Distance Regularized Level Set Evolution Research Articles

Level-set evolution for medical image segmentation with alternating direction method of multipliers

Segmentation via level-set methods plays a significant role in natural and medical image analysis. Recently, the distance regularized level-set evolution (DRLSE) model, which is based on variational level-set formulation has been proven very useful. However, the DRLSE model suffers from difficulties related to its computational cost due to the sign distance function (SDF) and Courant-Friedrichs-Lewy (CFL) condition applied by the gradient descent method. This paper analyzes and tests the well-known fast computing algorithm known as alternating direction method of multipliers (ADMM) for level-set based image segmentation. That is, we compute the DRLSE model by using the ADMM algorithm. The comparison between the classical gradient descent method and the proposed ADMM for two-phase and multi-phase level-set formulation is provided. The numerical experiments on standard public data sets, particularly in the case of medical image segmentation, demonstrate the effectiveness and efficiency of the proposed method. The average segmentation coefficients for Dice and Jaccard against ground truth are 0.97 and 0.92, respectively, and the average running time is 1.70 seconds. 0.0932, 0.993, 0.981, and 0.964 are average estimated values for mean absolute distance (MAD), accuracy, sensitivity, and specificity, respectively. The source code can be found at link https://github.com/WaliSamad/DRLSE-ADMM-Medical-Image-Segmentation.

Brain fMRI segmentation under emotion stimuli incorporating attention-based deep convolutional neural networks

Functional magnetic resonance imaging (fMRI) is widely used for clinical examinations, diagnosis, and treatment. By segmenting fMRI images, large-scale medical image data can be processed more efficiently. Most deep learning (DL)-based segmentation typically uses some type of encoding–decoding model. In this study, affective computing (AC) was developed using the brain fMRI dataset generated from an emotion simulation experiment. The brain fMRI dataset was segmented using an attention model, a deep convolutional neural network-32 (DCNN-32) based on Laplacian of Gaussian (LoG) filter, called ADCNN-32-G. For the evaluation of image segmentation, several indices are presented. By comparing the proposed ADCNN-32s-G model to distance regularized level set evolution (DRLSE), single-seeded region growing, and the single segNet full convolutional network model (FCN), the proposed model performs well in segmenting mass fMRI datasets. The proposed method can be applied to the real-time monitoring of patients with depression, and it can effectively advise human mental health.

Intelligent localization and quantitative evaluation of anterior talofibular ligament injury using magnetic resonance imaging of ankle

BackgroundThere is a high incidence of injury to the lateral ligament of the ankle in daily living and sports activities. The anterior talofibular ligament (ATFL) is the most frequent types of ankle injuries. It is of great clinical significance to achieve intelligent localization and injury evaluation of ATFL due to its vulnerability.MethodsAccording to the specific characteristics of bones in different slices, the key slice was extracted by image segmentation and characteristic analysis. Then, the talus and fibula in the key slice were segmented by distance regularized level set evolution (DRLSE), and the curvature of their contour pixels was calculated to find useful feature points including the neck of talus, the inner edge of fibula, and the outer edge of fibula. ATFL area can be located using these feature points so as to quantify its first-order gray features and second-order texture features. Support vector machine (SVM) was performed for evaluation of ATFL injury.ResultsData were collected retrospectively from 158 patients who underwent MRI, and were divided into normal (68) and tear (90) group. The positioning accuracy and Dice coefficient were used to measure the performance of ATFL localization, and the mean values are 87.7% and 77.1%, respectively, which is helpful for the following feature extraction. SVM gave a good prediction ability with accuracy of 93.8%, sensitivity of 88.9%, specificity of 100%, precision of 100%, and F1 score of 94.2% in the test set.ConclusionExperimental results indicate that the proposed method is reliable in diagnosing ATFL injury. This study may provide a potentially viable method for aided clinical diagnoses of some ligament injury.

AW-SDRLSE: Adaptive Weighting and Scalable Distance Regularized Level Set Evolution for Lymphoma Segmentation on PET Images.

Accurate lymphoma segmentation on Positron Emission Tomography (PET) images is of great importance for medical diagnoses, such as for distinguishing benign and malignant. To this end, this paper proposes an adaptive weighting and scalable distance regularized level set evolution (AW-SDRLSE) method for delineating lymphoma boundaries on 2D PET slices. There are three important characteristics with respect to AW-SDRLSE: 1) A scalable distance regularization term is proposed and a parameter q can control the contour's convergence rate and precision in theory. 2) A novel dynamic annular mask is proposed to calculate mean intensities of local interior and exterior regions and further define the region energy term. 3) As the level set method is sensitive to parameters, we thus propose an adaptive weighting strategy for the length and area energy terms using local region intensity and boundary direction information. AW-SDRLSE is evaluated on 90 cases of real PET data with a mean Dice coefficient of 0.8796. Comparative results demonstrate the accuracy and robustness of AW-SDRLSE as well as its performance advantages as compared with related level set methods. In addition, experimental results indicate that AW-SDRLSE can be a fine segmentation method for improving the lymphoma segmentation results obtained by deep learning (DL) methods significantly.

SHAPE CHARACTERIZATION OF MEDIASTINUM IN TUBERCULOSIS CHEST RADIOGRAPHS USING LEVEL SET SEGMENTATION

Mediastinum is considered as one of the substantial anatomical regions for the gross diagnosis of several chest related pathologies. The geometric variations of the mediastinum in Chest Radiographs (CXRs) could be utilised as potential image markers in the early detection of Tuberculosis (TB). This study attempts to segment mediastinum in CXRs using level sets for the shape characterization of TB conditions. The CXR images for this study are considered from a public database. An edge-based distance regularized level set evolution is employed to segment the lungs followed by a region-based Chan-Vese model that extracts mediastinum region. Features such as mediastinum area and lungs area are extracted from the segmented images. Further, mediastinum to lungs area ratio is calculated. Statistical analysis is performed on the features to differentiate normal and TB images. Results show that the proposed segmentation approach is able to segment the lungs and extract the mediastinum in CXRs. It is found that features namely mediastinum area and mediastinum to lungs area ratio are statistically significant in the differentiation of TB. Larger mediastinum area is observed in TB images as compared to normal. The performance of lung field segmentation is also observed to be in line with the literature. The mediastinum segmentation approach in CXRs obtains to be a novel method as compared to the existing methods. As the proposed approach based on mediastinum image analysis provides better shape characterization, the study could be clinically useful in the differentiation of TB conditions.

A fast, edge-preserving, distance-regularized model with bilateral filtering for oil spill segmentation of SAR images

Marine oil spills are among the most significant sources of marine pollution. Synthetic aperture radar (SAR) has been used to improve oil spill observations because of its advantages in oil spill detection and identification. However, speckle noise, weak boundaries, and intensity inhomogeneity often exist in the oil spill regions of SAR imagery, which will seriously affect the accurate identification of oil spills. To enhance marine oil spill segmentation of SAR images, a fast, edge-preserving framework based on the distance-regularized level set evolution (DRLSE) model was proposed. Specifically, a bilateral filter penalty term is designed and incorporated into the DRLSE energy function (BF-DRLSE) to preserve the edges of oil spills, and an adaptive initial box boundary was selected for the DRLSE model to reduce the operation time complexity. Two sets of RadarSat-2 SAR data were used to test the proposed method. The experimental results indicate that the bilateral filtering scheme incorporated into the energy function during level set evolution improved the stability of level set evolution. Compared with other methods, the proposed improved BF-DRLSE algorithm displayed a higher overall segmentation accuracy (97.83%). In addition, using an appropriate initial box boundary for the DRLSE method accelerated the global search process, improved the accuracy of oil spill segmentation, and reduced computational time. Therefore, the results suggest that the proposed framework is effective and applicable for marine oil spill segmentation.

Modified distance regularized level set evolution for brain ventricles segmentation

Neurodegenerative disorders are commonly characterized by atrophy of the brain which is caused by neuronal loss. Ventricles are one of the prominent structures in the brain; their shape changes, due to their content, the cerebrospinal fluid. Analyzing the morphological changes of ventricles, aids in the diagnosis of atrophy, for which the region of interest needs to be separated from the background. This study presents a modified distance regularized level set evolution segmentation method, incorporating regional intensity information. The proposed method is implemented for segmenting ventricles from brain images for normal and atrophy subjects of magnetic resonance imaging and computed tomography images. Results of the proposed method were compared with ground truth images and produced sensitivity in the range of 65%–90%, specificity in the range of 98%–99%, and accuracy in the range of 95%–98%. Peak signal to noise ratio and structural similarity index were also used as performance measures for determining segmentation accuracy: 95% and 0.95, respectively. The parameters of level set formulation vary for different datasets. An optimization procedure was followed to fine tune parameters. The proposed method was found to be efficient and robust against noisy images. The proposed method is adaptive and multimodal.

Pyramid graph cut: Integrating intensity and gradient information for grayscale medical image segmentation

Segmentation of grayscale medical images is challenging because of the similarity of pixel intensities and poor gradient strength between adjacent regions. The existing image segmentation approaches based on either intensity or gradient information alone often fail to produce accurate segmentation results. Previous approaches in the literature have approached the problem by embedded or sequential integration of different information types to improve the performance of the image segmentation on specific tasks. However, an effective combination or integration of such information is difficult to implement and not sufficiently generic for closely related tasks. Integration of the two information sources in a single graph structure is a potentially more effective way to solve the problem. In this paper we introduce a novel technique for grayscale medical image segmentation called pyramid graph cut, which combines intensity and gradient sources of information in a pyramid-shaped graph structure using a single source node and multiple sink nodes. The source node, which is the top of the pyramid graph, embeds intensity information into its linked edges. The sink nodes, which are the base of the pyramid graph, embed gradient information into their linked edges. The min-cut uses intensity information and gradient information, depending on which one is more useful or has a higher influence in each cutting location of each iteration. The experimental results demonstrate the effectiveness of the proposed method over intensity-based segmentation alone (i.e. Gaussian mixture model) and gradient-based segmentation alone (i.e. distance regularized level set evolution) on grayscale medical image datasets, including the public 3DIRCADb-01 dataset. The proposed method archives excellent segmentation results on the sample CT of abdominal aortic aneurysm, MRI of liver tumor and US of liver tumor, with dice scores of 90.49±5.23%, 88.86±11.77%, 90.68±2.45%, respectively.

Oriented distance regularized level set evolution for image segmentation

AbstractThe conventional distance regularized level set evolution method has been very popular in image segmentation, but usually it cannot converge to the desired boundary when there are multiple and unwanted boundaries in the image. By observation, the gradient direction between the target boundaries and the unwanted boundaries are usually different in one image. The gradient direction information of the boundaries can guide the orientation of the level set function evolution. In this study, the authors improved the conventional distance regularized level set evolution method, introduced new edge indicator functions and proposed an oriented distance regularized level set evolution method for image segmentation. The experiment results show the proposed method has a better segmentation result in images with multiple boundaries. Moreover, alternately selecting the edge indicator functions we proposed during the level set evolution can lead the zero level set contour to converge to the desired boundaries in complicated images.

Superpixel Approach in Computer-Aided Detection System of Dental Cavities in X-Ray Images

Dental caries can be characterized and segmented using a computer-aided diagnosis (CADx) system. In most CADx systems, segmentation is a key step in identifying dental cavities at an early stage. With well-segmented lesions, dentists can provide accurate diagnoses. This study proposes a semi-automated method based on superpixel segmentation to identify dental cavities in digital X-ray images. Superpixel segmentation partitions an image into multiple homogeneous segments, wherein the pixels of each segment contain certain comparable characteristics. The proposed method is compared with two previous methods; in one method, an active contour model, or snake, iteratively performs the initial contour, self-examination, and correction of the segmentation results, whereas in the other method, distance regularized level set evolution (DRLSE) eliminates the need for re-initialization, thereby avoiding its induced numerical errors. The proposed method combines DRLSE with superpixels. For the comparison of these three algorithms, 11 authentic teeth, ranging from premolars to molars, were used. With regard to the detection results and classification accuracies, the proposed superpixel DRLSE method received lower error metric scores than the other methods.

Fractional Distance Regularized Level Set Evolution With Its Application to Image Segmentation

To avoid the irregularities during the level set evolution, a fractional distance regularized variational model is proposed for image segmentation. We first define a fractional distance regularization term which punishes the deviation of the level set function (LSF) and the signed distance function. Since the fractional derivative of the constant value function outside the starting point is nonzero, the fractional gradient modular of the LSF does not approach infinity where the integer order gradient modular is close to 0. This prevents the sharp reverse diffusion of LSF in flat areas and ensures the smooth evolution of LSF. Then, we use the Grunwald-Letnikov (G-L) fractional derivative to derive the discrete forms of the conjugate of fractional derivatives and fractional divergence. To facilitate the calculation of fractional derivatives and their conjugates, we designed their covering templates. Finally, a numerical solution to the minimization of the energy functional is obtained from these discrete forms and covering templates. Numerical experiments of medical images with different modalities show that the model in this paper can well segment weak images and intensity inhomogeneity images.

Active contour model based on improved fuzzy c-means algorithm and adaptive functions

Distance regularized level set evolution (DRLSE) model, which solves the re-initialization problem in early active contours, is a ground breaking edge-based model for image segmentation. However, it has the disadvantages of unsatisfactory robustness to initialization and noise, unidirectional movement, slow convergence and poor stability. In this paper, we propose an active contour model driven by improved fuzzy c-means algorithm (FCM) and adaptive functions. An adaptive sign function based on image clustering information not only increases stability, but also solves the problem of unidirectional movement. Furthermore, it gives our model the ability to selectively segment targets in image. An adaptive edge indicator function accelerates convergence with better function performance. To further increase stability, a novel double-well potential function and the corresponding evolution speed function are proposed. Due to the improved FCM, the proposed model is robust to initialization and noise. In addition, our model exhibits an edge-based and region-based characteristic. Experimental results have proved that the proposed model can not only effectively segment images with intensity inhomogeneity, but also show a good robustness to initialization. Moreover, it has shorter time spent and higher segmentation accuracy compared with other models.

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.

Active contours driven by adaptive functions and fuzzy c-means energy for fast image segmentation

The shortcoming of geometric active contours is that they are extremely sensitive to positions of initialization. Initial contours of the distance regularized level set evolution model must be set inside or outside target completely. In addition, the model is prone to problems of falling into false boundary, leaking from weak edge and poor anti-noise ability. In this paper, we propose a robust active contour model driven by adaptive functions (including an adaptive edge indicator function and adaptive sign function) and fuzzy c-means energy. Utilize the adaptive edge indicator function which is composed of image intensity information to substitute traditional edge indicator function in the area term. Active contours can expand or shrink from initialization automatically, which improves the disadvantages of poor robustness to initialization and unidirectional movement. Due to the adaptive sign function and fuzzy c-means energy, the problems of slow convergence and leaking from weak edge have been solved. Moreover, a novel distance regularized term (mainly a potential function and evolution speed function) is proposed to make evolution more stable. Experimental results have proved that our model can segment images with intensity inhomogeneity effectively. Compared with other classic models, the proposed model not only shortens time spent and improves segmentation accuracy, but also shows a better robustness to initialization.

A study on the Development of a Korean Metabolic Syndrome Questionnaire

Segmentation of grayscale medical images is challenging because of the similarity of pixel intensities and poor gradient strength between adjacent regions. The existing image segmentation approaches based on either intensity or gradient information alone often fail to produce accurate segmentation results. Previous approaches in the literature have approached the problem by embedded or sequential integration of different information types to improve the performance of the image segmentation on specific tasks. However, an effective combination or integration of such information is difficult to implement and not sufficiently generic for closely related tasks. Integration of the two information sources in a single graph structure is a potentially more effective way to solve the problem. In this paper we introduce a novel technique for grayscale medical image segmentation called pyramid graph cut, which combines intensity and gradient sources of information in a pyramid-shaped graph structure using a single source node and multiple sink nodes. The source node, which is the top of the pyramid graph, embeds intensity information into its linked edges. The sink nodes, which are the base of the pyramid graph, embed gradient information into their linked edges. The min-cut uses intensity information and gradient information, depending on which one is more useful or has a higher influence in each cutting location of each iteration. The experimental results demonstrate the effectiveness of the proposed method over intensity-based segmentation alone (i.e. Gaussian mixture model) and gradient-based segmentation alone (i.e. distance regularized level set evolution) on grayscale medical image datasets, including the public 3DIRCADb-01 dataset. The proposed method archives excellent segmentation results on the sample CT of abdominal aortic aneurysm, MRI of liver tumor and US of liver tumor, with dice scores of 90.49±5.23%, 88.86±11.77%, 90.68±2.45%, respectively.

Automated Extraction of Parasite in the Microscopic Images by Distance Regularized Level Set Evolution initialized with Hough Transform

Automated Extraction of Parasite in the Microscopic Images by Distance Regularized Level Set Evolution initialized with Hough Transform

3D segmentation of exterior wall surface of abdominal aortic aneurysm from CT images using variable neighborhood search

A 3D model of abdominal aortic aneurysm (AAA) can provide useful anatomical information for clinical management and simulation. Thin-slice contiguous computed tomographic (CT) angiography is the best source of medical images for construction of 3D models, which requires segmentation of AAA in the images. Existing methods for segmentation of AAA rely on either manual process or 2D segmentation in each 2D CT slide. However, a traditional manual segmentation is a time consuming process which is not practical for routine use. The construction of a 3D model from 2D segmentation of each CT slice is not a fully satisfactory solution due to rough contours that can occur because of lack of constraints among segmented slices, as well as missed segmentation slices. To overcome such challenges, this paper proposes the 3D segmentation of AAA using the concept of variable neighborhood search by iteratively alternating between two different segmentation techniques in the two different 3D search spaces of voxel intensity and voxel gradient. The segmentation output of each method is used as the initial contour to the other method in each iteration. By alternating between search spaces, the technique can escape local minima that naturally occur in each search space. Also, the 3D search spaces provide more constraints across CT slices, when compared with the 2D search spaces in individual CT slices. The proposed method is evaluated with 10 easy and 10 difficult cases of AAA. The results show that the proposed 3D segmentation technique achieves the outstanding segmentation accuracy with an average dice similarity value (DSC) of 91.88%, when compared to the other methods using the same dataset, which are the 2D proposed method, classical graph cut, distance regularized level set evolution, and registration based geometric active contour with the DSCs of 87.57 ± 4.52%, 72.47 ± 8.11%, 58.50 ± 8.86% and 76.21 ± 10.49%, respectively.

Automated Parasite’s Detection in Microscopic Images of Stools Using Distance Regularized Level Set Evolution Initialized with Hough Transform

Background and purpose: The analysis of biomedical microscopic images is carried out manually in medical laboratories. The manual analysis of clinical images lets to both repetitive tasks and management of huge amounts of data. This is tedious and times consuming for laboratory technicians. Inevitably, it is also prone to human errors. Our objective in this work is to contribute to the automation of the analysis of microscopic images of stools using Distance Regularized Level Set Evolution automatically initialized by Hough transform. Method: We firstly converted the microscopic images to edge maps using canny algorithm. Next, we located the parasite through circular Hough transform and draw circles around them. Those circles stand as initial contours of DRLSE. The contours evolve until they fit the boundaries of the parasites. The final extraction is performed using a complementary method based on the signed distance character of the level set function. Results: The Distance Regularized Level Set Evolution has been automatically initialized. We applied our method to the detection of intestinal parasites in microscopic images. Experimental results show accurate, efficient and less time consuming of our scheme compared to others recently proposed in the literature. Conclusion: This is a notable contribution to the automation of stools examination in the medical laboratories. In forthcoming works, we plan to include this segmentation process in an expert system of parasitic diseases diagnosis.

A Distance Regularized Level-Set Evolution Model Based MRI Dataset Segmentation of Brain’s Caudate Nucleus

The caudate nucleus of the brain is highly correlated to the emotional decision-making of pessimism, which is an important process for improving the understanding and treatment of depression; and the segmentation of the caudate nucleus is the most basic step in the process of analysis and research concerning this region. In this paper, Level Set Method (LSM) is applied for caudate nucleus segmentation. Firstly, Distance Regularized Level Set Evolution (DRLSE), Region-Scalable Fitting (RSF) and Local Image Fitting (LIF) models are proposed for segmentation of the caudate nucleus of Magnetic Resonance Imaging (MRI) images of the brain, and the segmentation results are compared by using selected evaluation indices. The average Dice Similarity Coefficient (DSC) values of the proposed three methods all exceed 85%, and the average Jaccard Similarity (JS) values are over 77%, respectively. The results indicate that all these three models can have good segmentation results for medical images with intensity inhomogeneity and meet the general segmentation requirements, while the proposed DRLSE model performs better in segmentation.

Automating the clinical stools exam using image processing integrated in an expert system

Abstract Background and objective The diagnosis of intestinal parasitosis disease relies on physiological symptoms and stool examination. Often, few specialists are available, and manual stool exam is slow, prone to error, and can cause eye fatigue. Our aim was to design and implement a medical expert system that would be automated and helpful for diagnosis of human intestinal parasitosis. Methods The system was developed based on a decision algorithm. A knowledge base was constructed through information gleaned from books and physicians with information pertaining to the disease. The user interacts with the system by answering questions. The symptoms information collected led to a microscopic examination of stools, which was run on the system to detect parasites. The paradigm for automated microscopic examination of stools consisted of a combined distance regularized level set evolution, automatically initialized by a circular Hough transform, and a trained neuro-fuzzy classifier. The neuro-fuzzy classifier was trained for analysis of twenty human intestinal parasites. Results We combined the reasoning scheme of diagnosis and the automated clinical exam of stools in the same system. The parasites found in microscopic imagery confirmed the suspicious disease. The final recommendation of diagnosis was then completed, with appropriate proposed therapy. The system was evaluated with sixty cases of infection, and compared to the diagnosis of two expert doctors; we obtained fifty eight correct diagnoses, corresponding to a 96.6% accuracy. Conclusions The proposed system is automated, since the parameters of segmentation, feature extraction and classification are set to be computationally guided by the type of suspicious parasite. The system is potentially an important contribution for medical healthcare assistance.