Magnetic resonance imaging of the knee

Imaging of the medial collateral ligament of the knee: a systematic review

INTRODUCTION

The primary aim of this investigation was to systematically review relevant literature of various imaging modalities (magnetic resonance imaging (MRI), stress radiography and ultrasonography) in the assessment of patients with a medial collateral ligament (MCL) injury.

MATERIALS AND METHODS

A systematic literature review of articles indexed in PubMed and Cochrane library was performed. Original research reporting data associated with medial gapping, surgical, and clinical findings associated with MCL injuries were considered for inclusion. The methodological quality of each inclusion was also assessed using a verified tool.

RESULTS

Twenty-three imaging studies (magnetic resonance imaging (MRI) n = 14; ultrasonography n = 6; radiography n = 3) were ultimately included into the review. A total of 808 injured, and 294 control, knees were assessed. Interobserver reliabilities were reported in radiographic and ultrasonographic investigations with almost perfect agreement. MRI studies demonstrated agreement ranging between substantial to almost perfect. Intraobserver reliability was only reported in radiographic studies pertinent to medial gapping and was found to be almost perfect. Correlation of MRI with clinical findings was moderate to strong (65-92%). Additionally, MRI imaging was more sensitive in the detection of MCL lesions when compared to clinical examination. However, when compared to surgical findings, MRI underestimated the grade of instability in up to 21% of cases. Furthermore, MRI showed relatively inferior performance in the identification of the exact MCL-lesion location when compared to surgical findings. Interestingly, preoperative clinical examination was slightly inferior to stress radiography in the detection of MCL lesions. However, clinical testing under general anaesthesia performed similar to stress radiography. The methodological quality analysis showed a low risk of bias regarding patient selection and index testing in each imaging modality.

CONCLUSION

MRI can reliably diagnose an MCL lesion but demonstrates limitations in its ability to predict the specific lesion location or grade of MCL instability. Ultrasonography is a widely available, radiation free modality, but is rarely used in clinical practice for detecting MCL lesions and clinical or surgical correlates are scarce. Stress radiography findings correlate with surgical findings but clinical correlations are missing in the literature.

LEVEL OF EVIDENCE

IV.

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.

Knee Injury Detection Using Deep Learning on MRI Studies: A Systematic Review

The improved treatment of knee injuries critically relies on having an accurate and cost-effective detection. In recent years, deep-learning-based approaches have monopolized knee injury detection in MRI studies. The aim of this paper is to present the findings of a systematic literature review of knee (anterior cruciate ligament, meniscus, and cartilage) injury detection papers using deep learning. The systematic review was carried out following the PRISMA guidelines on several databases, including PubMed, Cochrane Library, EMBASE, and Google Scholar. Appropriate metrics were chosen to interpret the results. The prediction accuracy of the deep-learning models for the identification of knee injuries ranged from 72.5-100%. Deep learning has the potential to act at par with human-level performance in decision-making tasks related to the MRI-based diagnosis of knee injuries. The limitations of the present deep-learning approaches include data imbalance, model generalizability across different centers, verification bias, lack of related classification studies with more than two classes, and ground-truth subjectivity. There are several possible avenues of further exploration of deep learning for improving MRI-based knee injury diagnosis. Explainability and lightweightness of the deployed deep-learning systems are expected to become crucial enablers for their widespread use in clinical practice.

The efficacy of magnetic resonance imaging in acute multi-ligament injuries

This study compares MRI with examination under anaesthesia to surgical findings in evaluating soft tissue injuries in acute multi-ligament knee trauma. Pre-operative MRI was done for 44 patients who underwent surgery for grade III ACL and grade III medial collateral ligament (MCL) injury. In 21 cases both ACL and MCL were treated surgically, but in 23 only ACL. Intra-operative and MRI findings were compared. Accuracy of MRI for medial meniscal tears was 88.6%, sensitivity 80%, and specificity 91.2%; accuracy for lateral meniscal tears was 72.7%, sensitivity 55% and specificity 87.5%. Accuracy and sensitivity for severity of ACL tear was 93.2% and of MCL tear 86.4%. In 88.6% of the knees, bone bruises were visible, with anterolateral femoral and posterolateral tibial bone bruise being the most common. MRI revealed no chondral lesion, but arthroscopy revealed 11. In combined ACL-MCL ruptures, the incidence of concomitant injuries is high and the injuries are best detected with MRI.

Electrical stimulation and treadmill gait in tetraplegic patients: assessment of its effects on the knee with magnetic resonance imaging

STUDY DESIGN

Evaluation of knees of tetraplegic patients who have been walking for several months with the aid of a system that involves neuromuscular stimulation, treadmill and a harness support device.

OBJECTIVES

To investigate if the training program could cause knee injury to tetraplegic patients.

SETTING

Hospital das Clinicas - UNICAMP. Campinas-SP, Brazil.

METHODS

Nine patients were evaluated. Clinical exam and magnetic resonance images (MRIs) were used for evaluation. MRIs were taken before and after the training program, in a 6-month interval for each patient. There were two sessions of training every week. Each session lasted 20 min.

RESULTS

No severe clinical abnormality was observed in any patient. Mild knee injury was observed in four of nine patients studied.

CONCLUSIONS

Tetraplegic patients undergoing treadmill gait training deserve a close follow-up to prevent knee injury.