Segmentation Of Vertebral Bodies Research Articles

Using an Ensemble of Segmentation Methods to Detect Vertebral Bodies on Radiographs.

Vertebral compression fractures may indicate osteoporosis but are underdiagnosed and underreported by radiologists. We have developed an ensemble of vertebral body (VB) segmentation models for lateral radiographs as a critical component of an automated, opportunistic screening tool. Our goal is to detect the approximate location of thoracic and lumbar VBs, including fractured vertebra, on lateral radiographs. The Osteoporotic Fractures in Men Study (MrOS) data set includes spine radiographs of 5994 men aged ≥65 years from 6 clinical centers. Two segmentation models, U-Net and Mask-RCNN (Region-based Convolutional Neural Network), were independently trained on the MrOS data set retrospectively, and an ensemble was created by combining them. Primary performance metrics for VB detection success included precision, recall, and F1 score for object detection on a held-out test set. Intersection over union (IoU) and Dice coefficient were also calculated as secondary metrics of performance for the test set. A separate external data set from a quaternary health care enterprise was acquired to test generalizability, comprising diagnostic clinical radiographs from men and women aged ≥65 years. The trained models achieved F1 score of U-Net = 83.42%, Mask-RCNN = 86.30%, and ensemble = 88.34% in detecting all VBs, and F1 score of U-Net = 87.88%, Mask-RCNN = 92.31%, and ensemble = 97.14% in detecting severely fractured vertebrae. The trained models achieved an average IoU per VB of 0.759 for U-Net and 0.709 for Mask-RCNN. The trained models achieved F1 score of U-Net = 81.11%, Mask-RCNN = 79.24%, and ensemble = 87.72% in detecting all VBs in the external data set. An ensemble model combining predictions from U-Net and Mask-RCNN resulted in the best performance in detecting VBs on lateral radiographs and generalized well to an external data set. This model could be a key component of a pipeline to detect fractures on all vertebrae in a radiograph in an automated, opportunistic screening tool under development.

A 3D-CT Study of the Cortical Bone Trajectory Screw Placement Parameters Based on Lumbar CT.

The cortical bone trajectory (CBT) technology is an effective substitute for traditional pedicle screw (PS) technology. However, there is still controversy about the CBT screw technology placement strategy. The objective of this study was to simulate cortical screw placement with the help of three-dimensional (3D) software, to discuss the differences in screws between genders and vertebral segments, and to explore a safer and more efficient strategy for cortical screw placement. Mimics Medical software was used to construct a 3D model of the lumbar spine, and the placement of CBT screws was simulated. The volume of each vertebral body from L1 to L5, the pedicle isthmus height (IH), the pedicle isthmus width (IW), and the sagittal vertebral distance (SAVD) were measured. The transverse distance (TD) and the longitudinal distance (LD) between the ideal starting point (SP) and the clinical SP (the intersection Q of the midline of the superior articular process and the horizontal line 1 mm below the transverse process) were measured. The cephalad angle (CA), lateral angle (LA), maximum screw diameter (MSD), maximum screw length (MSL) of each trajectory of the L1 to L5 vertebral bodies, and the percentage of the screw insertion depth (PSID) into the vertebral body were measured. Data were statistically analyzed using Student's t-test, one-way analysis of variance (ANOVA), and Tukey's test. Vertebral anatomical parameters and CBT screw parameters differed between males and females. Female patients had lower IH, IW, SAVD, CA, LA, MSD, and MSL than males. IH was greatest in L1 (male, 17.81 mm; female, 16.12 mm) and the smallest in L5 (male, 14.11 mm; female, 13.05 mm). IW was smallest in L1 (male, 8.89 mm; female, 7.37 mm) and greatest in L5 (male, 16.59 mm; female, 15.43 mm). The MSD of males was smallest in L1 (6.05 mm) and greatest in L3 (7.06 mm); the MSD of females was smallest in L1 (5.13 mm) and greatest in L4 (6.64 mm). MSL was greatest at L3 (male, 33.63 mm; female, 32.28 mm) and smallest at L5 (male, 31.25 mm; female, 29.97 mm). CA was smallest in L1 (male, 22.80°; female, 21.92°) and greatest in L3 (male, 25.29°; female, 24.33°). LA was smallest in L1 (male 12.37°, female 11.27°) and greatest in L5 (male 13.56°, female 12.96°). Among the males, TD was smallest at L1 (-0.51 mm) and greatest at L5 (1.37 mm), while LD was greatest at L2 (3.46 mm) and smallest at L5 (2.40 mm). In females, TD was greatest at L1 (0.12 mm) and smallest at L3 (-0.51 mm), while LD was greatest at L1 (3.69 mm) and smallest at L5 (2.08 mm). In the overall sample, the incidence of SAVD and PSID gradually increased from L1 to L5. The optimal screw placement strategy for CBT screws varies significantly according to sex and vertebral body segments, particularly noting the specificity of screw placement at L5. The optimal screw placement strategy should be selected based on the patient's sex and segment characteristics before surgery to maximize the safety and accuracy of CBT screw placement.

Development of a deep learning model for detecting lumbar vertebral fractures on CT images: An external validation

ObjectiveTo develop and externally validate a binary classification model for lumbar vertebral body fractures based on CT images using deep learning methods. MethodsThis study involved data collection from two hospitals for AI model training and external validation. In Cohort A from Hospital 1, CT images from 248 patients, comprising 1508 vertebrae, revealed that 20.9% had fractures (315 vertebrae) and 79.1% were non-fractured (1193 vertebrae). In Cohort B from Hospital 2, CT images from 148 patients, comprising 887 vertebrae, indicated that 14.8% had fractures (131 vertebrae) and 85.2% were non-fractured (756 vertebrae). The AI model for lumbar spine fractures underwent two stages: vertebral body segmentation and fracture classification. The first stage utilized a 3D V-Net convolutional deep neural network, which produced a 3D segmentation map. From this map, region of each vertebra body were extracted and then input into the second stage of the algorithm. The second stage employed a 3D ResNet convolutional deep neural network to classify each proposed region as positive (fractured) or negative (not fractured). ResultsThe AI model’s accuracy for detecting vertebral fractures in Cohort A’s training set (n = 1199), validation set (n = 157), and test set (n = 152) was 100.0 %, 96.2 %, and 97.4 %, respectively. For Cohort B (n = 148), the accuracy was 96.3 %. The area under the receiver operating characteristic curve (AUC-ROC) values for the training, validation, and test sets of Cohort A, as well as Cohort B, and their 95 % confidence intervals (CIs) were as follows: 1.000 (1.000, 1.000), 0.978 (0.944, 1.000), 0.986 (0.969, 1.000), and 0.981 (0.970, 0.992). The area under the precision-recall curve (AUC-PR) values were 1.000 (0.996, 1.000), 0.964 (0.927, 0.985), 0.907 (0.924, 0.984), and 0.890 (0.846, 0.971), respectively. According to the DeLong test, there was no significant difference in the AUC-ROC values between the test set of Cohort A and Cohort B, both for the overall data and for each specific vertebral location (all P>0.05). ConclusionThe developed model demonstrates promising diagnostic accuracy and applicability for detecting lumbar vertebral fractures.

Artificial Intelligence Measurement of Preoperative Radiographs in Adolescent Idiopathic Scoliosis Based on Multiple-View Semantic Segmentation.

Cross-sectional study. Imaging classification of adolescent idiopathic scoliosis (AIS) is directly related to the surgical strategy, but the artificial classification is complex and depends on doctors' experience. This study investigated deep learning-based automated classification methods (DL group) for AIS and validated the consistency of machine classification and manual classification (M group). A total of 506 cases (81 males and 425 females) and 1812 AIS full spine images in the anteroposterior (AP), lateral (LAT), left bending (LB) and right bending (RB) positions were retrospectively used for training. The mean age was 13.6 ± 1.8. The mean maximum Cobb angle was 46.8 ± 12.0. U-Net semantic segmentation neural network technology and deep learning methods were used to automatically segment and establish the alignment relationship between multiple views of the spine, and to extract spinal features such as the Cobb angle. The type of each test case was automatically calculated according to Lenke's rule. An additional 107 cases of adolescent idiopathic scoliosis imaging were prospectively used for testing. The consistency of the DL group and M group was compared. Automatic vertebral body segmentation and recognition, multi-view alignment of the spine and automatic Cobb angle measurement were implemented. Compare to the M group, the consistency of the DL group was significantly higher in 3 aspects: type of lateral convexity (0.989 vs 0.566), lumbar curvature modifier (0.932 vs 0.738), and sagittal plane modifier (0.987 vs 0.522). Deep learning enables automated Cobb angle measurement and automated Lenke classification of idiopathic scoliosis whole spine radiographs with higher consistency than manual measurement classification.

Sagittal balance parameters measurement on cervical spine MR images based on superpixel segmentation.

Introduction: Magnetic Resonance Imaging (MRI) is essential in diagnosing cervical spondylosis, providing detailed visualization of osseous and soft tissue structures in the cervical spine. However, manual measurements hinder the assessment of cervical spine sagittal balance, leading to time-consuming and error-prone processes. This study presents the Pyramid DBSCAN Simple Linear Iterative Cluster (PDB-SLIC), an automated segmentation algorithm for vertebral bodies in T2-weighted MR images, aiming to streamline sagittal balance assessment for spinal surgeons. Method: PDB-SLIC combines the SLIC superpixel segmentation algorithm with DBSCAN clustering and underwent rigorous testing using an extensive dataset of T2-weighted mid-sagittal MR images from 4,258 patients across ten hospitals in China. The efficacy of PDB-SLIC was compared against other algorithms and networks in terms of superpixel segmentation quality and vertebral body segmentation accuracy. Validation included a comparative analysis of manual and automated measurements of cervical sagittal parameters and scrutiny of PDB-SLIC's measurement stability across diverse hospital settings and MR scanning machines. Result: PDB-SLIC outperforms other algorithms in vertebral body segmentation quality, with high accuracy, recall, and Jaccard index. Minimal error deviation was observed compared to manual measurements, with correlation coefficients exceeding 95%. PDB-SLIC demonstrated commendable performance in processing cervical spine T2-weighted MR images from various hospital settings, MRI machines, and patient demographics. Discussion: The PDB-SLIC algorithm emerges as an accurate, objective, and efficient tool for evaluating cervical spine sagittal balance, providing valuable assistance to spinal surgeons in preoperative assessment, surgical strategy formulation, and prognostic inference. Additionally, it facilitates comprehensive measurement of sagittal balance parameters across diverse patient cohorts, contributing to the establishment of normative standards for cervical spine MR imaging.

Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs.

This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics-compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)-from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

LumVertCancNet: A novel 3D lumbar vertebral body cancellous bone location and segmentation method based on hybrid Swin-transformer

Lumbar vertebral body cancellous bone location and segmentation is crucial in an automated lumbar spine processing pipeline. Accurate and reliable analysis of lumbar spine image is expected to advantage practical medical diagnosis and population-based analysis of bone strength. However, the design of automated algorithms for lumbar spine processing is demanding due to significant anatomical variations and scarcity of publicly available data. In recent years, convolutional neural network (CNN) and vision transformers (Vits) have been the de facto standard in medical image segmentation. Although adept at capturing global features, the inherent bias of locality and weight sharing of CNN constrains its capacity to model long-range dependency. In contrast, Vits excel at long-range dependency modeling, but they may not generalize well with limited datasets due to the lack of inductive biases inherent to CNN. In this paper, we propose a deep learning-based two-stage coarse-to-fine solution to address the problem of automatic location and segmentation of lumbar vertebral body cancellous bone. Specifically, in the first stage, a Swin-transformer based model is applied to predict the heatmap of lumbar vertebral body centroids. Considering the characteristic anatomical structure of lumbar spine, we propose a novel loss function called LumAnatomy loss, which enforces the order and bend of the predicted vertebral body centroids. To inherit the excellence of CNN and Vits while preventing their respective limitations, in the second stage, we propose an encoder–decoder network to segment the identified lumbar vertebral body cancellous bone, which consists of two parallel encoders, i.e., a Swin-transformer encoder and a CNN encoder. To enhance the combination of CNNs and Vits, we propose a novel multi-scale attention feature fusion module (MSA-FFM), which address issues that arise when fusing features given at different encoders. To tackle the issue of lack of data, we raise the first large-scale lumbar vertebral body cancellous bone segmentation dataset called LumVBCanSeg containing a total of 185 CT scans annotated at voxel level by 3 physicians. Extensive experimental results on the LumVBCanSeg dataset demonstrate the proposed algorithm outperform other state-of-the-art medical image segmentation methods. The data is publicly available at: https://zenodo.org/record/8181250. The implementation of the proposed method is available at: https://github.com/sia405yd/LumVertCancNet.

VertXNet: an ensemble method for vertebral body segmentation and identification from cervical and lumbar spinal X-rays

Accurate annotation of vertebral bodies is crucial for automating the analysis of spinal X-ray images. However, manual annotation of these structures is a laborious and costly process due to their complex nature, including small sizes and varying shapes. To address this challenge and expedite the annotation process, we propose an ensemble pipeline called VertXNet. This pipeline currently combines two segmentation mechanisms, semantic segmentation using U-Net, and instance segmentation using Mask R-CNN, to automatically segment and label vertebral bodies in lateral cervical and lumbar spinal X-ray images. VertXNet enhances its effectiveness by adopting a rule-based strategy (termed the ensemble rule) for effectively combining segmentation outcomes from U-Net and Mask R-CNN. It determines vertebral body labels by recognizing specific reference vertebral instances, such as cervical vertebra 2 (‘C2’) in cervical spine X-rays and sacral vertebra 1 (‘S1’) in lumbar spine X-rays. Those references are commonly relatively easy to identify at the edge of the spine. To assess the performance of our proposed pipeline, we conducted evaluations on three spinal X-ray datasets, including two in-house datasets and one publicly available dataset. The ground truth annotations were provided by radiologists for comparison. Our experimental results have shown that the proposed pipeline outperformed two state-of-the-art (SOTA) segmentation models on our test dataset with a mean Dice of 0.90, vs. a mean Dice of 0.73 for Mask R-CNN and 0.72 for U-Net. We also demonstrated that VertXNet is a modular pipeline that enables using other SOTA model, like nnU-Net to further improve its performance. Furthermore, to evaluate the generalization ability of VertXNet on spinal X-rays, we directly tested the pre-trained pipeline on two additional datasets. A consistently strong performance was observed, with mean Dice coefficients of 0.89 and 0.88, respectively. In summary, VertXNet demonstrated significantly improved performance in vertebral body segmentation and labeling for spinal X-ray imaging. Its robustness and generalization were presented through the evaluation of both in-house clinical trial data and publicly available datasets.

Biomechanical effects of endplate sagittal coverage change on cervical disc replacement: a finite element analysis.

In recent years, the number of artificial cervical disc replacements has increased, and paravertebral ectopic ossification is a common complication. Although the exact mechanism is not clear, some studies suggest that it is related to the concentration of tissue stress caused by incomplete coverage of the trailing edge of the endplate. Therefore, this study performed a quantitative analysis to compare the biomechanical effects of different sagittal distances at the posterior edge of the endplate of the upper and lower prosthesis on the cervical spine and to explore the mechanical response of incomplete coverage of the posterior edge of the endplate on the paravertebral tissues. A C2-C7 nonlinear finite element model of the cervical spine was established and validated. Based on the cervical spine model, cervical disc replacement surgery models were constructed with different distances of sagittal distance at the posterior edge of the upper prosthetic endplate (0, 1, 2, 3mm, respectively) and sagittal distance at the posterior edge of the lower prosthetic endplate (1, 2, 3mm, respectively). Each model was subjected to the same 1Nm torque and 73.6N driven compressive load. Range of motion (ROM), intervertebral disc pressure (IDP), facet joint force (FJF), and endplate stress were measured at the cervical surgical and other segments. Compared to the intact cervical spine model, the sagittal distance of the posterior edge of the prosthesis endplate at different distances increased the stress on the intervertebral disc and the capsular joint in the adjacent vertebral body segments to different degrees, especially in extension. In different directions of motion, the posterior margin sagittal distance of the posterior edge of the endplate of the lower prosthesis has a greater mechanical influence on the cervical spine compared to the posterior margin sagittal distance of the posterior edge of the endplate of the upper prosthesis. Compared with the intact model, the biomechanical parameters (ROM, FJF, endplate stress) of the C5-C6 segment increased the most when the sagittal distance of the posterior edge of the endplate of the upper prosthesis was 3mm. Compared with the intact model, the maximum intervertebral disc stress of C4-C5 and C6-C7 was 0.57MPa and 0.53MPa, respectively, when the sagittal distance of the posterior edge of the upper prosthetic endplate was 3mm. After the sagittal distance of the posterior edge of the prosthetic endplate was completely covered, the mechanical influence of the entire cervical spine was low. The sagittal distance at the posterior edge of the endplate of different sizes changed the motion pattern and load distribution of the implanted segment to some extent. When the sagittal distance between the prosthesis and the upper endplate was greater than or equal to 3mm, the mechanical indices of the implanted segment increased significantly, increasing the risk of local tissue injury, especially during extension motion. Compared to the sagittal distance at the posterior edge of the endplate of the lower prosthesis, increasing the sagittal distance at the posterior edge of the endplate of the upper prosthesis has a greater effect on the mechanics of the cervical spine.

Automated detection of vertebral fractures from X-ray images: A novel machine learning model and survey of the field

Vertebral fractures are a common problem and the most prevalent of thoracolumbar compression and burst fractures. However, vertebral fractures are difficult to diagnose: an experienced orthopedist or radiologist is required to detect and determine the type of vertebral fracture. Thus, artificial intelligence methods for diagnosing vertebral fractures are clinically useful.On the basis of a review of 12 studies in the literature, the earliest of which was published in 2020, we propose a machine learning model that detects and determines the type of vertebral fracture on the basis of X-ray data. In this method, YOLOv4 and ResUNet are used to segment vertebral bodies from X-ray images. In evaluation experiments, our method had a precision of 99%, 74%, and 94% in identifying healthy vertebrae, compression fractures, and burst fractures, respectively.

Automated vertebral bone mineral density measurement with phantomless internal calibration in chest LDCT scans using deep learning.

To develop and evaluate a fully automated method based on deep learning and phantomless internal calibration for bone mineral density (BMD) measurement and opportunistic low BMD (osteopenia and osteoporosis) screening using chest low-dose CT (LDCT) scans. A total of 1175 individuals were enrolled in this study, who underwent both chest LDCT and BMD examinations with quantitative computed tomography (QCT), by two different CT scanners (Siemens and GE). Two convolutional neural network (CNN) models were employed for vertebral body segmentation and labeling, respectively. A histogram technique was applied for vertebral BMD calculation using paraspinal muscle and surrounding fat as references. 195 cases (by Siemens scanner) as fitting cohort were used to build the calibration function. 698 cases as validation cohort I (VCI, by Siemens scanner) and 282 cases as validation cohort II (VCII, by GE scanner) were performed to evaluate the performance of the proposed method, with QCT as the standard for analysis. The average BMDs from the proposed method were strongly correlated with QCT (in VCI: r = 0.896, in VCII: r = 0.956, p < 0.001). Bland-Altman analysis showed a small mean difference of 1.1 mg/cm3, and large interindividual differences as seen by wide 95% limits of agreement (-29.9 to +32.0 mg/cm3) in VCI. The proposed method measured BMDs were higher than QCT measured BMDs in VCII (mean difference = 15.3 mg/cm3, p < 0.001). Osteoporosis and low BMD were diagnosed by proposed method with AUCs of 0.876 and 0.903 in VCI, 0.731 and 0.794 in VCII, respectively. The AUCs of the proposed method were increased to over 0.920 in both VCI and VCII after adjusting the cut-off. Without manual selection of the region of interest of body tissues, the proposed method based on deep learning and phantomless internal calibration has the potential for preliminary screening of patients with low BMD using chest LDCT scans. However, the agreement between the proposed method and QCT is insufficient to allow them to be used interchangeably in BMD measurement. This study proposed an automated vertebral BMD measurement method based on deep learning and phantomless internal calibration with paraspinal muscle and fat as reference.

Comparison of Hounsfield units and vertebral bone quality score for the prediction of time to pathologic fracture in mobile spine metastases treated with radiotherapy.

Spine metastases are commonly treated with radiotherapy for local tumor control; pathologic fracture is a potential complication of spinal radiotherapy. Both Hounsfield units (HUs) on CT and vertebral bone quality (VBQ) on MRI have been argued to predict stability as measured by odds of pathologic fracture, although it is unclear if there is a difference in the predictive power between the two methodologies. The objective of the present study was to examine whether one methodology is a better predictor of pathologic fracture following radiotherapy for mobile spine metastases. Patients who underwent radiotherapy (conventional external-beam radiation therapy, stereotactic body radiation therapy, or intensity-modulated radiation therapy) for mobile spine (C1-L5) metastases at a tertiary care center were retrospectively identified. Details regarding underlying pathology, patient demographics, and tumor morphology were collected. Vertebral involvement was assessed using the Weinstein-Boriani-Biagini (WBB) system. Bone quality of the non-tumor-involved bone was assessed on both pretreatment CT and MRI. Univariable analyses were conducted to identify independent predictors of fracture, and Kaplan-Meier analyses were used to identify significant predictors of time to pathologic fracture. Stepwise Cox regression analysis was used to determine independent predictors of time to fracture. One hundred patients were included (mean age 62.7 ± 11.9 years; 61% male), of whom 35 experienced postradiotherapy pathologic fractures. The most common histologies were lung (22%), prostate (21%), breast (14%), and renal cell (13%). On univariable analysis, the mean HUs of the vertebrae adjacent to the fractured vertebra were significantly lower among those experiencing fracture; VBQ was not significantly associated with fracture odds. Survival analysis showed that average HUs ≤ 132, nonprostate pathology, involvement of ≥ 3 vertebral body segments on the WBB system, Spine Instability Neoplastic Score (SINS) ≥ 7, and the presence of axial pain all predicted increased odds of fracture (all p < 0.001). Cox regression found that HUs ≤ 132 (OR 2.533, 95% CI 1.257-5.103; p = 0.009), ≥ 3 WBB vertebral body segments involved (OR 2.376, 95% CI 1.132-4.987; p = 0.022), and axial pain (OR 2.036, 95% CI 0.916-4.526; p = 0.081) predicted increased fracture odds, while prostate pathology predicted decreased odds (OR 0.076, 95% CI 0.009-0.613; p = 0.016). Sensitivity analysis suggested that an HU threshold of ≤ 132 and a SINS of ≥ 7 identified patients at increased risk of fracture. The present results suggest that bone density surrogates as measured on CT, but not MRI, can be used to predict the risk of pathologic fracture following radiotherapy for mobile spine metastases. More extensive vertebral body involvement and the presence of mechanical axial pain additionally predict increased fracture odds.

Finite element analysis of biomechanical effects of mineralized collagen modified bone cement on adjacent vertebral body after vertebroplasty.

Objective: To investigate whether mineralized collagen modified polymethyl methacrylate (MC-PMMA) bone cement impacts the implanted vertebral body and adjacent segments and the feasibility of biomechanical properties compared with common bone cement in the treatment of osteoporotic vertebral compression fractures (OVCF). Methods: A healthy volunteer was selected to perform a three-dimensional reconstruction of the T11-L1 vertebral body to establish the corresponding finite element model of the spine, and the changes in the stress distribution of different types of cement were biomechanically analyzed in groups by applying quantitative loads. Results: The stress distribution of the T11-L1 vertebral body was similar between the two bone types of cement under various stress conditions. Conclusion: Mineralized collagen modified bone cement had the advantages of promoting bone regeneration, good biocompatibility, good transformability, and coupling, and had support strength not inferior to common PMMA bone cement, indicating it has good development prospects and potential.

Semi-supervised hybrid spine network for segmentation of spine MR images.

Automatic segmentation of vertebral bodies (VBs) and intervertebral discs (IVDs) in 3D magnetic resonance (MR) images is vital in diagnosing and treating spinal diseases. However, segmenting the VBs and IVDs simultaneously is not trivial. Moreover, problems exist, including blurry segmentation caused by anisotropy resolution, high computational cost, inter-class similarity and intra-class variability, and data imbalances. We proposed a two-stage algorithm, named semi-supervised hybrid spine network (SSHSNet), to address these problems by achieving accurate simultaneous VB and IVD segmentation. In the first stage, we constructed a 2D semi-supervised DeepLabv3+ by using cross pseudo supervision to obtain intra-slice features and coarse segmentation. In the second stage, a 3D full-resolution patch-based DeepLabv3+ was built. This model can be used to extract inter-slice information and combine the coarse segmentation and intra-slice features provided from the first stage. Moreover, a cross tri-attention module was applied to compensate for the loss of inter-slice and intra-slice information separately generated from 2D and 3D networks, thereby improving feature representation ability and achieving satisfactory segmentation results. The proposed SSHSNet was validated on a publicly available spine MR image dataset, and remarkable segmentation performance was achieved. Moreover, results show that the proposed method has great potential in dealing with the data imbalance problem. Based on previous reports, few studies have incorporated a semi-supervised learning strategy with a cross attention mechanism for spine segmentation. Therefore, the proposed method may provide a useful tool for spine segmentation and aid clinically in spinal disease diagnoses and treatments. Codes are publicly available at: https://github.com/Meiyan88/SSHSNet.

Cervical Spine Magnetic Resonance Imaging Findings in Hirayama Disease.

Background Hirayama disease is an uncommon type of cervical cord myelopathy seen typically in young males due to trauma from flexion movements. This study aims to assess the clinical presentations and classify the extent of various cervical spine MRI findings for the local population. Methodology A retrospective study of 13 patients diagnosed with Hirayama disease oncervical MRI was performed from January 2017 to December 2022 at Dr. D. Y. Patil Medical College, Hospital and Research Center, Pune. Results Of the 13 patients, 12 (92%) were male, and one (8%) was female. Nine (69%) patients were in the 16-25-year age group, two (15%) were in the 26-35-year age group, and one (8%) each was in the 6-15-year and 66-75-year age groups. Upper limb weakness was the most common clinical symptom seen in 12 (92%) patients, followed by distal muscle atrophy in seven (54%) patients. Tremors in the hand were a rare symptom seen in two patients. Claw hand was an atypical symptom seen in one patient. On cervical MRI, all patients showed excessive forward shifting of the posterior dura on flexion, with resultant cord compression due to tightness of the dural sac. One (8%) patient had no signs of myelopathy, while 12 (92%) patients had developed chronic myelomalacia and showed abnormal cord hyperintensity and atrophy in the lower cervical cord. All 13 (100%) patients showed increased laminodural space on flexion; the mean thickness was 4.08 mm, with the minimum and maximum thickness being 2.4 mm and 6.7 mm, respectively. Classifying by length of the anterior bulging dura, one (8%) patient showed involvement of less than two vertebral body segments, eight (62%) patients showed involvement of two to four vertebral body segments, and four (30%) patients showed involvement of more than four vertebral body segments. Crescent-shaped post-contrast enhancement on flexion was seen in all eight (100%) patients who underwent a contrast study. Prominent epidural flow voids on flexion were seen in six (46%) patients. Conclusions Hirayama disease is an uncommon type of cervical myelopathy seen typically in juvenile males. The occult onset of distal upper limb weakness and atrophy during puberty, typical MRI features of lower cervical cord atrophy, and the presence of a crescent-shaped enhancing mass in the posterior epidural space are pathognomonic of the condition. A few atypical cases can also occur. Early diagnosis and treatment are crucial to avoiding serious dysfunction.

Vertebropexy as a semi-rigid ligamentous alternative to lumbar spinal fusion

PurposeTo develop ligamentous vertebral stabilization techniques (“vertebropexy”) that can be used after microsurgical decompression (intact posterior structures) and midline decompression (removed posterior structures) and to elaborate their biomechanical characteristics.MethodsFifteen spinal segments were biomechanically tested in a stepwise surgical decompression and ligamentous stabilization study. Stabilization was achieved with a gracilis or semitendinosus tendon allograft, which was attached to the spinous process (interspinous vertebropexy) or the laminae (interlaminar vertebropexy) in form of a loop. The specimens were tested (1) in the native state, after (2) microsurgical decompression, (3) interspinous vertebropexy, (4) midline decompression, and (5) interlaminar vertebropexy. In the intact state and after every surgical step, the segments were loaded in flexion–extension (FE), lateral shear (LS), lateral bending (LB), anterior shear (AS) and axial rotation (AR).ResultsInterspinous vertebropexy significantly reduced the range of motion (ROM) in all loading scenarios compared to microsurgical decompression: in FE by 70% (p < 0.001), in LS by 22% (p < 0.001), in LB by 8% (p < 0.001) in AS by 12% (p < 0.01) and in AR by 9% (p < 0.001). Interlaminar vertebropexy decreased ROM compared to midline decompression by 70% (p < 0.001) in FE, 18% (p < 0.001) in LS, 11% (p < 0.01) in LB, 7% (p < 0.01) in AS, and 4% (p < 0.01) in AR. Vertebral segment ROM was significantly smaller with the interspinous vertebropexy compared to the interlaminar vertebropexy for all loading scenarios except FE. Both techniques were able to reduce vertebral body segment ROM in FE, LS and LB beyond the native state.ConclusionVertebropexy is a new concept of semi-rigid spinal stabilization based on ligamentous reinforcement of the spinal segment. It is able to reduce motion, especially in flexion–extension. Studies are needed to evaluate its clinical application.

Analysis of the Curative Effect of Curved Angle Vertebroplasty in the Treatment of Osteoporotic Vertebral Compression Fracture.

To evaluate the clinical efficacy of Percutaneous curved vertebroplasty (PCVP) for osteoporotic vertebral compression fracture of the thoracolumbar spine. Retrospective analysis of 113 patients with osteoporotic vertebral compressive fractures (OVCFs) in our hospital from January 2017 to January 2020, a total of 120 diseased vertebrae, were divided into PCVP group (35 cases, 37 sections) and bilateral PVP(BVP) group (78 cases, 83 sections). To compare the distribution of baseline clinical data, pain relief (Visual Analog Scale, VAS), ODI (Oswestry Dability Index, ODI), operation time, intraoperative fluoroscopy, postoperative vertebral body re-fracture, and comparison of bone, and to compare the volume of cement penetration and the leakage rate of bone cement, etc. There was no significant difference in VAS and ODI before operation between the two groups (P > 0.05), and the VAS score and ODI after operation were significantly improved (P < 0.001). Compared with the bilateral PVP group, the operation time, the number of fluoroscopy, and the leakage rate of each layer of bone cement in the PCVP group were significantly reduced (P < 0.05); however, the amount of cement used in the two groups was similar (P > 0.05). There were no serious complications in both groups. In the bilateral PVP group, a total of seven patients had adjacent vertebral fractures or re-fractures of the original vertebral body. However, no patients in the PCVP group had re-fractures in any vertebral body segment. Both PCVP and bilateral PVP are safe and effective methods for the treatment of osteoporotic vertebral compression fractures, but PCVP has a short operation time, fewer fluoroscopy times, and a low bone cement leakage rate.

Automatic segmentation and radiomic texture analysis for osteoporosis screening using chest low-dose computed tomography.

This study developed a diagnostic tool combining machine learning (ML) segmentation and radiomic texture analysis (RTA) for bone density screening using chest low-dose computed tomography (LDCT). A total of 197 patients who underwent LDCT followed by dual-energy X-ray absorptiometry were analyzed. First, an autosegmentation model was trained using LDCT to delineate the thoracic vertebral body (VB). Second, a two-level classifier was developed using radiomic features extracted from VBs for the hierarchical pairwise classification of each patient's bone status. All the patients were initially classified as either normal or abnormal, and all patients with abnormal bone density were then subdivided into an osteopenia group and an osteoporosis group. The performance of the classifier was evaluated through fivefold cross-validation. The model for automated VB segmentation achieved a Sorenson-Dice coefficient of 0.87 ± 0.01. Furthermore, the area under the receiver operating characteristic curve scores for the two-level classifier were 0.96 ± 0.01 for detecting abnormal bone density (accuracy = 0.91 ± 0.02; sensitivity = 0.93 ± 0.03; specificity = 0.89 ± 0.03) and 0.98 ± 0.01 for distinguishing osteoporosis (accuracy = 0.94 ± 0.02; sensitivity = 0.95 ± 0.03; specificity = 0.93 ± 0.03). The testing prediction accuracy levels for the first- and second-level classifiers were 0.92 ± 0.04 and 0.94 ± 0.05, respectively. The overall testing prediction accuracy of our method was 0.90 ± 0.05. The combination of ML segmentation and RTA for automated bone density prediction based on LDCT scans is a feasible approach that could be valuable for osteoporosis screening during lung cancer screening. • This study developed an automatic diagnostic tool combining machine learning-based segmentation and radiomic texture analysis for bone density screening using chest low-dose computed tomography. • The developed method enables opportunistic screening without quantitative computed tomography or a dedicated phantom. • The developed method could be integrated into the current clinical workflow and used as an adjunct for opportunistic screening or for patients who are ineligible for screening with dual-energy X-ray absorptiometry.

Study on anatomical parameters of adult lumbar intervertebral disc and endplate based on CT

To provide basic data for clinical application and individualized design of lumbar disc prostheses by measuring the anatomical parameters of lumbar intervertebral discs and endplates in healthy adults with CT three-dimensional reconstruction technology. A retrospective analysis was performed on 200 males and 200 females with normal lumbar spine who were admitted to the imaging center or outpatient department of the Second Affiliated Hospital of Xinjiang Medical University from September 2019 to December 2020. The age ranged from 20 to 60 years old, with an average of (40.61±11.22) years old. The measurement segment was L1-S1 intervertebral disc, and the measurement indicators included the axial anteroposterior diameter and transverse diameter of the intervertebral disc, sagittal anterior, middle and posterior height, coronal left and right height, intervertebral space angle, and transverse and anteroposterior diameters of the upper and lower endplates of each vertebral body. ①In terms of gender, the anatomical parameters of L1-S1 disc axial diameter, transverse diameter, sagittal anterior, middle and posterior height, left and right coronal height and intervertebral space angle were all higher in males than in females(P<0.05), and the anatomical parameters of upper and lower endplates of L1-S1 vertebral body were higher in males than in females(P<0.001). ②In comparison of sagittal height of anterior, middle and posterior intervertebral discs, the sagittal height of L1-L5 intervertebral discs was middle-high > anterior-high > posterior-high(P<0.001), while that of L5S1 intervertebral disc was anterior-high > middle-high > posterior-high (P<0.001). ③In the comparison of left and right coronal height, there was no statistical significance in the left and right coronal height of L1-S1 disc between male and female(P>0.05). ④The L1-S1 intervertebral spaces angle between male and female increased with the increase of vertebral body segments. ⑤The anterior and posterior diameters and transverse diameters of upper and lower of L1-S1 vertebral bodies endplates were height in males than in females(P<0.001). The results suggest that gender differences should be considered in the design of adult lumbar disc prostheses. The anatomical parameters of the lumbar intervertebral disc varied with the increase of the vertebral body sequence, suggesting that different anatomical parameters of the intervertebral disc should be considered in the design of the artificial intervertebral disc, and the changes in the height of the sagittal position suggest that the design of the intervertebral disc should be wedge-shaped.

Lumbar intervertebral disc segmentation for computer modeling and simulation

Background and objectiveThe present work had as its main objective the development of a method for localizing and automatically segmenting lumbar intervertebral discs (IVD) in 3D from magnetic resonance imaging (MRI), with the goal of supporting the generation of finite element (FE) models from actual lumbar spine anatomy, by providing accurate and personalized information on the shape of the patient's IVD. The extension of the method to allow performing separate segmentations of the IVD's two main structures – annulus fibrosus (AF) and nucleus pulposus (NP) – as well as automatically detecting degenerated IVD where this distinction is no longer possible was also an objective of the work. MethodsThe method presented here evolves from 2D segmentations in the sagittal profile using Gabor filters towards 3D segmentations. It works by detecting the spine curves and intensity regions corresponding to IVD. As so, the 2D method from Zhu et al. (2013) was partially implemented, modified and adapted to 3D use, and then tested with eight spines from two separated online datasets. The 3D adaptation was achieved by using vertebral body segmentation masks to approximate the shape of the vertebrae and to adjust the spine curves accordingly. ResultsThe method showed average values of 85%, 83% and 96% for the Dice coefficient, sensitivity and specificity, respectively. The method correctly identified 65 of 68 (96%) IVD as either healthy or degenerated. The method's Dice coefficient is within the range of existing 3D IVD segmentation methods in the literature (81–92%). The method took on average 6–7 s to perform a full 3D segmentation, which is well within the range of the existing methods (2 s – 19 min). ConclusionsThe developed method can be used to generate accurate 3D models of the IVD based on MRI, with AF/NP distinction and detection of marked degeneration by comparing each IVD with the remaining spine levels. Further work shall improve the method towards distinguishing between specific levels of degeneration for clinically oriented FE modeling.