Magnetic resonance imaging of the musculoskeletal system. Part 8. The spine, section 2

Role of Artificial intelligence model in prediction of low back pain using T2 weighted MRI of Lumbar spine

Background

Low back pain (LBP), the primary cause of disability, is the most common musculoskeletal disorder globally and the primary cause of disability. Magnetic resonance imaging (MRI) studies are inconclusive and less sensitive for identifying and classifying patients with LBP. Hence, this study aimed to investigate the role of artificial intelligence (AI) models in the prediction of LBP using T2 weighted MRI image of the lumbar spine.

Methods

This was a prospective case-control study. A total of 200 MRI patients (100 cases and controls each) referred for lumbar spine and whole spine screening were included. The scans were performed using 3.0 Tesla MRI (United Imaging Healthcare). T2 weighted images of the lumbar spine were segmented to extract radiomic features. Machine learning (ML) models, such as random forest, decision tree, logistic regression, K-nearest neighbors, adaboost, and deep learning methods (DL), such as ResNet and GoogleNet, were used, and performance measures were calculated.

Results

Our study showed that Random forest and AdaBoost are the most reliable ML models for predicting LBP. Random forest showed high performance with area under curve (AUC) values from 0.83 to 0.88 across all lumbar vertebrae and L2-L3, L3-L4, and L4-L5 intervertebral discs (IVDs), with AUCs of 0.88 the highest at L5-S1 IVD (0.92). Adaboost demonstrated high performance at the L2-L5 vertebrae with AUC values of 0.82 to 0.90, with the highest AUC (0.97) at the L5-S1 IVD. Among the DL models, GoogleNet outperformed the other models at 30 epochs with an accuracy of 0.85, followed by ResNet 18 (30 epochs) with an accuracy of 0.84.

Conclusion

The study demonstrated that ML and DL models can effectively predict LBP from MRI T2 weighted image of the lumbar spine. ML and DL models could also enhance the diagnostic accuracy of LBP, potentially leading to better patient management and outcomes.

Morphology and Composition of Lumbar Intervertebral Discs: Comparative Analyses of Manual Measurement and Computer-Assisted Algorithms

BACKGROUND

The morphology and internal composition, particularly the nucleus-to-cross sectional area (NP-to-CSA) ratio of the lumbar intervertebral discs (IVDs), is important information for finite element models (FEMs) of spinal loadings and biomechanical behaviors, and, yet, this has not been well investigated and reported.

METHODS

Anonymized MRI scans were retrieved from a previously established database, including a total of 400 lumbar IVDs from 123 subjects (58 F and 65 M). Measurements were conducted manually by a spine surgeon and using two computer-assisted segmentation algorithms, i.e., fuzzy C-means (FCM) and region growing (RG). The respective results were compared. The influence of gender and spinal level was also investigated.

RESULTS

Ratios derived from manual measurements and the two computer-assisted algorithms (FCM and RG) were 46%, 39%, and 38%, respectively. Ratios derived manually were significantly larger.

CONCLUSIONS

Computer-assisted methods provide reliable outcomes that are traditionally difficult for the manual measurement of internal composition. FEMs should consider the variability of NP-to-CSA ratios when studying the biomechanical behavior of the spine.

Machine Learning Assisting the Prediction of Clinical Outcomes following Nucleoplasty for Lumbar Degenerative Disc Disease

BACKGROUND

Lumbar degenerative disc disease (LDDD) is a leading cause of chronic lower back pain; however, a lack of clear diagnostic criteria and solid LDDD interventional therapies have made predicting the benefits of therapeutic strategies challenging. Our goal is to develop machine learning (ML)-based radiomic models based on pre-treatment imaging for predicting the outcomes of lumbar nucleoplasty (LNP), which is one of the interventional therapies for LDDD.

METHODS

The input data included general patient characteristics, perioperative medical and surgical details, and pre-operative magnetic resonance imaging (MRI) results from 181 LDDD patients receiving lumbar nucleoplasty. Post-treatment pain improvements were categorized as clinically significant (defined as a ≥80% decrease in the visual analog scale) or non-significant. To develop the ML models, T2-weighted MRI images were subjected to radiomic feature extraction, which was combined with physiological clinical parameters. After data processing, we developed five ML models: support vector machine, light gradient boosting machine, extreme gradient boosting, extreme gradient boosting random forest, and improved random forest. Model performance was measured by evaluating indicators, such as the confusion matrix, accuracy, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC), which were acquired using an 8:2 allocation of training to testing sequences.

RESULTS

Among the five ML models, the improved random forest algorithm had the best performance, with an accuracy of 0.76, a sensitivity of 0.69, a specificity of 0.83, an F1 score of 0.73, and an AUC of 0.77. The most influential clinical features included in the ML models were pre-operative VAS and age. In contrast, the most influential radiomic features had the correlation coefficient and gray-scale co-occurrence matrix.

CONCLUSIONS

We developed an ML-based model for predicting pain improvement after LNP for patients with LDDD. We hope this tool will provide both doctors and patients with better information for therapeutic planning and decision-making.

Correlation between T2 relaxation time and intervertebral disk degeneration

OBJECTIVE

Magnetic resonance T2 mapping allows for the quantification of water and proteoglycan content within tissues and can be used to detect early cartilage abnormalities as well as to track the response to therapy. The goal of the present study was to use T2 mapping to quantify intervertebral disk water content according to the Pfirrmann classification.

MATERIALS AND METHODS

This study involved 60 subjects who underwent lumbar magnetic resonance imaging (a total of 300 lumbar disks). The degree of disk degeneration was assessed in the midsagittal section on T2-weighted images according to the Pfirrmann classification (grades I to V). Receiver operating characteristic (ROC) analysis was performed among grades to determine the cut-off values.

RESULTS

In the nucleus pulposus, T2 values tended to decrease with increasing grade, and there was a significant difference in T2 values between each grade from grades I to IV. However, there was no significant difference in T2 values in the anterior or posterior annulus fibrosus. T2 values according to disk degeneration level classification were as follows: grade I (>116.8 ms), grade II (92.7-116.7 ms), grade III (72.1-92.6 ms), grade IV (<72.0 ms).

CONCLUSION

T2 values decreased with increasing Pfirrmann classification grade in the nucleus pulposus, likely reflecting a decrease in proteoglycan and water content. Thus, T2 value-based measurements of intervertebral disk water content may be useful for future clinical research on degenerative disk diseases.

Imaging modalities for cervical spondylotic stenosis and myelopathy

Cervical spondylosis is a spectrum of pathology presenting as neck pain, radiculopathy, and myelopathy or all in combination. Diagnostic imaging is essential to diagnosis and preoperative planning. We discuss the modalities of imaging in common practice. We examine the use of imaging to differentiate among central, subarticular, and lateral stenosis and in the assessment of myelopathy.

Scheuermann's kyphosis

Scheuermann's disease is a kyphotic deformity of the spine that develops in early adolescence. This condition has been reported to occur in 0.4% to 8% of the general population, with an equal distribution between sexes. Diagnosis of Scheuermann's disease is suggested on clinical examination; however, parents of children affected often confuse it with poor posture. Radiographs are the standard imaging modality used to confirm the diagnosis of Scheuermann's disease. Classic signs include vertebral end plate irregularity, disk space narrowing, and anterior wedging of involved vertebral bodies. Other diagnostic tools such as CT scans or magnetic resonance imaging may also be of value in the evaluation of Scheuermann's disease. The mode of treatment for this condition depends upon the severity of the deformity, remaining growth, and presence or absence of symptoms. Early treatment may be limited to observation and exercises, whereas patients who have kyphosis of up to 75 degrees and how have growth remaining may benefit from bracing. Surgical correction is reserved for severe cases that are symptomatic and refractory to conservative management.