A region-based segmentation of tumour from brain CT images using nonlinear support vector machine classifier

The proposed system provides new textural information for segmenting tumours, efficiently and accurately and with less computational time, from benign and malignant tumour images, especially in smaller dimensions of tumour regions of computed tomography (CT) images. Region-based segmentation of tumour from brain CT image data is an important but time-consuming task performed manually by medical experts. The objective of this work is to segment brain tumour from CT images using combined grey and texture features with new edge features and nonlinear support vector machine (SVM) classifier. The selected optimal features are used to model and train the nonlinear SVM classifier to segment the tumour from computed tomography images and the segmentation accuracies are evaluated for each slice of the tumour image. The method is applied on real data of 80 benign, malignant tumour images. The results are compared with the radiologist labelled ground truth. Quantitative analysis between ground truth and the segmented tumour is presented in terms of segmentation accuracy and the overlap similarity measure dice metric. From the analysis and performance measures such as segmentation accuracy and dice metric, it is inferred that better segmentation accuracy and higher dice metric are achieved with the normalized cut segmentation method than with the fuzzy c-means clustering method.     

Feature Selection and Performance Evaluation of Support Vector Machine (SVM)-Based Classifier for Differentiating Benign and Malignant Pulmonary Nodules by Computed Tomography

There are lots of work being done to develop computer-assisted diagnosis and detection (CAD) technologies and systems to improve the diagnostic quality for pulmonary nodules. Another way to improve accuracy of diagnosis on new images is to recall or find images with similar features from archived historical images which already have confirmed diagnostic results, and the content-based image retrieval (CBIR) technology has been proposed for this purpose. In this paper, we present a method to find and select texture features of solitary pulmonary nodules (SPNs) detected by computed tomography (CT) and evaluate the performance of support vector machine (SVM)-based classifiers in differentiating benign from malignant SPNs. Seventy-seven biopsy-confirmed CT cases of SPNs were included in this study. A total of 67 features were extracted by a feature extraction procedure, and around 25 features were finally selected after 300 genetic generations. We constructed the SVM-based classifier with the selected features and evaluated the performance of the classifier by comparing the classification results of the SVM-based classifier with six senior radiologists′ observations. The evaluation results not only showed that most of the selected features are characteristics frequently considered by radiologists and used in CAD analyses previously reported in classifying SPNs, but also indicated that some newly found features have important contribution in differentiating benign from malignant SPNs in SVM-based feature space. The results of this research can be used to build the highly efficient feature index of a CBIR system for CT images with pulmonary nodules.     

Unsupervised tumour segmentation in PET using local and global intensity-fitting active surface and alpha matting

This paper proposes an unsupervised tumour segmentation approach for PET data. The method computes the volumes of interest (VOIs) with sub-voxel precision by considering the limited image resolution and partial volume effects. First, an improved anisotropic diffusion filter is used to remove image noise. A hierarchical local and global intensity active surface modelling scheme is then applied to segment VOIs, followed by an alpha matting step to further refine the segmentation boundary. The proposed method is validated on real PET images of head-and-neck cancer patients with ground truth provided by human experts, as well as custom-designed phantom PET images with objective ground truth. Experimental results show that our method outperforms previous automatic approaches in terms of segmentation accuracy.     

Metastatic liver tumour segmentation with a neural network-guided 3D deformable model

The segmentation of liver tumours in CT images is useful for the diagnosis and treatment of liver cancer. Furthermore, an accurate assessment of tumour volume aids in the diagnosis and evaluation of treatment response. Currently, segmentation is performed manually by an expert, and because of the time required, a rough estimate of tumour volume is often done instead. We propose a semi-automatic segmentation method that makes use of machine learning within a deformable surface model. Specifically, we propose a deformable model that uses a voxel classifier based on a multilayer perceptron (MLP) to interpret the CT image. The new deformable model considers vertex displacement towards apparent tumour boundaries and regularization that promotes surface smoothness. During operation, a user identifies the target tumour and the mesh then automatically delineates the tumour from the MLP processed image. The method was tested on a dataset of 40 abdominal CT scans with a total of 95 colorectal metastases collected from a variety of scanners with variable spatial resolution. The segmentation results are encouraging with a Dice similarity metric of \(0.80 \pm 0.11\) and demonstrates that the proposed method can deal with highly variable data. This work motivates further research into tumour segmentation using machine learning with more data and deeper neural networks.     

A methodology for evaluation of boundary detection algorithms on breast ultrasound images

Image segmentation is the partition of an image into a set of non-overlapping regions that comprise the entire image. The image is decomposed into meaningful parts, which are uniform with respect to certain characteristics, such as grey level or texture. This study presents a novel methodology to evaluate ultrasound image segmentation algorithms. The sonographic features can differentiate between various sized malignant and benign breast tumours. The clinical experiment can determine whether a tumour is benign or not, based on contour, shape, echogenicity and echo texture. Further study of the standardized sonographic features, especially the tumour contour and shape, will improve the positive predictive value and accuracy rate in breast tumour detection. The effectiveness of using this methodology is illustrated by evaluating image segmentation on breast ultrasound images. Via definite segmentation, the appreciated tumour shape and contour can be ascertained. Furthermore, this method can enhance the ability of ultrasound to distinguish between benign and malignant breast lesions.     

Automatic Organ Segmentation for CT Scans Based on Super-Pixel and Convolutional Neural Networks

Accurate segmentation of specific organ from computed tomography (CT) scans is a basic and crucial task for accurate diagnosis and treatment. To avoid time-consuming manual optimization and to help physicians distinguish diseases, an automatic organ segmentation framework is presented. The framework utilized convolution neural networks (CNN) to classify pixels. To reduce the redundant inputs, the simple linear iterative clustering (SLIC) of super-pixels and the support vector machine (SVM) classifier are introduced. To establish the perfect boundary of organs in one-pixel-level, the pixels need to be classified step-by-step. First, the SLIC is used to cut an image into grids and extract respective digital signatures. Next, the signature is classified by the SVM, and the rough edges are acquired. Finally, a precise boundary is obtained by the CNN, which is based on patches around each pixel-point. The framework is applied to abdominal CT scans of livers and high-resolution computed tomography (HRCT) scans of lungs. The experimental CT scans are derived from two public datasets (Sliver 07 and a Chinese local dataset). Experimental results show that the proposed method can precisely and efficiently detect the organs. This method consumes 38 s/slice for liver segmentation. The Dice coefficient of the liver segmentation results reaches to 97.43%. For lung segmentation, the Dice coefficient is 97.93%. This finding demonstrates that the proposed framework is a favorable method for lung segmentation of HRCT scans.     

Wavelet energy-guided level set-based active contour: A segmentation method to segment highly similar regions

This paper introduces an approach to perform segmentation of regions in computed tomography (CT) images that exhibit intra-region intensity variations and at the same time have similar intensity distributions with surrounding/adjacent regions. In this work, we adapt a feature computed from wavelet transform called wavelet energy to represent the region information. The wavelet energy is embedded into a level set model to formulate the segmentation model called wavelet energy-guided level set-based active contour (WELSAC). The WELSAC model is evaluated using several synthetic and CT images focusing on tumour cases, which contain regions demonstrating the characteristics of intra-region intensity variations and having high similarity in intensity distributions with the adjacent regions. The obtained results show that the proposed WELSAC model is able to segment regions of interest in close correspondence with the manual delineation provided by the medical experts and to provide a solution for tumour detection.     

Registration-based Automatic 3D Segmentation of Cardiac CT Images

Segmenting whole heart from cardiac computed tomography(CT images can provide an important basis for the evaluation of cardiac function and help improve the accuracy of clinical diagnosis. Manual segmentation is the most accurate method for cardiac segmentation. But it is time consuming and not sufficiently reproducible. However, clinicians still rely on this method in practical applications. So a fully automatic method is needed to improve the segmentation efficiency. This pape proposes a registration-based automatic approach for three-dimensional(3D segmentation of cardiac CT images. The proposed method utilizes the similarity o cardiac CT images between different individuals, and uses registration to achieve the segmentation. Affine transformation is firstly implemented to achieve global coarse registration. Then, cubic B-splines are used to refine the local details in locally accurate registration. Mutual information(Ml) is used as the similarity measure, and adaptive stochastic gradient descent(ASGD) as the optimization algorithm. Ou method is applied to the dual-source cardiac CT images to segment whole heart Experimental results show that the proposed method can automatically segment whole heart from cardiac CT images.     

Applying Quantitative Radiographic Image Markers to Predict Clinical Complications After Aneurysmal Subarachnoid Hemorrhage: A Pilot Study

Accurately predicting clinical outcome of aneurysmal subarachnoid hemorrhage (aSAH) patients is difficult. The purpose of this study was to develop and test a new fully-automated computer-aided detection (CAD) scheme of brain computed tomography (CT) images to predict prognosis of aSAH patients. A retrospective dataset of 59 aSAH patients was assembled. Each patient had 2 sets of CT images acquired at admission and prior-to-discharge. CAD scheme was applied to segment intracranial brain regions into four subregions, namely, cerebrospinal fluid (CSF), white matter (WM), gray matter (GM), and leaked extraparenchymal blood (EPB), respectively. CAD then detects sulci and computes 9 image features related to 5 volumes of the segmented sulci, EPB, CSF, WM, and GM and 4 volumetrical ratios to sulci. Subsequently, applying a leave-one-case-out cross-validation method embedded with a principal component analysis (PCA) algorithm to generate optimal feature vector, 16 support vector machine (SVM) models were built using CT images acquired either at admission or prior-to-discharge to predict each of eight clinically relevant parameters commonly used to assess patients' prognosis. Finally, a receiver operating characteristics (ROC) method was used to evaluate SVM model performance. Areas under ROC curves of 16 SVM models range from 0.62 ± 0.07 to 0.86 ± 0.07. In general, SVM models trained using CT images acquired at admission yielded higher accuracy to predict short-term clinical outcomes, while SVM models trained using CT images acquired prior-to-discharge demonstrated higher accuracy in predicting long-term clinical outcomes. This study demonstrates feasibility to predict prognosis of aSAH patients using new quantitative image markers generated by SVM models.

     

Automated Lung Segmentation from HRCT Scans with Diffuse Parenchymal Lung Diseases

Performing accurate and fully automated lung segmentation of high-resolution computed tomography (HRCT) images affected by dense abnormalities is a challenging problem. This paper presents a novel algorithm for automated segmentation of lungs based on modified convex hull algorithm and mathematical morphology techniques. Sixty randomly selected lung HRCT scans with different abnormalities are used to test the proposed algorithm, and experimental results show that the proposed approach can accurately segment the lungs even in the presence of disease patterns, with some limitations in the apices and bases of lungs. The algorithm demonstrates a high segmentation accuracy (dice similarity coefficient?=?98.62 and shape differentiation metrics d_mean?=?1.39 mm, and d_rms?=?2.76 mm). Therefore, the developed automated lung segmentation algorithm is a good candidate for the first stage of a computer-aided diagnosis system for diffuse lung diseases.     

Automated lung nodule classification following automated nodule detection on CT: a serial approach

We have evaluated the performance of an automated classifier applied to the task of differentiating malignant and benign lung nodules in low-dose helical computed tomography (CT) scans acquired as part of a lung cancer screening program. The nodules classified in this manner were initially identified by our automated lung nodule detection method, so that the output of automated lung nodule detection was used as input to automated lung nodule classification. This study begins to narrow the distinction between the "detection task" and the "classification task." Automated lung nodule detection is based on two- and three-dimensional analyses of the CT image data. Gray-level-thresholding techniques are used to identify initial lung nodule candidates, for which morphological and gray-level features are computed. A rule-based approach is applied to reduce the number of nodule candidates that correspond to non-nodules, and the features of remaining candidates are merged through linear discriminant analysis to obtain final detection results. Automated lung nodule classification merges the features of the lung nodule candidates identified by the detection algorithm that correspond to actual nodules through another linear discriminant classifier to distinguish between malignant and benign nodules. The automated classification method was applied to the computerized detection results obtained from a database of 393 low-dose thoracic CT scans containing 470 confirmed lung nodules (69 malignant and 401 benign nodules). Receiver operating characteristic (ROC) analysis was used to evaluate the ability of the classifier to differentiate between nodule candidates that correspond to malignant nodules and nodule candidates that correspond to benign lesions. The area under the ROC curve for this classification task attained a value of 0.79 during a leave-one-out evaluation.     

Efficient COVID-19 Segmentation from CT Slices Exploiting Semantic Segmentation with Integrated Attention Mechanism

Coronavirus (COVID-19) is a pandemic, which caused suddenly unexplained pneumonia cases and caused a devastating effect on global public health. Computerized tomography (CT) is one of the most effective tools for COVID-19 screening. Since some specific patterns such as bilateral, peripheral, and basal predominant ground-glass opacity, multifocal patchy consolidation, crazy-paving pattern with a peripheral distribution can be observed in CT images and these patterns have been declared as the findings of COVID-19 infection. For patient monitoring, diagnosis and segmentation of COVID-19, which spreads into the lung, expeditiously and accurately from CT, will provide vital information about the stage of the disease. In this work, we proposed a SegNet-based network using the attention gate (AG) mechanism for the automatic segmentation of COVID-19 regions in CT images. AGs can be easily integrated into standard convolutional neural network (CNN) architectures with a minimum computing load as well as increasing model precision and predictive accuracy. Besides, the success of the proposed network has been evaluated based on dice, Tversky, and focal Tversky loss functions to deal with low sensitivity arising from the small lesions. The experiments were carried out using a fivefold cross-validation technique on a COVID-19 CT segmentation database containing 473 CT images. The obtained sensitivity, specificity, and dice scores were reported as 92.73%, 99.51%, and 89.61%, respectively. The superiority of the proposed method has been highlighted by comparing with the results reported in previous studies and it is thought that it will be an auxiliary tool that accurately detects automatic COVID-19 regions from CT images.     

Classification of Breast Masses Using Selected Shape,Edge-sharpness,and Texture Features with Linear and Kernel-based Classifiers

Breast masses due to benign disease and malignant tumors related to breast cancer differ in terms of shape, edge-sharpness, and texture characteristics. In this study, we evaluate a set of 22 features including 5 shape factors, 3 edge-sharpness measures, and 14 texture features computed from 111 regions in mammograms, with 46 regions related to malignant tumors and 65 to benign masses. Feature selection is performed by a genetic algorithm based on several criteria, such as alignment of the kernel with the target function, class separability, and normalized distance. Fisher's linear discriminant analysis, the support vector machine (SVM), and our strict two-surface proximal (S2SP) classifier, as well as their corresponding kernel-based nonlinear versions, are used in the classification task with the selected features. The nonlinear classification performance of kernel Fisher's discriminant analysis, SVM, and S2SP, with the Gaussian kernel, reached 0.95 in terms of the area under the receiver operating characteristics curve. The results indicate that improvement in classification accuracy may be gained by using selected combinations of shape, edge-sharpness, and texture features.     

Model-Based Pancreas Segmentation in Portal Venous Phase Contrast-Enhanced CT Images

This study aims to automatically detect and segment the pancreas in portal venous phase contrast-enhanced computed tomography (CT) images. The institutional review board of the University of Erlangen-Nuremberg approved this study and waived the need for informed consent. Discriminative learning is used to build a pancreas tissue classifier incorporating spatial relationships between the pancreas and surrounding organs and vessels. Furthermore, discrete cosine and wavelet transforms are used to build texture features to describe local tissue appearance. Classification is used to guide a constrained statistical shape model to fit the data. The algorithm to detect and segment the pancreas was evaluated on 40 consecutive CT data that were acquired in the portal venous contrast agent phase. Manual segmentation of the pancreas was carried out by experienced radiologists and served as reference standard. Threefold cross validation was performed. The algorithm-based detection and segmentation yielded an average surface distance of 1.7 mm and an average overlap of 61.2 % compared with the reference standard. The overall runtime of the system was 20.4 min. The presented novel approach enables automatic pancreas segmentation in portal venous phase contrast-enhanced CT images which are included in almost every clinical routine abdominal CT examination. Reliable pancreatic segmentation is crucial for computer-aided detection systems and an organ-specific decision support.     

Learning image context for segmentation of the prostate in CT-guided radiotherapy

Accurate segmentation of the prostate is the key to the success of external beam radiotherapy of prostate cancer. However, accurate segmentation of the prostate in computer tomography (CT) images remains challenging mainly due to three factors: (1) low image contrast between the prostate and its surrounding tissues, (2) unpredictable prostate motion across different treatment days and (3) large variations of intensities and shapes of the bladder and rectum around the prostate. In this paper, an online-learning and patient-specific classification method based on the location-adaptive image context is presented to deal with all these challenging issues and achieve the precise segmentation of the prostate in CT images. Specifically, two sets of location-adaptive classifiers are placed, respectively, along the two coordinate directions of the planning image space of a patient, and further trained with the planning image and also the previous-segmented treatment images of the same patient to jointly perform prostate segmentation for a new treatment image (of the same patient). In particular, each location-adaptive classifier, which itself consists of a set of sequential sub-classifiers, is recursively trained with both the static image appearance features and the iteratively updated image context features (extracted at different scales and orientations) for better identification of each prostate region. The proposed learning-based prostate segmentation method has been extensively evaluated on 161 images of 11 patients, each with more than nine daily treatment three-dimensional CT images. Our method achieves the mean Dice value 0.908 and the mean ± SD of average surface distance value 1.40 ± 0.57 mm. Its performance is also compared with three prostate segmentation methods, indicating the best segmentation accuracy by the proposed method among all methods under comparison.     

A Knowledge-Based Approach for Carpal Tunnel Segmentation from Magnetic Resonance Images

Carpal tunnel syndrome (CTS) has been reported as one of the most common peripheral neuropathies. Carpal tunnel segmentation from magnetic resonance (MR) images is important for the evaluation of CTS. To date, manual segmentation, which is time-consuming and operator dependent, remains the most common approach for the analysis of the carpal tunnel structure. Therefore, we propose a new knowledge-based method for automatic segmentation of the carpal tunnel from MR images. The proposed method first requires the segmentation of the carpal tunnel from the most proximally cross-sectional image. Three anatomical features of the carpal tunnel are detected by watershed and polygonal curve fitting algorithms to automatically initialize a deformable model as close to the carpal tunnel in the given image as possible. The model subsequently deforms toward the tunnel boundary based on image intensity information, shape bending degree, and the geometry constraints of the carpal tunnel. After the deformation process, the carpal tunnel in the most proximal image is segmented and subsequently applied to a contour propagation step to extract the tunnel contours sequentially from the remaining cross-sectional images. MR volumes from 15 subjects were included in the validation experiments. Compared with the ground truth of two experts, our method showed good agreement on tunnel segmentations by an average margin of error within 1 mm and dice similarity coefficient above 0.9.     

Segmentation accuracy of long bones

The use of three-dimensional imaging methodologies in new applications in the orthopaedic field has introduced a need for high accuracy, in addition to a correct diagnosis. The aim of this study was to quantify the absolute dimensional errors between models reconstructed from computed tomography and magnetic resonance images compared to a ground truth for various regions of the bone. Clinical CT and MRI scans were acquired from nine lower leg cadavers and the bones were subsequently cleaned from soft tissues. 3D models of the tibia were created from the segmented CT and MRI images and compared to optical scans of the cleaned bones (considered as ground truth). The 3D reconstruction using CT images resulted in an RMS error of 0.55 mm, corresponding to an overestimated CT bone model compared to the cleaned bone. MR imaging resulted in an RMS error of 0.56 mm; however, the MRI bone model was on average a small underestimation of the cleaned bone. Different regions of the bones were analysed, indicating a difference in accuracy between diaphysis and epiphysis. This study demonstrates a high accuracy for both CT and MRI imaging, supporting the feasibility of using MRI technology for the 3D reconstruction of bones in medical applications.     

Knowledge-based segmentation of thoracic computed tomography images for assessment of split lung function

The assessment of differential left and right lung function is important for patients under consideration for lung resection procedures such as single lung transplantation. We developed an automated, knowledge-based segmentation algorithm for purposes of deriving functional information from dynamic computed tomography (CT) image data. Median lung attenuation (HU) and area measurements were automatically calculated for each lung from thoracic CT images acquired during a forced expiratory maneuver as indicators of the amount and rate of airflow. The accuracy of these derived measures from fully automated segmentation was validated against those from segmentation using manual editing by an expert observer. A total of 1313 axial images were analyzed from 49 patients. The images were segmented using our knowledge-based system that identifies the chest wall, mediastinum, trachea, large airways and lung parenchyma on CT images. The key components of the system are an anatomical model, an inference engine and image processing routines, and segmentation involves matching objects extracted from the image to anatomical objects described in the model. The segmentation results from all images were inspected by the expert observer. Manual editing was required to correct 183 (13.94%) of the images, and the sensitivity, specificity, and accuracy of the knowledge-based segmentation were greater than 98.55% in classifying pixels as lung or nonlung. There was no significant difference between median lung attenuation or area values from automated and edited segmentations (p > 0.70). Using the knowledge-based segmentation method we can automatically derive indirect quantitative measures of single lung function that cannot be obtained using conventional pulmonary function tests.     

Comparative Performance Analysis of State-of-the-Art Classification Algorithms Applied to Lung Tissue Categorization

In this paper, we compare five common classifier families in their ability to categorize six lung tissue patterns in high-resolution computed tomography (HRCT) images of patients affected with interstitial lung diseases (ILD) and with healthy tissue. The evaluated classifiers are naive Bayes, k-nearest neighbor, J48 decision trees, multilayer perceptron, and support vector machines (SVM). The dataset used contains 843 regions of interest (ROI) of healthy and five pathologic lung tissue patterns identified by two radiologists at the University Hospitals of Geneva. Correlation of the feature space composed of 39 texture attributes is studied. A grid search for optimal parameters is carried out for each classifier family. Two complementary metrics are used to characterize the performances of classification. These are based on McNemar’s statistical tests and global accuracy. SVM reached best values for each metric and allowed a mean correct prediction rate of 88.3% with high class-specific precision on testing sets of 423 ROIs.     

Automated curved planar reformation of 3D spine images

Traditional techniques for visualizing anatomical structures are based on planar cross-sections from volume images, such as images obtained by computed tomography (CT) or magnetic resonance imaging (MRI). However, planar cross-sections taken in the coordinate system of the 3D image often do not provide sufficient or qualitative enough diagnostic information, because planar cross-sections cannot follow curved anatomical structures (e.g. arteries, colon, spine, etc). Therefore, not all of the important details can be shown simultaneously in any planar cross-section. To overcome this problem, reformatted images in the coordinate system of the inspected structure must be created. This operation is usually referred to as curved planar reformation (CPR). In this paper we propose an automated method for CPR of 3D spine images, which is based on the image transformation from the standard image-based to a novel spine-based coordinate system. The axes of the proposed spine-based coordinate system are determined on the curve that represents the vertebral column, and the rotation of the vertebrae around the spine curve, both of which are described by polynomial models. The optimal polynomial parameters are obtained in an image analysis based optimization framework. The proposed method was qualitatively and quantitatively evaluated on five CT spine images. The method performed well on both normal and pathological cases and was consistent with manually obtained ground truth data. The proposed spine-based CPR benefits from reduced structural complexity in favour of improved feature perception of the spine. The reformatted images are diagnostically valuable and enable easier navigation, manipulation and orientation in 3D space. Moreover, reformatted images may prove useful for segmentation and other image analysis tasks.