A deep-learning model for identifying fresh vertebral compression fractures on digital radiography

Fully Automatic Deep Learning Model for Spine Refracture in Patients with OVCF: A Multi-Center Study

BACKGROUND

The reaserch of artificial intelligence (AI) model for predicting spinal refracture is limited to bone mineral density, X-ray and some conventional laboratory indicators, which has its own limitations. Besides, it lacks specific indicators related to osteoporosis and imaging factors that can better reflect bone quality, such as computed tomography (CT).

OBJECTIVE

To construct a novel predicting model based on bone turn-over markers and CT to identify patients who were more inclined to suffer spine refracture.

METHODS

CT images and clinical information of 383 patients (training set = 240 cases of osteoporotic vertebral compression fractures (OVCF), validation set = 63, test set = 80) were retrospectively collected from January 2015 to October 2022 at three medical centers. The U-net model was adopted to automatically segment ROI. Three-dimensional (3D) cropping of all spine regions was used to achieve the final ROI regions including 3D_Full and 3D_RoiOnly. We used the Densenet 121-3D model to model the cropped region and simultaneously build a T-NIPT prediction model. Diagnostics of deep learning models were assessed by constructing ROC curves. We generated calibration curves to assess the calibration performance. Additionally, decision curve analysis (DCA) was used to assess the clinical utility of the predictive models.

RESULTS

The performance of the test model is comparable to its performance on the training set (dice coefficients of 0.798, an mIOU of 0.755, an SA of 0.767, and an OS of 0.017). Univariable and multivariable analysis indicate that T_P1NT was an independent risk factor for refracture. The performance of predicting refractures in different ROI regions showed that 3D_Full model exhibits the highest calibration performance, with a Hosmer-Lemeshow goodness-of-fit (HL) test statistic exceeding 0.05. The analysis of the training and test sets showed that the 3D_Full model, which integrates clinical and deep learning results, demonstrated superior performance with significant improvement (p-value < 0.05) compared to using clinical features independently or using only 3D_RoiOnly.

CONCLUSION

T_P1NT was an independent risk factor of refracture. Our 3D-FULL model showed better performance in predicting high-risk population of spine refracture than other models and junior doctors do. This model can be applicable to real-world translation due to its automatic segmentation and detection.

Artificial Intelligence in Spinal Imaging and Patient Care: A Review of Recent Advances

Artificial intelligence (AI) is transforming spinal imaging and patient care through automated analysis and enhanced decision-making. This review presents a clinical task-based evaluation, highlighting the specific impact of AI techniques on different aspects of spinal imaging and patient care. We first discuss how AI can potentially improve image quality through techniques like denoising or artifact reduction. We then explore how AI enables efficient quantification of anatomical measurements, spinal curvature parameters, vertebral segmentation, and disc grading. This facilitates objective, accurate interpretation and diagnosis. AI models now reliably detect key spinal pathologies, achieving expert-level performance in tasks like identifying fractures, stenosis, infections, and tumors. Beyond diagnosis, AI also assists surgical planning via synthetic computed tomography generation, augmented reality systems, and robotic guidance. Furthermore, AI image analysis combined with clinical data enables personalized predictions to guide treatment decisions, such as forecasting spine surgery outcomes. However, challenges still need to be addressed in implementing AI clinically, including model interpretability, generalizability, and data limitations. Multicenter collaboration using large, diverse datasets is critical to advance the field further. While adoption barriers persist, AI presents a transformative opportunity to revolutionize spinal imaging workflows, empowering clinicians to translate data into actionable insights for improved patient care.

Deep learning-based automated high-accuracy location and identification of fresh vertebral compression fractures from spinal radiographs: a multicenter cohort study

Background

Digital radiography (DR) is a common and widely available examination. However, spinal DR cannot detect bone marrow edema, therefore, determining vertebral compression fractures (VCFs), especially fresh VCFs, remains challenging for clinicians.

Methods

We trained, validated, and externally tested the deep residual network (DRN) model that automated the detection and identification of fresh VCFs from spinal DR images. A total of 1,747 participants from five institutions were enrolled in this study and divided into the training cohort, validation cohort and external test cohorts (YHDH and BMUH cohorts). We evaluated the performance of DRN model based on the area under the receiver operating characteristic curve (AUC), feature attention maps, sensitivity, specificity, and accuracy. We compared it with five other deep learning models and validated and tested the model internally and externally and explored whether it remains highly accurate for an external test cohort. In addition, the influence of old VCFs on the performance of the DRN model was assessed.

Results

The AUC was 0.99, 0.89, and 0.88 in the validation, YHDH, and BMUH cohorts, respectively, for the DRN model for detecting and discriminating fresh VCFs. The accuracies were 81.45% and 72.90%, sensitivities were 84.75% and 91.43%, and specificities were 80.25% and 63.89% in the YHDH and BMUH cohorts, respectively. The DRN model generated correct activation on the fresh VCFs and accurate peak responses on the area of the target vertebral body parts and demonstrated better feature representation learning and classification performance. The AUC was 0.90 (95% confidence interval [CI] 0.84-0.95) and 0.84 (95% CI 0.72-0.93) in the non-old VCFs and old VCFs groups, respectively, in the YHDH cohort (p = 0.067). The AUC was 0.89 (95% CI 0.84-0.94) and 0.85 (95% CI 0.72-0.95) in the non-old VCFs and old VCFs groups, respectively, in the BMUH cohort (p = 0.051).

Conclusion

In present study, we developed the DRN model for automated diagnosis and identification of fresh VCFs from spinal DR images. The DRN model can provide interpretable attention maps to support the excellent prediction results, which is the key that most clinicians care about when using the model to assist decision-making.

Constructing a Deep Learning Radiomics Model Based on X-ray Images and Clinical Data for Predicting and Distinguishing Acute and Chronic Osteoporotic Vertebral Fractures: A Multicenter Study

RATIONALE AND OBJECTIVES

To construct and validate a deep learning radiomics (DLR) model based on X-ray images for predicting and distinguishing acute and chronic osteoporotic vertebral fractures (OVFs).

METHODS

A total of 942 cases (1076 vertebral bodies) with both vertebral X-ray examination and MRI scans were included in this study from three hospitals. They were divided into a training cohort (n = 712), an internal validation cohort (n = 178), an external validation cohort (n = 111), and a prospective validation cohort (n = 75). The ResNet-50 model architecture was used for deep transfer learning (DTL), with pre-training performed on RadImageNet and ImageNet datasets. DTL features and radiomics features were extracted from lateral X-ray images of OVFs patients and fused together. A logistic regression model with the least absolute shrinkage and selection operator was established, with MRI showing bone marrow edema as the gold standard for acute OVFs. The performance of the model was evaluated using receiver operating characteristic curves. Eight machine learning classification models were evaluated for their ability to distinguish between acute and chronic OVFs. The Nomogram was constructed by combining clinical baseline data to achieve visualized classification assessment. The predictive performance of the best RadImageNet model and ImageNet model was compared using the Delong test. The clinical value of the Nomogram was evaluated using decision curve analysis (DCA).

RESULTS

Pre-training resulted in 34 and 39 fused features after feature selection and fusion. The most effective machine learning algorithm in both DLR models was Light Gradient Boosting Machine. Using the Delong test, the area under the curve (AUC) for distinguishing between acute and chronic OVFs in the training cohort was 0.979 and 0.972 for the RadImageNet and ImageNet models, respectively, with no statistically significant difference between them (P = 0.235). In the internal validation cohort, external validation cohort, and prospective validation cohort, the AUCs for the two models were 0.967 vs 0.629, 0.886 vs 0.817, and 0.933 vs 0.661, respectively, with statistically significant differences in all comparisons (P < 0.05). The deep learning radiomics nomogram (DLRN) was constructed by combining the predictive model of RadImageNet with clinical baseline features, resulting in AUCs of 0.981, 0.974, 0.895, and 0.902 in the training cohort, internal validation cohort, external validation cohort, and prospective validation cohort, respectively. Using the Delong test, the AUCs for the fused feature model and the DLRN in the training cohort were 0.979 and 0.981, respectively, with no statistically significant difference between them (P = 0.169). In the internal validation cohort, external validation cohort, and prospective validation cohort, the AUCs for the two models were 0.967 vs 0.974, 0.886 vs 0.895, and 0.933 vs 0.902, respectively, with statistically significant differences in all comparisons (P < 0.05). The Nomogram showed a slight improvement in predictive performance in the internal and external validation cohort, but a slight decrease in the prospective validation cohort (0.933 vs 0.902). DCA showed that the Nomogram provided more benefits to patients compared to the DLR models.

CONCLUSION

Compared to the ImageNet model, the RadImageNet model has higher diagnostic value in distinguishing between acute and chronic OVFs. Furthermore, the diagnostic performance of the model is further improved when combined with clinical baseline features to construct the Nomogram.

The Classification of Lumbar Spondylolisthesis X-Ray Images Using Convolutional Neural Networks

We aimed to develop and validate a deep convolutional neural network (DCNN) model capable of accurately identifying spondylolysis or spondylolisthesis on lateral or dynamic X-ray images. A total of 2449 lumbar lateral and dynamic X-ray images were collected from two tertiary hospitals. These images were categorized into lumbar spondylolysis (LS), degenerative lumbar spondylolisthesis (DLS), and normal lumbar in a proportional manner. Subsequently, the images were randomly divided into training, validation, and test sets to establish a classification recognition network. The model training and validation process utilized the EfficientNetV2-M network. The model's ability to generalize was assessed by conducting a rigorous evaluation on an entirely independent test set and comparing its performance with the diagnoses made by three orthopedists and three radiologists. The evaluation metrics employed to assess the model's performance included accuracy, sensitivity, specificity, and F1 score. Additionally, the weight distribution of the network was visualized using gradient-weighted class activation mapping (Grad-CAM). For the doctor group, accuracy ranged from 87.9 to 90.0% (mean, 89.0%), precision ranged from 87.2 to 90.5% (mean, 89.0%), sensitivity ranged from 87.1 to 91.0% (mean, 89.2%), specificity ranged from 93.7 to 94.7% (mean, 94.3%), and F1 score ranged from 88.2 to 89.9% (mean, 89.1%). The DCNN model had accuracy of 92.0%, precision of 91.9%, sensitivity of 92.2%, specificity of 95.7%, and F1 score of 92.0%. Grad-CAM exhibited concentrations of highlighted areas in the intervertebral foraminal region. We developed a DCNN model that intelligently distinguished spondylolysis or spondylolisthesis on lumbar lateral or lumbar dynamic radiographs.

A Deep-Learning Model for Diagnosing Fresh Vertebral Fractures on Magnetic Resonance Images

BACKGROUND

The accurate diagnosis of fresh vertebral fractures (VFs) was critical to optimizing treatment outcomes. Existing studies, however, demonstrated insufficient accuracy, sensitivity, and specificity in detecting fresh fractures using magnetic resonance imaging (MRI), and fall short in localizing the fracture sites.

METHODS

This prospective study comprised 716 patients with fresh VFs. We obtained 849 Short TI Inversion Recovery (STIR) image slices for training and validation of the AI model. The AI models employed were yolov7 and resnet50, to detect fresh VFs.

RESULTS

The AI model demonstrated a diagnostic accuracy of 97.6% for fresh VFs, with a sensitivity of 98% and a specificity of 97%. The performance of the model displayed a high degree of consistency when compared to the evaluations by spine surgeons. In the external testing dataset, the model exhibited a classification accuracy of 92.4%, a sensitivity of 93%, and a specificity of 92%.

CONCLUSIONS

Our findings highlighted the potential of AI in diagnosing fresh VFs, offering an accurate and efficient way to aid physicians with diagnosis and treatment decisions.

Artificial intelligence in fracture detection with different image modalities and data types: A systematic review and meta-analysis

Artificial Intelligence (AI), encompassing Machine Learning and Deep Learning, has increasingly been applied to fracture detection using diverse imaging modalities and data types. This systematic review and meta-analysis aimed to assess the efficacy of AI in detecting fractures through various imaging modalities and data types (image, tabular, or both) and to synthesize the existing evidence related to AI-based fracture detection. Peer-reviewed studies developing and validating AI for fracture detection were identified through searches in multiple electronic databases without time limitations. A hierarchical meta-analysis model was used to calculate pooled sensitivity and specificity. A diagnostic accuracy quality assessment was performed to evaluate bias and applicability. Of the 66 eligible studies, 54 identified fractures using imaging-related data, nine using tabular data, and three using both. Vertebral fractures were the most common outcome (n = 20), followed by hip fractures (n = 18). Hip fractures exhibited the highest pooled sensitivity (92%; 95% CI: 87-96, p< 0.01) and specificity (90%; 95% CI: 85-93, p< 0.01). Pooled sensitivity and specificity using image data (92%; 95% CI: 90-94, p< 0.01; and 91%; 95% CI: 88-93, p < 0.01) were higher than those using tabular data (81%; 95% CI: 77-85, p< 0.01; and 83%; 95% CI: 76-88, p < 0.01), respectively. Radiographs demonstrated the highest pooled sensitivity (94%; 95% CI: 90-96, p < 0.01) and specificity (92%; 95% CI: 89-94, p< 0.01). Patient selection and reference standards were major concerns in assessing diagnostic accuracy for bias and applicability. AI displays high diagnostic accuracy for various fracture outcomes, indicating potential utility in healthcare systems for fracture diagnosis. However, enhanced transparency in reporting and adherence to standardized guidelines are necessary to improve the clinical applicability of AI. Review Registration: PROSPERO (CRD42021240359).

Artificial Intelligence-Assisted Detection of Osteoporotic Vertebral Fractures on Lateral Chest Radiographs in Post-Menopausal Women

Osteoporotic vertebral fractures (OVFs) are often not reported by radiologists on routine chest radiographs. This study aims to investigate the clinical value of a newly developed artificial intelligence (AI) tool, Ofeye 1.0, for automated detection of OVFs on lateral chest radiographs in post-menopausal women (>60 years) who were referred to undergo chest x-rays for other reasons. A total of 510 de-identified lateral chest radiographs from three clinical sites were retrieved and analysed using the Ofeye 1.0 tool. These images were then reviewed by a consultant radiologist with findings serving as the reference standard for determining the diagnostic performance of the AI tool for the detection of OVFs. Of all the original radiologist reports, missed OVFs were found in 28.8% of images but were detected using the AI tool. The AI tool demonstrated high specificity of 92.8% (95% CI: 89.6, 95.2%), moderate accuracy of 80.3% (95% CI: 76.3, 80.4%), positive predictive value (PPV) of 73.7% (95% CI: 65.2, 80.8%), and negative predictive value (NPV) of 81.5% (95% CI: 79, 83.8%), but low sensitivity of 49% (95% CI: 40.7, 57.3%). The AI tool showed improved sensitivity compared with the original radiologist reports, which was 20.8% (95% CI: 14.5, 28.4). The new AI tool can be used as a complementary tool in routine diagnostic reports for the reduction in missed OVFs in elderly women.

A Deep Learning-Based Model for Classifying Osteoporotic Lumbar Vertebral Fractures on Radiographs: A Retrospective Model Development and Validation Study

Early diagnosis and initiation of treatment for fresh osteoporotic lumbar vertebral fractures (OLVF) are crucial. Magnetic resonance imaging (MRI) is generally performed to differentiate between fresh and old OLVF. However, MRIs can be intolerable for patients with severe back pain. Furthermore, it is difficult to perform in an emergency. MRI should therefore only be performed in appropriately selected patients with a high suspicion of fresh fractures. As radiography is the first-choice imaging examination for the diagnosis of OLVF, improving screening accuracy with radiographs will optimize the decision of whether an MRI is necessary. This study aimed to develop a method to automatically classify lumbar vertebrae (LV) conditions such as normal, old, or fresh OLVF using deep learning methods with radiography. A total of 3481 LV images for training, validation, and testing and 662 LV images for external validation were collected. Visual evaluation by two radiologists determined the ground truth of LV diagnoses. Three convolutional neural networks were ensembled. The accuracy, sensitivity, and specificity were 0.89, 0.83, and 0.92 in the test and 0.84, 0.76, and 0.89 in the external validation, respectively. The results suggest that the proposed method can contribute to the accurate automatic classification of LV conditions on radiography.

The use of deep learning in medical imaging to improve spine care: A scoping review of current literature and clinical applications

Background

Artificial intelligence is a revolutionary technology that promises to assist clinicians in improving patient care. In radiology, deep learning (DL) is widely used in clinical decision aids due to its ability to analyze complex patterns and images. It allows for rapid, enhanced data, and imaging analysis, from diagnosis to outcome prediction. The purpose of this study was to evaluate the current literature and clinical utilization of DL in spine imaging.

Methods

This study is a scoping review and utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology to review the scientific literature from 2012 to 2021. A search in PubMed, Web of Science, Embased, and IEEE Xplore databases with syntax specific for DL and medical imaging in spine care applications was conducted to collect all original publications on the subject. Specific data was extracted from the available literature, including algorithm application, algorithms tested, database type and size, algorithm training method, and outcome of interest.

Results

A total of 365 studies (total sample of 232,394 patients) were included and grouped into 4 general applications: diagnostic tools, clinical decision support tools, automated clinical/instrumentation assessment, and clinical outcome prediction. Notable disparities exist in the selected algorithms and the training across multiple disparate databases. The most frequently used algorithms were U-Net and ResNet. A DL model was developed and validated in 92% of included studies, while a pre-existing DL model was investigated in 8%. Of all developed models, only 15% of them have been externally validated.

Conclusions

Based on this scoping review, DL in spine imaging is used in a broad range of clinical applications, particularly for diagnosing spinal conditions. There is a wide variety of DL algorithms, database characteristics, and training methods. Future studies should focus on external validation of existing models before bringing them into clinical use.

Performance of deep learning-based autodetection of arterial stenosis on head and neck CT angiography: an independent external validation study

PURPOSE

To externally validate the performance of automated stenosis detection on head and neck CT angiography (CTA) and investigate the impact factors using an independent bi-center dataset with digital subtraction angiography (DSA) as the ground truth.

MATERIAL AND METHODS

Patients who underwent head and neck CTA and DSA between January 2019 and December 2021 were retrospectively included. The degree of stenosis was automatically evaluated using CerebralDoc based on CTA. The performance of CerebralDoc across levels (per-patient, per-region, per-vessel, and per-segment) and thresholds (≥ 50%, ≥ 70%, and = 100%) was evaluated. Logistic regression was performed to identify independent factors associated with false negative results.

RESULTS

296 patients were analyzed. Specificity across levels and thresholds was high, exceeding 92%. The area under the curve ranged from poor (0.615, 95% CI: 0.544, 0.686; at the region-based analysis for stenosis ≥ 70%) to excellent (0.945, 95% CI: 0.905, 0.985; at the patient-based analysis for stenosis ≥ 50%). Sensitivity ranged from 0.714 (95% CI: 0.675, 0.750) at the segment-based analysis for stenosis ≥ 70% to 0.895 (95% CI: 0.849, 0.919) at the patient-based analysis for stenosis ≥ 50%. The multiple logistic regression analysis revealed that false negative results were primarily more likely to specific stenosis locations (particularly the M2 segment and skull base segment of the internal carotid artery) and occlusion.

CONCLUSIONS

CerebralDoc has the potential to automated stenosis detection on head and neck CTA, but further efforts are needed to optimize its performance.

Benign vs malignant vertebral compression fractures with MRI: a comparison between automatic deep learning network and radiologist's assessment

OBJECTIVE

To test the diagnostic performance of a deep-learning Two-Stream Compare and Contrast Network (TSCCN) model for differentiating benign and malignant vertebral compression fractures (VCFs) based on MRI.

METHODS

We tested a deep-learning system in 123 benign and 86 malignant VCFs. The median sagittal T1-weighted images (T1WI), T2-weighted images with fat suppression (T2WI-FS), and a combination of both (thereafter, T1WI/T2WI-FS) were used to validate TSCCN. The receiver operator characteristic (ROC) curve was analyzed to evaluate the performance of TSCCN. The accuracy, sensitivity, and specificity of TSCCN in differentiating benign and malignant VCFs were calculated and compared with radiologists' assessments. Intraclass correlation coefficients (ICCs) were tested to find intra- and inter-observer agreement of radiologists in differentiating malignant from benign VCFs.

RESULTS

The AUC of the ROC plots of TSCCN according to T1WI, T2WI-FS, and T1WI/T2WI-FS images were 99.2%, 91.7%, and 98.2%, respectively. The accuracy of T1W, T2WI-FS, and T1W/T2WI-FS based on TSCCN was 95.2%, 90.4%, and 96.2%, respectively, greater than that achieved by radiologists. Further, the specificity of T1W, T2WI-FS, and T1W/T2WI-FS based on TSCCN was higher at 98.4%, 94.3%, and 99.2% than that achieved by radiologists. The intra- and inter-observer agreements of radiologists were 0.79-0.85 and 0.79-0.80 for T1WI, 0.65-0.72 and 0.70-0.74 for T2WI-FS, and 0.83-0.88 and 0.83-0.84 for T1WI/T2WI-FS.

CONCLUSION

The TSCCN model showed better diagnostic performance than radiologists for automatically identifying benign or malignant VCFs, and is a potentially helpful tool for future clinical application.

CLINICAL RELEVANCE STATEMENT

TSCCN-assisted MRI has shown superior performance in distinguishing benign and malignant vertebral compression fractures compared to radiologists. This technology has the value to enhance diagnostic accuracy, sensitivity, and specificity. Further integration into clinical practice is required to optimize patient management.

KEY POINTS

• The Two-Stream Compare and Contrast Network (TSCCN) model showed better diagnostic performance than radiologists for identifying benign vs malignant vertebral compression fractures. • The processing of TSCCN is fast and stable, better than the subjective evaluation by radiologists in diagnosing vertebral compression fractures. • The TSCCN model provides options for developing a fully automated, streamlined artificial intelligence diagnostic tool.

Radiographic assessment of acute vs chronic vertebral compression fractures

PURPOSE

Distinguishing between acute and chronic vertebral compression fractures typically requires advanced imaging techniques such as magnetic resonance imaging (MRI). Recognizing specific radiographic findings associated with fracture acuity may improve the accuracy of radiographic assessment.

METHODS

Patients with compression fractures that had both radiographic and MRI studies of the lumbar spine within a 30-day time frame were retrospectively reviewed. MRI studies were used to determine compression fracture acuity. Radiographs were interpreted by a separate group of radiologists blinded to the MRI results. Radiographic findings of endplate osteophyte, subendplate density, subendplate cleft, and subendplate cyst were recorded as was the overall impression of fracture acuity.

RESULTS

Sensitivity and specificity for radiographic reporting of acute fracture were 0.52 (95% CI: 0.42, 0.61) and 0.95 (95% CI: 0.93, 0.97) respectively. For chronic fractures, the sensitivity and specificity were 0.52 (95% CI: 0.41, 0.63) and 0.94 (95% CI: 0.92, 0.96). The radiographic presence of a subendplate cleft increased the odds of a fracture being acute by a factor of 1.75 (95% CI: 1.09, 2.81; P = 0.0202). The radiographic presence of subendplate density increased the odds of a fracture being acute by a factor of 1.78 (95% CI: 1.21, 2.63; P = 0.0037). The presence of an endplate osteophyte or subendplate cyst was not significantly associated with fracture acuity.

CONCLUSION

Radiographs are relatively insensitive in distinguishing between acute and chronic lumbar compression fractures but the presence of a subendplate cleft or subendplate density increases the likelihood that a given fracture is acute.

Deep learning-based artificial intelligence model for classification of vertebral compression fractures: A multicenter diagnostic study

OBJECTIVE

To develop and validate an artificial intelligence diagnostic system based on X-ray imaging data for diagnosing vertebral compression fractures (VCFs).

METHODS

In total, 1904 patients who underwent X-ray at four independent hospitals were retrospectively (n=1847) and prospectively (n=57) enrolled. The participants were separated into a development cohort, a prospective test cohort and three external test cohorts. The proposed model used a transfer learning method based on the ResNet-18 architecture. The diagnostic performance of the model was evaluated using receiver operating characteristic curve (ROC) analysis and validated using a prospective validation set and three external sets. The performance of the model was compared with three degrees of musculoskeletal expertise: expert, competent, and trainee.

RESULTS

The diagnostic accuracy for identifying compression fractures was 0.850 in the testing set, 0.829 in the prospective set, and ranged from 0.757 to 0.832 in the three external validation sets. In the human and deep learning (DL) collaboration dataset, the area under the ROC curves(AUCs) in acute, chronic, and pathological compression fractures were as follows: 0.780, 0.809, 0.734 for the DL model; 0.573, 0.618, 0.541 for the trainee radiologist; 0.701, 0.782, 0.665 for the competent radiologist; 0.707,0.732, 0.667 for the expert radiologist; 0.722, 0.744, 0.610 for the DL and trainee; 0.767, 0.779, 0.729 for the DL and competent; 0.801, 0.825, 0.751 for the DL and expert radiologist.

CONCLUSIONS

Our study offers a high-accuracy multi-class deep learning model which could assist community-based hospitals in improving the diagnostic accuracy of VCFs.

Automated detection and classification of acute vertebral body fractures using a convolutional neural network on computed tomography

BACKGROUND

Acute vertebral fracture is usually caused by low-energy injury with osteoporosis and high-energy trauma. The AOSpine thoracolumbar spine injury classification system (AO classification) plays an important role in the diagnosis and treatment of the disease. The diagnosis and description of vertebral fractures according to the classification scheme requires a great deal of time and energy for radiologists.

PURPOSE

To design and validate a multistage deep learning system (multistage AO system) for the automatic detection, localization and classification of acute thoracolumbar vertebral body fractures according to AO classification on computed tomography.

MATERIALS AND METHODS

The CT images of 1,217 patients who came to our hospital from January 2015 to December 2019 were collected retrospectively. The fractures were marked and classified by 2 junior radiology residents according to the type A standard in the AO classification. Marked fracture sites included the upper endplate, lower endplate and posterior wall. When there were inconsistent opinions on classification labels, the final result was determined by a director radiologist. We integrated different networks into different stages of the overall framework. U-net and a graph convolutional neural network (U-GCN) are used to realize the location and classification of the thoracolumbar spine. Next, a classification network is used to detect whether the thoracolumbar spine has a fracture. In the third stage, we detect fractures in different parts of the thoracolumbar spine by using a multibranch output network and finally obtain the AO types.

RESULTS

The mean age of the patients was 61.87 years with a standard deviation of 17.04 years, consisting of 760 female patients and 457 male patients. On vertebrae level, sensitivity for fracture detection was 95.23% in test dataset, with an accuracy of 97.93% and a specificity of 98.35%. For the classification of vertebral body fractures, the balanced accuracy was 79.56%, with an AUC of 0.904 for type A1, 0.945 for type A2, 0.878 for type A3 and 0.942 for type A4.

CONCLUSION

The multistage AO system can automatically detect and classify acute vertebral body fractures in the thoracolumbar spine on CT images according to AO classification with high accuracy.

Performance of a deep convolutional neural network for MRI-based vertebral body measurements and insufficiency fracture detection

OBJECTIVES

The aim is to validate the performance of a deep convolutional neural network (DCNN) for vertebral body measurements and insufficiency fracture detection on lumbar spine MRI.

METHODS

This retrospective analysis included 1000 vertebral bodies in 200 patients (age 75.2 ± 9.8 years) who underwent lumbar spine MRI at multiple institutions. 160/200 patients had ≥ one vertebral body insufficiency fracture, 40/200 had no fracture. The performance of the DCNN and that of two fellowship-trained musculoskeletal radiologists in vertebral body measurements (anterior/posterior height, extent of endplate concavity, vertebral angle) and evaluation for insufficiency fractures were compared. Statistics included (a) interobserver reliability metrics using intraclass correlation coefficient (ICC), kappa statistics, and Bland-Altman analysis, and (b) diagnostic performance metrics (sensitivity, specificity, accuracy). A statistically significant difference was accepted if the 95% confidence intervals did not overlap.

RESULTS

The inter-reader agreement between radiologists and the DCNN was excellent for vertebral body measurements, with ICC values of > 0.94 for anterior and posterior vertebral height and vertebral angle, and good to excellent for superior and inferior endplate concavity with ICC values of 0.79-0.85. The performance of the DCNN in fracture detection yielded a sensitivity of 0.941 (0.903-0.968), specificity of 0.969 (0.954-0.980), and accuracy of 0.962 (0.948-0.973). The diagnostic performance of the DCNN was independent of the radiological institution (accuracy 0.964 vs. 0.960), type of MRI scanner (accuracy 0.957 vs. 0.964), and magnetic field strength (accuracy 0.966 vs. 0.957).

CONCLUSIONS

A DCNN can achieve high diagnostic performance in vertebral body measurements and insufficiency fracture detection on heterogeneous lumbar spine MRI.

KEY POINTS

• A DCNN has the potential for high diagnostic performance in measuring vertebral bodies and detecting insufficiency fractures of the lumbar spine.

A software program for automated compressive vertebral fracture detection on elderly women's lateral chest radiograph: Ofeye 1.0

Background

Because osteoporotic vertebral fracture (OVF) on chest radiographs is commonly missed in radiological reports, we aimed to develop a software program which offers automated detection of compressive vertebral fracture (CVF) on lateral chest radiographs, and which emphasizes CVF detection specificity with a low false positivity rate.

Methods

For model training, we retrieved 3,991 spine radiograph cases and 1,979 chest radiograph cases from 16 sources, with among them in total 1,404 cases had OVF. For model testing, we retrieved 542 chest radiograph cases and 162 spine radiograph cases from four independent clinics, with among them 215 cases had OVF. All cases were female subjects, and except for 31 training data cases which were spine trauma cases, all the remaining cases were post-menopausal women. Image data included DICOM (Digital Imaging and Communications in Medicine) format, hard film scanned PNG (Portable Network Graphics) format, DICOM exported PNG format, and PACS (Picture Archiving and Communication System) downloaded resolution reduced DICOM format. OVF classification included: minimal and mild grades with <20% or ≥20–25% vertebral height loss respectively, moderate grade with ≥25–40% vertebral height loss, severe grade with ≥40%–2/3 vertebral height loss, and collapsed grade with ≥2/3 vertebral height loss. The CVF detection base model was mainly composed of convolution layers that include convolution kernels of different sizes, pooling layers, up-sampling layers, feature merging layers, and residual modules. When the model loss function could not be further decreased with additional training, the model was considered to be optimal and termed ‘base-model 1.0’. A user-friendly interface was also developed, with the synthesized software termed ‘Ofeye 1.0’.

Results

Counting cases and with minimal and mild OVFs included, base-model 1.0 demonstrated a specificity of 97.1%, a sensitivity of 86%, and an accuracy of 93.9% for the 704 testing cases. In total, 33 OVFs in 30 cases had a false negative reading, which constituted a false negative rate of 14.0% (30/215) by counting all OVF cases. Eighteen OVFs in 15 cases had OVFs of ≥ moderate grades missed, which constituted a false negative rate of 7.0% (15/215, i.e., sensitivity 93%) if only counting cases with ≥ moderate grade OVFs missed. False positive reading was recorded in 13 vertebrae in 13 cases (one vertebra in each case), which constituted a false positivity rate of 2.7% (13/489). These vertebrae with false positivity labeling could be readily differentiated from a true OVF by a human reader. The software Ofeye 1.0 allows ‘batch processing’, for example, 100 radiographs can be processed in a single operation. This software can be integrated into hospital PACS, or installed in a standalone personal computer.

Conclusions

A user-friendly software program was developed for CVF detection on elderly women’s lateral chest radiographs. It has an overall low false positivity rate, and for moderate and severe CVFs an acceptably low false negativity rate. The integration of this software into radiological practice is expected to improve osteoporosis management for elderly women.

Research hotspots and emerging trends of deep learning applications in orthopedics: A bibliometric and visualized study

Background

As a research hotspot, deep learning has been continuously combined with various research fields in medicine. Recently, there is a growing amount of deep learning-based researches in orthopedics. This bibliometric analysis aimed to identify the hotspots of deep learning applications in orthopedics in recent years and infer future research trends.

Methods

We screened global publication on deep learning applications in orthopedics by accessing the Web of Science Core Collection. The articles and reviews were collected without language and time restrictions. Citespace was applied to conduct the bibliometric analysis of the publications.

Results

A total of 822 articles and reviews were finally retrieved. The analysis showed that the application of deep learning in orthopedics has great prospects for development based on the annual publications. The most prolific country is the USA, followed by China. University of California San Francisco, and Skeletal Radiology are the most prolific institution and journal, respectively. LeCun Y is the most frequently cited author, and Nature has the highest impact factor in the cited journals. The current hot keywords are convolutional neural network, classification, segmentation, diagnosis, image, fracture, and osteoarthritis. The burst keywords are risk factor, identification, localization, and surgery. The timeline viewer showed two recent research directions for bone tumors and osteoporosis.

Conclusion

Publications on deep learning applications in orthopedics have increased in recent years, with the USA being the most prolific. The current research mainly focused on classifying, diagnosing and risk predicting in osteoarthritis and fractures from medical images. Future research directions may put emphasis on reducing intraoperative risk, predicting the occurrence of postoperative complications, screening for osteoporosis, and identification and classification of bone tumors from conventional imaging.

Image Quality Control in Lumbar Spine Radiography Using Enhanced U-Net Neural Networks

Purpose

To standardize the radiography imaging procedure, an image quality control framework using the deep learning technique was developed to segment and evaluate lumbar spine x-ray images according to a defined quality control standard.

Materials and Methods

A dataset comprising anteroposterior, lateral, and oblique position lumbar spine x-ray images from 1,389 patients was analyzed in this study. The training set consisted of digital radiography images of 1,070 patients (800, 798, and 623 images of the anteroposterior, lateral, and oblique position, respectively) and the validation set included 319 patients (200, 205, and 156 images of the anteroposterior, lateral, and oblique position, respectively). The quality control standard for lumbar spine x-ray radiography in this study was defined using textbook guidelines of as a reference. An enhanced encoder-decoder fully convolutional network with U-net as the backbone was implemented to segment the anatomical structures in the x-ray images. The segmentations were used to build an automatic assessment method to detect unqualified images. The dice similarity coefficient was used to evaluate segmentation performance.

Results

The dice similarity coefficient of the anteroposterior position images ranged from 0.82 to 0.96 (mean 0.91 ± 0.06); the dice similarity coefficient of the lateral position images ranged from 0.71 to 0.95 (mean 0.87 ± 0.10); the dice similarity coefficient of the oblique position images ranged from 0.66 to 0.93 (mean 0.80 ± 0.14). The accuracy, sensitivity, and specificity of the assessment method on the validation set were 0.971-0.990 (mean 0.98 ± 0.10), 0.714-0.933 (mean 0.86 ± 0.13), and 0.995-1.000 (mean 0.99 ± 0.12) for the three positions, respectively.

Conclusion

This deep learning-based algorithm achieves accurate segmentation of lumbar spine x-ray images. It provides a reliable and efficient method to identify the shape of the lumbar spine while automatically determining the radiographic image quality.