.01). The conspicuity of all tissues assessed through contrast-to-noise rate was higher on AFSMRI than on original FSPD images (P < 0.05).

CONCLUSIONS

Artificial intelligence-based FS MRI (3D AFSMRI) is feasible and offers a method for fast imaging, with similar detection rates for structural abnormalities of the knee, compared with original 3D MR sequences.

Intra-substance meniscal changes and their clinical significance: a meta-analysis

The degeneration of radial tie fibres of the central meniscal layer, and thinning of its lamellar layer results in increased intensity signals on magnetic resonance imaging, making it difficult to differentiate from true meniscal tear. This study aimed to assess the rate of encountered MRI grades 1 and 2 intrasubstance meniscal changes, and to set guidelines to report these changes based on predicted clinical outcome. A systematic review approach was employed using search engines, libraries, and databases (Google Scholar, ERIC, PubMed, and Medline) to search for scholarly sources on meniscal lesions and their significance in MRI published between 1 January 2000 and 30 June 2019. It retrieved 2750 abstracts, out of which 2738 were excluded and 13 studies meeting inclusion criteria were meta-analysed. It found an association between intrasubstances meniscal changes and outcomes. It resulted that intrasubstance meniscal changes were preservable through the protective functioning of the meniscus. Other than weight gain, no other significant risk factor of developing true meniscal tears later in life was found. It is important to examine intrasubstance meniscal change when patients suffer from mechanical meniscal symptoms especially in old age.

Accuracy measures of 1.5-tesla MRI for the diagnosis of ACL, meniscus and articular knee cartilage damage and characteristics of false negative lesions: a level III prognostic study

BACKGROUND

MRI is the most accurate imaging modality for diagnosing knee pathologies. However, there is uncertainty concerning factors predicting false negative MRI, such as meniscal tear patterns as well as patient factors. The aims of this study were to report 1.5-Tesla MRI accuracy of ACL, meniscus and articular cartilage damage and characterize false negative lesions.

METHODS

Two hundred eighteen consecutive knee arthroscopies performed in our institution between 2013 and 2016 and their respective prospectively-collected MRI reports were reviewed. Inclusion criteria were age > 15 years-old, primary arthroscopy, 1.5-Tesla MRI performed at the same institution, and time interval MRI-surgery < 6 months. Exclusion criteria were revision arthroscopy and arthroscopic-assisted fracture fixation or multiligament surgery. Accuracy measures and Kappa coefficients were calculated comparing the MRI diagnosis to the arthroscopic findings. Moreover, the arthroscopic findings of false negative MRI were compared to the findings of true positive MRI using the Fisher-exact test. Pearson correlation was used for testing the correlation between MRI accuracy and patient age.

RESULTS

The highest accuracy was observed in medial meniscus and in ACL findings. For the medial meniscus sensitivity, specificity, agreement, and Kappa coefficient were 77, 92, 86%, and 0.7, and for the ACL these measures were 82, 97, 87%, and 0.73. MRI accuracy was lower in the lateral meniscus and articular cartilage with Kappa coefficient 0.42 and 0.3, respectively. More specifically, short peripheral tears in the posterior horn of the medial meniscus were characteristic of false negative findings compared to true positive findings of the MRI (p <  0.01). MRI accuracy correlated negatively compared to arthroscopic findings with patient age for the medial meniscus (r = - 0.21, p = 0.002) and for articular cartilage damage (r = - 0.45, p <  0.001).

CONCLUSION

1.5-Tesla MRI will accurately diagnose ACL and medial meniscal tears and can reliably complete the diagnostic workup following physical examination, particularly in young adults. This modality however is not reliable for diagnosing short peripheral tears at the posterior horn of the medial meniscus and partial thickness articular cartilage lesion of the femoral condyles. For these lesions, definitive diagnosis may require cartilage-specific MRI sequences or direct arthroscopic evaluation.

LEVEL OF EVIDENCE

Prognostic study, Level III.

Clinical practice and postoperative rehabilitation after knee arthroscopy vary according to surgeons' expertise: a survey among polish arthroscopy society members

BACKGROUND

Meniscus repair is a challenging task in knee arthroscopy. Currently, there are a variety of arthroscopic methods available for meniscus repair. The purpose of this study was to determine a consensus in meniscus tear treatment in the environment of Polish orthopaedists.

METHODS

A total of 205 registered orthopaedic surgeons participated in the surveys. The survey consisted of 35 questions regarding general arthroscopy and postoperative management, including physicians' level of expertise, anaesthesia, postoperative treatment, rehabilitation and procedures performed. Comparisons were made between knee arthroscopy experts (> 100 arthroscopies performed per year) and non-experts (≤ 100 cases) on aspects of patient care.

RESULTS

The most important finding of this study was the agreement among almost all aspects of the knee arthroscopy approach. Consensus among Polish surgeons was noticed in choosing regional anaesthesia for knee arthroscopy, the lack of need for knee braces and knee medications, the of use of LMW heparin for thromboprophylaxis, 1-2 days of hospitalization, the recommendation of rehabilitation and the use of magnetic resonance as a diagnostic test for meniscus damage. Surgical expertise was significantly associated with the performance of meniscus suture procedures (p = 0.009). Experts recommended starting rehabilitation on the day of surgery (p = 0.007) and were more likely to use objective physical tests (p = 0.003). Non-expert surgeons recommended a longer period from meniscus suture to full-range knee motion (p = 0.001) and admitted that patient age does matter for meniscus repair qualification (p = 0.002).

CONCLUSIONS

There is consensus among almost all issues of meniscus tear treatment in the environment of Polish orthopaedists; however, the issues of rehabilitation and the use of advanced meniscus repair techniques are associated with surgical expertise.

Magnetic Resonance Imaging Evaluation of the Anterolateral Ligament and the Iliotibial Band in Acute Anterior Cruciate Ligament Injuries Associated With Segond Fractures

PURPOSE

To evaluate the magnetic resonance imaging (MRI) characteristics of Segond fractures, including the structures attached to the avulsed fragment, the integrity of anterolateral ligament (ALL) and iliotibial band (ITB), and fragment size and location.

METHODS

This was a retrospective, cross-sectional study with MRI scans from 2016 to 2019 with the term "Segond" in the reports, signs of acute trauma, and a bony anterolateral tibial avulsion (Segond) fracture. Two experienced observers accessed images to evaluate fragment dimensions (anteroposterior, mediolateral, and craniocaudal) and distances from anatomic landmarks (Gerdy's tubercle, articular surface, and posterolateral tibial corner). ALLs and ITBs also were evaluated, both for integrity and for attachment to the avulsed bony fragment. Data were statistically evaluated for significant correlations.

RESULTS

Forty-eight knee MRIs of patients suffering from a combined anterior cruciate ligament and Segond injury were evaluated. The ALL presented with edema in 28 cases (58.3%) and was torn in 3 cases (6.3%). The ALL was attached to the bone fragment in all cases and the ITB also was attached in 25 cases (52.1%). Receiver operating characteristic curves also demonstrated that the larger each of the dimensions and the volume of the fragment, the greater the probability of ITB fibers being inserted. Also, the narrower the distance from the fragment to the center of Gerdy's tubercle, the greater the probability of iliotibial band fibers being attached. Interobserver correlation varied from 0.87 to 0.97 for all measurements.

CONCLUSIONS

In all combined ACL injuries and Segond fracture MRI cases, the complete ALL inserted on the avulsed bony fragment, whereas the posterior part of the ITB was only attached in 52.1% of the cases. Segond fracture fragment size was predictive for the structures attached to it. The ALL was abnormal in 64.5% of cases and presented a clear discontinuity in 6.3%.

LEVEL OF EVIDENCE

Case Series, Level IV.

Correlation between notch width index assessed via magnetic resonance imaging and risk of anterior cruciate ligament injury: an updated meta-analysis

PURPOSE

To analyze the correlation between notch width index (NWI) and/or femoral intercondylar notch width (NW) assessed by magnetic resonance imaging (MRI) and risk of anterior cruciate ligament (ACL) injury.

METHODS

We searched the PubMed, Embase, China National Knowledge Infrastructure and Wanfang databases for literature reporting a correlation between ACL injury and NWI and/or NW. Subgroup analyses were stratified by ethnicity, sex and control source. The weighted mean difference (WMD) and 95% confidence intervals (95% CIs) were calculated for the ACL injury cases and controls using random- or fixed-effects models. Begg's test and sensitivity analyses were applied to assess publication bias and stability of the results, respectively.

RESULTS

Twenty-eight eligible studies were finally enrolled. The NW was significantly narrowerin the ACL injury cases than in the control cases (pooled WMD, - 1.88 [95% CI, - 2.43 to - 1.32]). The results were similar when stratified by ethnicity and sex. Similarly, the NWI was lower in ACL injury cases than in the controls. Asian populations presented similar results when stratified by ethnicity, among the self-control group when stratified by control source, and among men when stratified by sex. No publication bias was identified; however, the sensitivity analysis suggested unstable results in the NWI subgroup analysis.

CONCLUSIONS

The current meta-analysis evidenced that the NW assessed via MRI was significantly smaller in ACL injury cases than in the controls. The NWI was lower in ACL injury cases among men. Prevention strategies for ACL injury could be applied for people with intercondylar notch stenosis.

Increased signal in the proximal patellar tendon: normal or pathologic?

OBJECTIVE

To determine the clinical significance of T2 signal hyperintensity in the proximal patellar tendon seen on MRI of the knee.

MATERIALS AND METHODS

MRIs of 100 patients who underwent MRI of the knee between 1 May 2018 and 15 July 2018 were retrospectively evaluated. All examinations were performed on 3-Tesla MRI scanners with a dedicated knee coil and in accordance with our institution's standard knee MRI protocol. The presence of increased T2 signal was assessed on both sagittal and axial T2-weighted fat-saturated images. The amount of increased signal in the proximal patellar tendon on T2-weighted images was characterized as: none, mild, moderate, or severe. A corresponding chart review of the referring physicians' notes was performed to determine the presence of clinical symptoms of patellar tendinopathy. Patellar tendinopathy was considered present if the clinical notes described tenderness on palpation of the inferior patellar pole, infrapatellar tenderness, or patellar tendinosis/tendinitis.

RESULTS

The majority (66%) of knee MRIs demonstrated the presence of increased T2 signal in the proximal patellar tendon. Only 4.5% of these patients had associated clinical findings of patellar tendinopathy.

CONCLUSION

Although increased T2 signal in the proximal patellar tendon is a common finding, only in rare cases are there associated clinical symptoms. Thus, increased T2 signal in the proximal patellar tendon may not be a pathological finding in the absence of clinical findings of patellar tendinopathy.

Comparison Between 3-Dimensional Multiple-Echo Recombined Gradient Echo Magnetic Resonance Imaging and Arthroscopic Findings for the Evaluation of Acetabular Labrum Tear

PURPOSE

To evaluate radially reconstructed 3.0-Tesla 3-dimensional multiple-echo recombined gradient echo (MERGE) magnetic resonance imaging (MRI) without arthrography for the assessment of acetabular labrum tears, using arthroscopic evaluation as the reference standard.

METHODS

A total of 71 consecutive hips, including 29 with femoroacetabular impingement, 26 with borderline developmental dysplasia of the hip, and 16 with early-stage osteoarthritis, were evaluated in this retrospective study. MERGE MRI findings were evaluated according to the modified Czerny classification for 3 regions of interest: anterior region, anterolateral region, and lateral region. Cases with severe degeneration that was not concordant with any stage in the original Czerny classification were defined as stage Ⅳ. MERGE MRI findings were compared with arthroscopic findings, and the sensitivity, specificity, positive predictive value, and negative predictive value in terms of the existence of labrum tears were calculated.

RESULTS

MERGE MRI findings revealed labrum tears more frequently in the anterolateral region than in the anterior and lateral regions (P < .01). In cases of femoroacetabular impingement and borderline developmental dysplasia of the hip in particular, labrum tears were more frequently observed on MRI in the anterolateral region than in the lateral region (P < .05). In comparison with MRI findings and arthroscopic findings, our newly defined stage IV in the modified Czerny classification was more frequently observed in cases with a Multicenter Arthroscopy of the Hip Outcomes Research Network (MAHORN) classification of degenerative or complex (P < .01). The average sensitivity and specificity of all regions for the existence of labrum tears were 85% and 56%, respectively. Sensitivity and specificity were 79% and 50%, respectively, in the anterior region; 96% and 50%, respectively, in the anterolateral region; and 70% and 57%, respectively, in the lateral region.

CONCLUSIONS

We validated the diagnostic performance of 3.0-Tesla 3-dimensional MERGE MRI for evaluating acetabular labrum tears and made comparisons with arthroscopic findings. Radially reconstructed MERGE magnetic resonance images showed excellent sensitivity for the diagnosis of labrum tears, particularly in the anterolateral region. The newly defined stage IV was distinctive of early-stage osteoarthritis cases with degeneration and/or complex arthroscopic findings. The noninvasive imaging modality of radially reconstructed MERGE MRI may be an alternative to magnetic resonance arthrography for evaluating labrum tears.

LEVELS OF EVIDENCE

Level Ⅱ, development of diagnostic criteria.

Deep-learning-assisted diagnosis for knee magnetic resonance imaging: Development and retrospective validation of MRNet

BACKGROUND

Magnetic resonance imaging (MRI) of the knee is the preferred method for diagnosing knee injuries. However, interpretation of knee MRI is time-intensive and subject to diagnostic error and variability. An automated system for interpreting knee MRI could prioritize high-risk patients and assist clinicians in making diagnoses. Deep learning methods, in being able to automatically learn layers of features, are well suited for modeling the complex relationships between medical images and their interpretations. In this study we developed a deep learning model for detecting general abnormalities and specific diagnoses (anterior cruciate ligament [ACL] tears and meniscal tears) on knee MRI exams. We then measured the effect of providing the model's predictions to clinical experts during interpretation.

METHODS AND FINDINGS

Our dataset consisted of 1,370 knee MRI exams performed at Stanford University Medical Center between January 1, 2001, and December 31, 2012 (mean age 38.0 years; 569 [41.5%] female patients). The majority vote of 3 musculoskeletal radiologists established reference standard labels on an internal validation set of 120 exams. We developed MRNet, a convolutional neural network for classifying MRI series and combined predictions from 3 series per exam using logistic regression. In detecting abnormalities, ACL tears, and meniscal tears, this model achieved area under the receiver operating characteristic curve (AUC) values of 0.937 (95% CI 0.895, 0.980), 0.965 (95% CI 0.938, 0.993), and 0.847 (95% CI 0.780, 0.914), respectively, on the internal validation set. We also obtained a public dataset of 917 exams with sagittal T1-weighted series and labels for ACL injury from Clinical Hospital Centre Rijeka, Croatia. On the external validation set of 183 exams, the MRNet trained on Stanford sagittal T2-weighted series achieved an AUC of 0.824 (95% CI 0.757, 0.892) in the detection of ACL injuries with no additional training, while an MRNet trained on the rest of the external data achieved an AUC of 0.911 (95% CI 0.864, 0.958). We additionally measured the specificity, sensitivity, and accuracy of 9 clinical experts (7 board-certified general radiologists and 2 orthopedic surgeons) on the internal validation set both with and without model assistance. Using a 2-sided Pearson's chi-squared test with adjustment for multiple comparisons, we found no significant differences between the performance of the model and that of unassisted general radiologists in detecting abnormalities. General radiologists achieved significantly higher sensitivity in detecting ACL tears (p-value = 0.002; q-value = 0.019) and significantly higher specificity in detecting meniscal tears (p-value = 0.003; q-value = 0.019). Using a 1-tailed t test on the change in performance metrics, we found that providing model predictions significantly increased clinical experts' specificity in identifying ACL tears (p-value < 0.001; q-value = 0.006). The primary limitations of our study include lack of surgical ground truth and the small size of the panel of clinical experts.

CONCLUSIONS

Our deep learning model can rapidly generate accurate clinical pathology classifications of knee MRI exams from both internal and external datasets. Moreover, our results support the assertion that deep learning models can improve the performance of clinical experts during medical imaging interpretation. Further research is needed to validate the model prospectively and to determine its utility in the clinical setting.

High field magnetic resonance imaging is comparable with gross anatomy for description of the normal appearance of soft tissues in the equine stifle

High field magnetic resonance imaging (MRI) is increasingly used for horses with suspected stifle disease, however there is limited available information on normal imaging anatomy and potential incidental findings. The aim of this prospective, anatomic study was to develop an optimized high field MRI protocol for evaluation of the equine stifle and provide detailed descriptions of the normal MRI appearance of the stifle soft tissues, using ultrasound and gross pathological examination as comparison tests. Nine cadaver limbs were acquired from clinically normal horses. Stifles were evaluated ultrasonographically and then by an extensive 1.5 T MRI protocol. Subsequently, all stifles were evaluated for gross pathologic change. Findings were compared between gross evaluation and MRI imaging and described. No soft tissue structure abnormalities were identified on any evaluation. Specific descriptive findings of the meniscotibial, meniscofemoral, collateral, patellar and cruciate ligaments, and the menisci were reported. The high field MRI protocol described in this study provided high spatial and contrast resolution of the soft tissue structures, and this in turn allowed visualization of detailed structural characteristics, such as striations and variations in signal intensity. Findings supported the use of high field MRI as a modality for the evaluation of the soft tissues of the equine stifle. As clinical availability of this modality increases in the future, authors anticipate that new stifle diseases will be detected that have not previously been identified with other imaging modalities.

MRI analysis and clinical significance of lower extremity muscle cross-sectional area after spinal cord injury

Shortly after spinal cord injury (SCI), the musculoskeletal system undergoes detrimental changes in size and composition, predominantly below the level of injury. The loss of muscle size and strength, along with increased immobility, predisposes persons with SCI to rapid and severe loss in bone mineral density and other health related consequences. Previous studies have highlighted the significance of measuring thigh muscle cross-sectional area, however, measuring the size and composition of muscles of the lower leg may provide insights on how to decrease the risk of various comorbidities. The purpose of the current review was to summarize the methodological approach to manually trace and measure the muscles of the lower leg in individuals with SCI, using magnetic resonance imaging. We also intend to highlight the significance of analyzing lower leg muscle cross-sectional area and its relationship to musculoskeletal and vascular systems in persons with SCI.