scispace - formally typeset

Book ChapterDOI

Automated Cobb Angle Computation from Scoliosis Radiograph

19 Jan 2018-Communications in computer and information science (Springer, Singapore)-Vol. 836, pp 140-155

TL;DR: A fully automatic technique for Cobb angle computation from Scoliosis radiograph image where the objectives are to have no user intervention and to increase the reliability of spinal curvature magnitude quantification.

AbstractIn this paper we propose a fully automatic technique for Cobb angle computation from Scoliosis radiograph image where the objectives are to have no user intervention and to increase the reliability of spinal curvature magnitude quantification. The automatic technique mainly comprises of four steps, namely: Preprocessing, ROI identification, Object centerline extraction and Cobb angle computation from the extracted spine centerline. Bilateral image denoising is considered as the preprocessing step. Support Vector Machine classifier is used for object identification. We have assumed that the spine is a continuous contour rather than a series of discrete vertebral bodies with individual orientations. Morphological operation, Gaussian blurring, spine centerline approximation and polynomial fit are used to extract the centerline of spine. The tangent at every point of the extracted centerline is taken and Cobb angle is evaluated from these tangent values. To analyze the automated diagnosis technique, the proposed approach was evaluated on a set of 21 coronal radiograph images. Identification of ROI based on Support Vector Machine classifier is effective enough with a sensitivity and specificity of 100% and the center line extraction from this ROI gave correct results for 57.14% subjects with very less or negligible angular variability. As the vertebral endplates in radiograph images have poor contrast due to reduced radiation dose, the continuous contour based approach gives better reliability.

...read more


Citations
More filters
09 Nov 2011
TL;DR: The widespread availability of inclinometer-equipped mobile phones and the ability to store measurements in later versions of the angle measurement software may make these new technologies attractive for clinical measurement applications.
Abstract: Purpose: The Cobb technique is the universally accepted method for measuring the severity of spinal deformities. Traditionally, Cobb angles have been measured using protractor and pencil on hardcopy radiographic films. The new generation of mobile phones make accurate angle measurement possible using an integrated accelerometer, providing a potentially useful clinical tool for assessing Cobb angles. The purpose of this study was to compare Cobb angle measurements performed using an Apple iPhone and traditional protractor in a series of twenty Adolescent Idiopathic Scoliosis patients. Methods: Seven observers measured major Cobb angles on twenty pre-operative postero-anterior radiographs of Adolescent Idiopathic Scoliosis patients with both a standard protractor and using an Apple iPhone. Five of the observers repeated the measurements at least a week after the original measurements. Results: The mean absolute difference between pairs of iPhone/protractor measurements was 2.1°, with a small (1°) bias toward lower Cobb angles with the iPhone. 95% confidence intervals for intra-observer variability were ±3.3° for the protractor and ±3.9° for the iPhone. 95% confidence intervals for inter-observer variability were ±8.3° for the iPhone and ±7.1° for the protractor. Both of these confidence intervals were within the range of previously published Cobb measurement studies. Conclusions: We conclude that the iPhone is an equivalent Cobb measurement tool to the manual protractor, and measurement times are about 15% less. The widespread availability of inclinometer-equipped mobile phones and the ability to store measurements in later versions of the angle measurement software may make these new technologies attractive for clinical measurement applications.

11 citations

Journal ArticleDOI
TL;DR: It was concluded that CLT method is more reproducible than the Cobb method for measuring spinal curvature and more repeatable than Cobb Method.
Abstract: Background : Scoliosis is a health problem that causes a side-to-side curvature in the spine. The curvature may have an “S” or “C” shape. To evaluate scoliosis, the Cobb angle has been commonly used. However, digital image processing allows the Cobb angle to be obtained easily and quickly, several researchers have determined that Cobb angle contains high variations (errors) in the measurements. Therefore, a more reproducible computer aided-method to evaluate scoliosis is presented. Material and Methods: In this analytical study, several polynomial curves were fitted to the spine curvature (4 th to 8 th order) of thirty plain films of scoliosis patients to obtain the Curvature-Length of the spine. Each plain film was evaluated by 3 physician observers. Curvature was measured twice using the Cobb method and the proposed Curvature-Length Technique (CLT). Data were analyzed by a paired-sample Student t-test and Pearson correlation method using SPSS Statistics 25. Results: The curve of 7 th order polynomial had the best fit on the spine curvature and was also used for our proposed method (CLT) obtaining a significant positive correlation when compared to Cobb measurements (r=0.863, P<0.001). The Intraclass Correlation (ICC) was between 0.863 and 0.948 for Cobb method and0.974 to 0.984 for CLT method. In addition, mean measurement of the inter-observer COV (Coefficient of Variation) for Cobb method was of 0.185, that was significantly greater than the obtained with CLT method of 0.155, this means that CLT method is 16.2% more repeatable than Cobb Method. Conclusion: Based on results, it was concluded that CLT method is more reproducible than the Cobb method for measuring spinal curvature.

References
More filters
Journal ArticleDOI
TL;DR: High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated and the performance of the support- vector network is compared to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Abstract: The support-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data. High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.

35,157 citations

Proceedings ArticleDOI
04 Jan 1998
TL;DR: In contrast with filters that operate on the three bands of a color image separately, a bilateral filter can enforce the perceptual metric underlying the CIE-Lab color space, and smooth colors and preserve edges in a way that is tuned to human perception.
Abstract: Bilateral filtering smooths images while preserving edges, by means of a nonlinear combination of nearby image values. The method is noniterative, local, and simple. It combines gray levels or colors based on both their geometric closeness and their photometric similarity, and prefers near values to distant values in both domain and range. In contrast with filters that operate on the three bands of a color image separately, a bilateral filter can enforce the perceptual metric underlying the CIE-Lab color space, and smooth colors and preserve edges in a way that is tuned to human perception. Also, in contrast with standard filtering, bilateral filtering produces no phantom colors along edges in color images, and reduces phantom colors where they appear in the original image.

8,053 citations

Journal ArticleDOI
TL;DR: The grating cell operator is the only one that selectively responds only to texture and does not give false response to nontexture features such as object contours and the texture detection capabilities of the operators are compared.
Abstract: Texture features that are based on the local power spectrum obtained by a bank of Gabor filters are compared. The features differ in the type of nonlinear post-processing which is applied to the local power spectrum. The following features are considered: Gabor energy, complex moments, and grating cell operator features. The capability of the corresponding operators to produce distinct feature vector clusters for different textures is compared using two methods: the Fisher (1923) criterion and the classification result comparison. Both methods give consistent results. The grating cell operator gives the best discrimination and segmentation results. The texture detection capabilities of the operators and their robustness to nontexture features are also compared. The grating cell operator is the only one that selectively responds only to texture and does not give false response to nontexture features such as object contours.

719 citations

Journal ArticleDOI
TL;DR: To quantitate the intrinsic error in measurement, fifty anteroposterior radiographs of patients who had scoliosis were each measured on six separate occasions by four orthopaedic surgeons using the Cobb method.
Abstract: To quantitate the intrinsic error in measurement, fifty anteroposterior radiographs of patients who had scoliosis were each measured on six separate occasions by four orthopaedic surgeons using the Cobb method For the first two measurements (Set I), each observer selected the end-vertebrae of the curve; for the next two measurements (Set II), the end-vertebrae were pre-selected and constant The last two measurements (Set III) were obtained in the same manner as Set II, except that each examiner used the same protractor rather than the one that he carried with him The pooled results of all four observers suggested that the 95 per cent confidence limit for intraobserver variability was 49 degrees for Set I, 38 degrees for Set II, and 28 degrees for Set III The interobserver variability was 72 degrees for Set I and 63 degrees for Sets II and III The mean angles differed significantly between observers, but the difference was smaller when the observers used the same protractor

493 citations

Book ChapterDOI
01 Jan 2008
TL;DR: The objective of this study was to evaluate SVMs for their effectiveness and prospects for object-based image analysis as a modern computational intelligence method and the SVM methodology seems very promising for Object Based Image Analysis.
Abstract: The Support Vector Machine is a theoretically superior machine learning methodology with great results in pattern recognition. Especially for supervised classification of high-dimensional datasets and has been found competitive with the best machine learning algorithms. In the past, SVMs were tested and evaluated only as pixel-based image classifiers. During recent years, advances in Remote Sensing occurred in the field of Object-Based Image Analysis (OBIA) with combination of low level and high level computer vision techniques. Moving from pixel-based techniques towards object-based representation, the dimensions of remote sensing imagery feature space increases significantly. This results to increased complexity of the classification process, and causes problems to traditional classification schemes. The objective of this study was to evaluate SVMs for their effectiveness and prospects for object-based image analysis as a modern computational intelligence method. Here, an SVM approach for multi-class classification was followed, based on primitive image objects provided by a multi-resolution segmentation algorithm. Then, a feature selection step took place in order to provide the features for classification which involved spectral, texture and shape information. After the feature selection step, a module that integrated an SVM classifier and the segmentation algorithm was developed in C++. For training the SVM, sample image objects derived from the segmentation procedure were used. The proposed classification procedure followed, resulting in the final object classification. The classification results were compared to the Nearest Neighbor object-based classifier results, and were found satisfactory. The SVM methodology seems very promising for Object Based Image Analysis and future work will focus on integrating SVM classifiers with rule-based classifiers.

127 citations