Centroid method: an alternative method of determining coronal curvature in scoliosis. A comparative study versus Cobb method in the degenerative spine

W-Transformer: Accurate Cobb angles estimation by using a transformer-based hybrid structure

BACKGROUND

Scoliosis is a type of spinal deformity, which is harmful to a person's health. In severe cases, it can trigger paralysis or death. The measurement of Cobb angle plays an essential role in assessing the severity of scoliosis.

PURPOSE

The aim of this paper is to propose an automatic system for landmark detection and Cobb angle estimation, which can effectively help clinicians diagnose and treat scoliosis.

METHODS

A novel hybrid framework was proposed to measure Cobb angle precisely for clinical diagnosis, which was referred as W-Transformer due to its w-shaped architecture. First, a convolutional neural network of cascade residual blocks as our backbone was designed. Then a transformer was fused to learn the dependency information between spine and landmarks. In addition, a reinforcement branch was designed to improve the overlap of landmarks, and an improved prediction module was proposed to fine-tune the final coordinates of landmarks in Cobb angles estimation. Besides, the public AASCE MICCAI 2019 challenge was served as dataset. It supplies 609 manually labeled spine AP X-ray images, each of which contains a total of 68 landmark labels and three Cobb Angles tags.

RESULTS

From the perspective of the AASCE MICCAI 2019 challenge, we achieved a lower symmetric mean absolute percentage error (SMAPE) of 8.26% for all Cobb angles and the lowest averaged detection error of 50.89 in terms of landmark detection, compared with many state-of-the-art methods. We also provided the SMAPEs for the Cobb angles of the Proximal-Thoracic (PT), the Main-Thoracic (MT) and the Thoracic-Lumbar (TL) area, which are 5.27%, 14.59% and 20.97% respectively, however, these data were not covered in most previous studies. Statistical analysis demonstrates that our model has obtained a high level of Pearson correlation coefficient of 0.9398 (p<0.001), which shows excellent reliability of our model. Our model can yield 0.9489 (p<0.001), 0.8817 (p<0.001) and 0.9149 (p<0.001$) for PT, MT and TL, respectively. The overall variability of Cobb angle measurement is less than 4°, implying clinical value. And the mean absolute deviation (Standard Deviation) for three regions is 3.64° (4.13°), 3.84° (4.66°) and 3.80° (4.19°). The results of Student paired t-test indicate that no statistically significant differences are observed between manual measurement and our automatic approach (p value is always > 0.05). Regarding the diagnosis of scoliosis (Cobb angle > 10°), the proposed method achieves a high sensitivity of 0.9577 and a specificity of 0.8475 for all spinal regions.

CONCLUSIONS

This study offers a brand-new automatic approach that is potentially of great benefit of the complex task of landmark detection and Cobb angle evaluation, which can provide helpful navigation information about the early diagnosis of scoliosis. This article is protected by copyright. All rights reserved.

Intelligent Measurement of Spinal Curvature Using Cascade Gentle AdaBoost Classifier and Region-Based DRLSE

For spinal curvature measurements, because of the anatomical complexity of the spine CT image, developing an automated method to avoid manual landmark is a challenging task. In this study, we propose an intelligent framework that integrates the cascade AdaBoost classifier and region-based distance regularized level set evolution (DRLSE) with the vertebral centroid measurement. First, the histogram-of-oriented-gradients based cascade gentle AdaBoost classifier is used to detect automatically and localize vertebral bodies from computer tomography (CT) spinal images. Considering these vertebral pathological images enables us to produce a diverse training dataset. Then, the DRLSE method introduces the local region information to converge the vertebral boundary quickly. The located bounding box is regarded as an accurate initial contour. This avoids the negative impact of manual initialization. Finally, we perform vertebral centroid extraction and spinal curve fitting. The spinal curvature angle is determined by calculating the angle between two tangents to the curve. We verified the effectiveness of the proposed method on 10 spine CT volumes. Quantitative comparison against the ground-truth centroids yielded a detection accuracy rate of 98.3% and a mean centroid location error of 1.15 mm. The comparative results with existing methods demonstrate that the proposed method can accurately detect and segment vertebral bodies. Furthermore, the spinal curvature can be automatically measured without manual landmark.

Reliability analysis of Cobb measurement in degenerative lumbar scoliosis using endplate versus pedicle as bony landmarks

OBJECTIVES

Degenerative changes of endplates in older patients and tilting of vertebral body in lumbosacral lordosis could make an accurate identification of endplates for the Cobb measurement difficult. Pedicles have been proposed as alternative landmarks because they are usually better visualized, and offer similar clinical validity to the endplates. The objective of this study was to investigate the reliability of the pedicle method of Cobb measurement in degenerative lumbar scoliosis and compare it with the traditional endplate method.

METHODS

Two hundred and eighty-four radiographs of degenerative lumbar scoliosis were evaluated. The radiographs were classified into groups based on the patient's age (< 60 years, 60 to 80 years, and > 80 years), level of lower end vertebra (LEV) (LEV at L5, and LEV at or above L4), and curve severity (< 20°, 20° to 40°, and > 40°). Three observers independently measured the radiographs using the endplate and pedicle methods twice with an interval of 1 week. The intra- and interobserver reliabilities were calculated using intraclass correlation coefficients (ICC).

RESULTS

The intra- and interobserver ICC values were better for all observers in the > 80 years age group using the pedicle method. The intraobserver ICC values of pedicle method were also better in the LEV at L5 group, and the interobserver ICC values showed a slightly better consistency with the pedicle method. For patients with > 40° curves, the intraobserver ICC values for all observers as well as interobserver ICC values were better using the endplate method.

CONCLUSION

The reliabilities of the endplate and pedicle methods for degenerative lumbar scoliosis were both excellent. The pedicle method might be better in older patients (> 80 years) and those with LEV at L5; while the endplate method could have some strength in severe cases (> 40°).

What is the optimal cutoff value of the axis-line-angle technique for evaluating trunk imbalance in coronal plane?

BACKGROUND CONTEXT

Accurately evaluating the extent of trunk imbalance in the coronal plane is significant for patients before and after treatment. We preliminarily practiced a new method, axis-line-angle technique (ALAT), for evaluating coronal trunk imbalance with excellent intra-observer and interobserver reliability. Radiologists and surgeons were encouraged to use this method in clinical practice. However, the optimal cutoff value of the ALAT for determination of the extent of coronal trunk imbalance has not been calculated up to now.

PURPOSE

The purpose of this study was to identify the cutoff value of the ALAT that best predicts a positive measurement point to assess coronal balance or imbalance.

STUDY DESIGN/SETTING

A retrospective study at a university affiliated hospital was carried out.

PATIENT SAMPLE

A total of 130 patients with C7-central sacral vertical line (CSVL) >0 mm and aged 10-18 years were recruited in this study from September 2013 to December 2014.

OUTCOME MEASURES

Data were analyzed to determine the optimal cutoff value of the ALAT measurement.

METHODS

The C7-CSVL and ALAT measurements were conducted respectively twice on plain film within a 2-week interval by two radiologists. The optimal cutoff value of the ALAT was analyzed via receiver operating characteristic (ROC) curve. Comparison variables were performed with chi-square test between the C7-CSVL and ALAT measurements for evaluating trunk imbalance. Kappa agreement coefficient method was used to test the intra-observer and interobserver agreement of C7-CSVL and ALAT.

RESULTS

The ROC curve area for the ALAT was 0.82 (95% confidence interval: 0.753-0.894, p<.001). The maximum Youden index was 0.51, and the corresponding cutoff point was 2.59°. No statistical difference was found between the C7-CSVL and ALAT measurements for evaluating trunk imbalance (p>.05). Intra-observer agreement values for the C7-CSVL measurements by observers 1 and 2 were 0.79 and 0.91 (p<.001), respectively, whereas intra-observer agreement values for the ALAT measurements were both 0.89 by observers 1 and 2 (p<.001). The interobserver agreement values for the first and second measurements with the C7-CSVL were 0.78 and 0.85 (p<.001), respectively, whereas the interobserver agreement values for the first and second measurements with the ALAT were 0.91 and 0.88 (p<.001), respectively.

CONCLUSIONS

The newly developed ALAT provided an acceptable optimal cutoff value for evaluating trunk imbalance in the coronal plane with a high level of intra-observer and interobserver agreement, which suggests that the ALAT is suitable for clinical use.

Reliability of a new method for measuring coronal trunk imbalance, the axis-line-angle technique

BACKGROUND CONTEXT

Accurate determination of the extent of trunk imbalance in the coronal plane plays a key role in an evaluation of patients with trunk imbalance, such as patients with adolescent idiopathic scoliosis. An established, widely used practice in evaluating trunk imbalance is to drop a plumb line from the C7 vertebra to a key reference axis, the central sacral vertical line (CSVL) in full-spine standing anterioposterior radiographs, and measuring the distance between them, the C7-CSVL. However, measuring the CSVL is subject to intraobserver differences, is error-prone, and is of poor reliability. Therefore, the development of a different way to measure trunk imbalance is needed.

PURPOSE

This study aimed to describe a new method to measure coronal trunk imbalance, the axis-line-angle technique (ALAT), which measures the angle at the intersection between the C7 plumb line and an axis line drawn from the vertebral centroid of the C7 to the middle of the superior border of the symphysis pubis, and to compare the reliability of the ALAT with that of the C7-CSVL.

STUDY DESIGN/SETTING

A prospective study at a university hospital was used.

PATIENT SAMPLE

The patient sample consisted of sixty-nine consecutively enrolled men and women patients, aged 10-18 years, who had trunk imbalance defined as C7-CSVL longer than 20 mm on computed full-spine standing anterioposterior radiographs.

OUTCOME MEASURES

Data were analyzed to determine the correlation between C7-CSVL and ALAT measurements and to determine intraobserver and interobserver reliabilities.

METHODS

Using a picture archiving and communication system, three radiologists independently evaluated trunk imbalance on the 69 computed radiographs by measuring the C7-CSVL and by measuring the angle determined by the ALAT. Data were analyzed to determine the correlations between the two measures of trunk imbalance, and to determine intraobserver and interobserver reliabilities of each of them.

RESULTS

Overall results from the measurements by the C7-CSVL and the ALAT were significantly moderately correlated. Intraobserver assessments by measuring the C7-CSVL and by doing the ALAT failed to find any significant differences between the findings from the first and second assessments by the same radiologist. Interobserver assessments significantly differed between radiologists 1 and 2 for the first assessment measuring the C7-CSVL, and between radiologists 2 and 3 for the second assessment measuring the C7-CSVL. Interobserver assessments by doing the ALAT failed to find any significant differences among the three radiologists for either of the two assessments.

CONCLUSIONS

Our results indicated that using the ALAT, which is simple and convenient, is of great value in measuring trunk imbalance. For measuring trunk imbalance, the ALAT has essential advantages compared with measuring the C7-CSVL. We encourage spine surgeons to consider using the ALAT in evaluating trunk imbalance.

Comparison in clinical outcome of two surgical treatments in degenerative scoliosis

To evaluate the clinical outcome of two different surgical treatments in treating degenerative scoliosis. Forty patients with degenerative scoliosis hospitalized in our department from June 2010 to June 2012 were selected. They were randomly divided into two groups. The first group was performed on the points with nerve or spinal compression for decompression, bone grafting, and short-segmental fixation in situ; The second group was treated with sufficient decompression, long-segmental fixation, and short-segmental fusion to operate orthopedic on scoliosis in three dimensions. All patients completed the follow-up period for more than 1 year, with the average of 18 months. Bone grafting fusion was achieved in all of the patients. The second group showed significantly better result in remission rate of postoperative pain and ODI improvement rate than the first group. Long-segmental internal fixation orthopedic is a better surgical option for patients with degenerative scoliosis to achieve sufficient decompression and three-dimensional orthopedic; therefore, it is a better solution for biomechanical reconstruction of spine.