Dataset Information

Automatically Diagnosing Disk Bulge and Disk Herniation With Lumbar Magnetic Resonance Images by Using Deep Convolutional Neural Networks: Method Development Study.

ABSTRACT:

Background

Disk herniation and disk bulge are two common disorders of lumbar intervertebral disks (IVDs) that often result in numbness, pain in the lower limbs, and lower back pain. Magnetic resonance (MR) imaging is one of the most efficient techniques for detecting lumbar diseases and is widely used for making clinical diagnoses at hospitals. However, there is a lack of efficient tools for effectively interpreting massive amounts of MR images to meet the requirements of many radiologists.

Objective

The aim of this study was to present an automatic system for diagnosing disk bulge and herniation that saves time and can effectively and significantly reduce the workload of radiologists.

Methods

The diagnosis of lumbar vertebral disorders is highly dependent on medical images. Therefore, we chose the two most common diseases-disk bulge and herniation-as research subjects. This study is mainly about identifying the position of IVDs (lumbar vertebra [L] 1 to L2, L2-L3, L3-L4, L4-L5, and L5 to sacral vertebra [S] 1) by analyzing the geometrical relationship between sagittal and axial images and classifying axial lumbar disk MR images via deep convolutional neural networks.

Results

This system involved 4 steps. In the first step, it automatically located vertebral bodies (including the L1, L2, L3, L4, L5, and S1) in sagittal images by using the faster region-based convolutional neural network, and our fourfold cross-validation showed 100% accuracy. In the second step, it spontaneously identified the corresponding disk in each axial lumbar disk MR image with 100% accuracy. In the third step, the accuracy for automatically locating the intervertebral disk region of interest in axial MR images was 100%. In the fourth step, the 3-class classification (normal disk, disk bulge, and disk herniation) accuracies for the L1-L2, L2-L3, L3-L4, L4-L5, and L5-S1 IVDs were 92.7%, 84.4%, 92.1%, 90.4%, and 84.2%, respectively.

Conclusions

The automatic diagnosis system was successfully built, and it could classify images of normal disks, disk bulge, and disk herniation. This system provided a web-based test for interpreting lumbar disk MR images that could significantly improve diagnostic efficiency and standardized diagnosis reports. This system can also be used to detect other lumbar abnormalities and cervical spondylosis.

SUBMITTER: Pan Q

PROVIDER: S-EPMC8178733 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Aim Accurate severity grading of lumbar spine disease by magnetic resonance images (MRIs) plays an important role in selecting appropriate treatment for the disease. However, interpreting these complex MRIs is a repetitive and time-consuming workload for clinicians, especially radiologists. Here, we aim to develop a multi-task classification model based on artificial intelligence for automated grading of lumbar disc herniation (LDH), lumbar central canal stenosis (LCCS) and lumbar nerve roots compression (LNRC) at lumbar axial MRIs. Methods Total 15254 lumbar axial T2W MRIs as the internal dataset obtained from the Fifth Affiliated Hospital of Sun Yat-sen University from January 2015 to May 2019 and 1273 axial T2W MRIs as the external test dataset obtained from the Third Affiliated Hospital of Southern Medical University from June 2016 to December 2017 were analyzed in this retrospective study. Two clinicians annotated and graded all MRIs using the three international classification systems. In agreement, these results served as the reference standard; In disagreement, outcomes were adjudicated by an expert surgeon to establish the reference standard. The internal dataset was randomly split into an internal training set (70%), validation set (15%) and test set (15%). The multi-task classification model based on ResNet-50 consists of a backbone network for feature extraction and three fully-connected (FC) networks for classification and performs the classification tasks of LDH, LCCS, and LNRC at lumbar MRIs. Precision, accuracy, sensitivity, specificity, F1 scores, confusion matrices, receiver-operating characteristics and interrater agreement (Gwet k) were utilized to assess the model’s performance on the internal test dataset and external test datasets. Results A total of 1115 patients, including 1015 patients from the internal dataset and 100 patients from the external test dataset [mean age, 49 years ± 15 (standard deviation); 543 women], were evaluated in this study. The overall accuracies of grading for LDH, LCCS and LNRC were 84.17% (74.16%), 86.99% (79.65%) and 81.21% (74.16%) respectively on the internal (external) test dataset. Internal and external testing of three spinal diseases showed substantial to the almost perfect agreement (k, 0.67 - 0.85) for the multi-task classification model. Conclusion The multi-task classification model has achieved promising performance in the automated grading of LDH, LCCS and LNRC at lumbar axial T2W MRIs.

Project description:ContextFor patients with lumbar disk herniation, the Spine Patient Outcomes Research Trial (SPORT) randomized trial intent-to-treat analysis showed small but not statistically significant differences in favor of diskectomy compared with usual care. However, the large numbers of patients who crossed over between assigned groups precluded any conclusions about the comparative effectiveness of operative therapy vs usual care.ObjectiveTo compare the treatment effects of diskectomy and usual care.Design, setting, and patientsProspective observational cohort of surgical candidates with imaging-confirmed lumbar intervertebral disk herniation who were treated at 13 spine clinics in 11 US states and who met the SPORT eligibility criteria but declined randomization between March 2000 and March 2003.InterventionsStandard open diskectomy vs usual nonoperative care.Main outcome measuresChanges from baseline in the Medical Outcomes Study Short-Form Health Survey (SF-36) bodily pain and physical function scales and the modified Oswestry Disability Index (American Academy of Orthopaedic Surgeons/MODEMS version).ResultsOf the 743 patients enrolled in the observational cohort, 528 patients received surgery and 191 received usual nonoperative care. At 3 months, patients who chose surgery had greater improvement in the primary outcome measures of bodily pain (mean change: surgery, 40.9 vs nonoperative care, 26.0; treatment effect, 14.8; 95% confidence interval, 10.8-18.9), physical function (mean change: surgery, 40.7 vs nonoperative care, 25.3; treatment effect, 15.4; 95% CI, 11.6-19.2), and Oswestry Disability Index (mean change: surgery, -36.1 vs nonoperative care, -20.9; treatment effect, -15.2; 95% CI, -18.5. to -11.8). These differences narrowed somewhat at 2 years: bodily pain (mean change: surgery, 42.6 vs nonoperative care, 32.4; treatment effect, 10.2; 95% CI, 5.9-14.5), physical function (mean change: surgery, 43.9 vs nonoperavtive care 31.9; treatment effect, 12.0; 95% CI; 7.9-16.1), and Oswestry Disability Index (mean change: surgery -37.6 vs nonoperative care -24.2; treatment effect, -13.4; 95% CI, -17.0 to -9.7).ConclusionsPatients with persistent sciatica from lumbar disk herniation improved in both operated and usual care groups. Those who chose operative intervention reported greater improvements than patients who elected nonoperative care. However, nonrandomized comparisons of self-reported outcomes are subject to potential confounding and must be interpreted cautiously.Trial registrationclinicaltrials.gov Identifier: NCT00000410.

Project description:PurposeLumbar disc herniation (LDH) is one of the leading causes of low-back pain and results in a series of clinical symptoms, including pain, reflex loss, and muscle weakness. Spinal manipulative therapy (SMT) can relieve pain and promote internal and external stabilization of the lumbar spine. In this study, we investigated whether the brain alterations of LDH patients with SMT were frequency-dependent based on the calculation of Amplitude of Low-Frequency Fluctuations (ALFF) and fractional ALFF (fALFF). Further, we established a cohort of LDH patients to evaluate the contribution of SMT treatments to brain functional reorganization.MethodsA total of 55 participants, including 27 LDH patients and 28 health controls (HCs), were collected. All LDH patients underwent two fMRI scans (before SMT and after the sixth SMT session). To represent LDH-related brain oscillatory activities, we calculated the ALFF and fALFF in the conventional band (0.01-0.08 Hz), the slow-4 band (0.027-0.073 Hz), and the slow-5 band (0.01-0.027 Hz). Moreover, we extracted ALFF and fALFF values in clusters with significant differences to evaluate the SMT effect.ResultsCompared with HCs, the LDH patients before SMT (LDH-pre) exhibited increased fALFF in right lingual gyri in the conventional band, and showed increased fALFF in left Cerebelum_Crus1 in the slow-4 band. We further examined the abnormal brain activities changes before and after the SMT intervention. The ALFF and fALFF values of LDH-pre group were higher than those of the HCs and LDH-pos groups. After SMT, the increased ALFF and fALFF values were suppressed for patients in conventional band and slow-4 band.ConclusionThe present study characterized the altered regional patterns in spontaneous neural activity in patients with LDH. Meanwhile, SMT is an effective treatment of LDH, and we supposed that it might have been involved in modulating dysfunctional brain regions which are important for the processing of pain. The findings of the current study may provide new insights to understand pathological mechanism of LDH.

Project description:ContextLumbar diskectomy is the most common surgical procedure performed for back and leg symptoms in US patients, but the efficacy of the procedure relative to nonoperative care remains controversial.ObjectiveTo assess the efficacy of surgery for lumbar intervertebral disk herniation.Design, setting, and patientsThe Spine Patient Outcomes Research Trial, a randomized clinical trial enrolling patients between March 2000 and November 2004 from 13 multidisciplinary spine clinics in 11 US states. Patients were 501 surgical candidates (mean age, 42 years; 42% women) with imaging-confirmed lumbar intervertebral disk herniation and persistent signs and symptoms of radiculopathy for at least 6 weeks.InterventionsStandard open diskectomy vs nonoperative treatment individualized to the patient.Main outcome measuresPrimary outcomes were changes from baseline for the Medical Outcomes Study 36-item Short-Form Health Survey bodily pain and physical function scales and the modified Oswestry Disability Index (American Academy of Orthopaedic Surgeons MODEMS version) at 6 weeks, 3 months, 6 months, and 1 and 2 years from enrollment. Secondary outcomes included sciatica severity as measured by the Sciatica Bothersomeness Index, satisfaction with symptoms, self-reported improvement, and employment status.ResultsAdherence to assigned treatment was limited: 50% of patients assigned to surgery received surgery within 3 months of enrollment, while 30% of those assigned to nonoperative treatment received surgery in the same period. Intent-to-treat analyses demonstrated substantial improvements for all primary and secondary outcomes in both treatment groups. Between-group differences in improvements were consistently in favor of surgery for all periods but were small and not statistically significant for the primary outcomes.ConclusionsPatients in both the surgery and the nonoperative treatment groups improved substantially over a 2-year period. Because of the large numbers of patients who crossed over in both directions, conclusions about the superiority or equivalence of the treatments are not warranted based on the intent-to-treat analysis.Trial registrationclinicaltrials.gov Identifier: NCT00000410.