Level Set Method For Segmentation Research Articles

Corona virus disease 2019 lesion segmentation network based on an adaptive joint loss function

Corona virus disease 2019 (COVID-19) is an acute respiratory infectious disease with strong contagiousness, strong variability, and long incubation period. The probability of misdiagnosis and missed diagnosis can be significantly decreased with the use of automatic segmentation of COVID-19 lesions based on computed tomography images, which helps doctors in rapid diagnosis and precise treatment. This paper introduced the level set generalized Dice loss function (LGDL) in conjunction with the level set segmentation method based on COVID-19 lesion segmentation network and proposed a dual-path COVID-19 lesion segmentation network (Dual-SAUNet++) to address the pain points such as the complex symptoms of COVID-19 and the blurred boundaries that are challenging to segment. LGDL is an adaptive weight joint loss obtained by combining the generalized Dice loss of the mask path and the mean square error of the level set path. On the test set, the model achieved Dice similarity coefficient of (87.81 ± 10.86)%, intersection over union of (79.20 ± 14.58)%, sensitivity of (94.18 ± 13.56)%, specificity of (99.83 ± 0.43)% and Hausdorff distance of 18.29 ± 31.48 mm. Studies indicated that Dual-SAUNet++ has a great anti-noise capability and it can segment multi-scale lesions while simultaneously focusing on their area and border information. The method proposed in this paper assists doctors in judging the severity of COVID-19 infection by accurately segmenting the lesion, and provides a reliable basis for subsequent clinical treatment.

Modified convolutional neural network for lung cancer detection: Improved cat swarm-based optimal training

Lung cancer is the most lethal and severe illness in existence. However, lung cancer patients may live longer if they receive early detection and treatment. In the medical field, the best imaging technique is CT scan imaging as it is more complex for doctors to identify cancer and interpret from CT scan images. Consequently, the computer-aided diagnosis (CAD) is more useful for doctors to find out cancerous nodules. To identify lung cancer, a number of CAD techniques utilising machine learning (ML) and image processing are used nowadays. The goal of this study is to present a novel method for detecting lung cancer that entails four main steps: (i) Pre-processing, (ii) Segmentation, (iii) Feature extraction, and (iv) Classification. ”The input image is first put through a pre-processing step in which the CLAHE model is used to pre-process the image. The segmentation phase of the pre-processed images is then initiated, and it makes use of a modified Level set segmentation method. The retrieved features from the segmented images include statistical features, colour features, and texture features (GLCM, GLRM, and LBP). The Layer Fused Conventional Neural Network (LF-CNN) is then utilised to classify these features in the end. Particularly, layer-wise modification is carried out, and along with that, the LF-CNN is trained by the Modified Cat swarm Optimization (MCSO) Algorithm via selecting optimal weights. The accepted scheme is then compared to the current models in terms of several metrics, including recall, FNR, MCC, FDR, Threat score, FPR, precision, FOR, accuracy, specificity, NPV, FMS, and sensitivity.

An underwater small target boundary segmentation method in forward-looking sonar images

Underwater target boundary segmentation is integral to forward-looking sonar image processing. However, small target boundary segmentation is always challenging due to the low resolution, low signal-to-noise ratio, and inhomogeneity of intensity of forward-looking sonar images. This paper proposes an improved level set segmentation method to accurately obtain the contour of small targets in the forward-looking sonar images: morphological reconstruction combined with the level set method (MRLSM). Compared with the classical level set and the level set method combined with the morphological method, MRLSM improves the accuracy and stability for forward-looking sonar image segmentation. Furthermore, the fuzzy C-means, Markov random field, and MRLSM are applied to the numerical simulations and experimental data for comparison. The deep learning methods are also used to compare the performances. The results demonstrated that the proposed method is more accurate, robust, and considerable for underwater small target boundary segmentation.

Sea-Crossing Bridge Detection in Polarimetric SAR Images Based on Windowed Level Set Segmentation and Polarization Parameter Discrimination

As sea-crossing bridges are important hubs connecting separated land areas, their detection in SAR images is of great significance. However, under complex scenarios, the sea surface conditions, the distribution of coastal terrain morphologies, and the scattering components of different structures in the bridge area are very complex and diverse, which makes the accurate and robust detection of sea-crossing bridges difficult, including the sea–land segmentation and bridge feature extraction on which the detection depends. In this paper, we propose a polarimetric SAR image detection method for sea-crossing bridges based on windowed level set segmentation and polarization parameter discrimination. Firstly, the sea and land are segmented by a proposed windowed level set segmentation method, which replaces the construction of the level set segmentation energy function based on the isolated pixel distribution with a joint distribution of pixels in a certain window region. Secondly, water regions of interest are extracted by a proposed water region merging algorithm combining the distances of the water contour and polarization similarity parameter. Finally, the bridge regions of interest (ROIs) are extracted by merging close water contours, and the ROIs are discriminated by the polarimetric parameters of the polarization entropy and scattering angle. Experimental results using multiple AirSAR, RADARSAT-2, and TerraSAR-X quad-polarization SAR data from the coastal areas of San Francisco in the USA, Singapore, and Fuzhou, Fujian, and Zhanjiang, Guangdong, in China show that the proposed method can achieve 100% detection of sea-crossing bridges in different bands for different scenes, and the accuracy of the intersection of the ground-truth (IoG) index of bridge body recognition can reach more than 85%. The proposed method can improve the detection rate and reduce the false alarm rate compared with the traditional spatial-based method.

Detection of skin defects on loquat using hyperspectral imaging combining both band radio and improved three-phase level set segmentation method

AbstractBackground and objectivesSkin defects are one of the primary problems that occur in post-harvest grading and processing of loquats. Skin defects lead to the loquat being easily destroyed during transportation and storage, which causes the risk of other loquats being infected, affecting the selling price.Materials and MethodsIn this paper, a method combining band radio image with an improved three-phase level set segmentation algorithm (ITPLSSM) is proposed to achieve high accuracy, rapid, and non-destructive detection of skin defects of loquats. Principal component analysis (PCA) was used to find the characteristic wavelength and PC images to distinguish four types of skin defects. The best band ratio image based on characteristic wavelength was determined.ResultsThe band ratio image (Q782/944) based on PC2 image is the best segmented image. Based on pseudo-color image enhancement, morphological processing, and local clustering criteria, the band ratio image (Q782/944) has better contrast between defective and normal areas in loquat. Finally, the ITPLSSM was used to segment the processing band ratio image (Q782/944), with an accuracy of 95.28%.ConclusionsThe proposed ITPLSSM method is effective in distinguishing four types of skin defects. Meanwhile, it also effectively segments images with intensity inhomogeneities.

Intracranial Artery Morphology in Pediatric Moya Moya Disease and Moya Moya Syndrome.

Moya Moya disease (MMD) and Moya Moya syndrome (MMS) are cerebrovascular disorders, which affect the internal carotid arteries (ICAs). Diagnosis and surveillance of MMD/MMS in children mostly rely on qualitative evaluation of vascular imaging, especially MR angiography (MRA). To quantitatively characterize arterial differences in pediatric patients with MMD/MMS compared with normal controls. MRA data sets from 17 presurgery MMD/MMS (10M/7F, mean age = 10.0 years) patients were retrospectively collected and compared with MRA data sets of 98 children with normal vessel morphology (49 male patients; mean age = 10.6 years). Using a level set segmentation method with anisotropic energy weights, the cerebral arteries were automatically extracted and used to compute the radius of the ICA, middle cerebral artery (MCA), anterior cerebral artery (ACA), posterior cerebral artery (PCA), and basilar artery (BA). Moreover, the density and the average radius of all arteries in the MCA, ACA, and PCA flow territories were quantified. Statistical analysis revealed significant differences comparing children with MMD/MMS and those with normal vasculature ( P < .001), whereas post hoc analyses identified significantly smaller radii of the ICA, MCA-M1, MCA-M2, and ACA ( P < .001) in the MMD/MMS group. No significant differences were found for the radii of the PCA and BA or any artery density and average artery radius measurement in the flow territories ( P > .05). His study describes the results of an automatic approach for quantitative characterization of the cerebrovascular system in patients with MMD/MMS with promising preliminary results for quantitative surveillance in pediatric MMD/MMS management.

Offshore Bridge Detection in Polarimetric SAR Images Based on Water Network Construction Using Markov Tree

It is difficult to detect bridges in synthetic aperture radar (SAR) images due to the inherent speckle noise of SAR images, the interference generated by strong coastal scatterers, and the diversity of bridge and coastal terrain morphologies. In this paper, we present a two-step bridge detection method for polarimetric SAR imagery, in which the probability graph model of a Markov tree is used to build the water network, and bridges are detected by traversing the graph of the water network to determine all adjacent water branch pairs. In the step of the water network construction, candidate water branches are first extracted by using a region-based level set segmentation method. The water network is then built globally as a tree by connecting the extracted water branches based on the probabilistic graph model of a Markov tree, in which a node denotes a single branch and an edge denotes the connection of two adjacent branches. In the step of the bridge detection, all adjacent water branch pairs related to bridges are searched by traversing the constructed tree. Each bridge is finally detected by merging the two contours of the corresponding branch pair. Three polarimetric SAR data acquired by RADARSAT-2 covering Singapore and Lingshui, China, and by TerraSAR-X covering Singapore, are used for testing. The experimental results show that the detection rate, the false alarm rate, and the intersection over union (IoU) between the recognized bridge body and the ground truth are all improved by using the proposed method, compared to the method that constructs a water network based on water branches merging by contour distance.

Offshore Oil Platform Detection in Polarimetric SAR Images Using Level Set Segmentation of Limited Initial Region and Convolutional Neural Network

Offshore oil platforms are difficult to detect due to the complex sea state, the sparseness of target distribution, and the similarity of targets with ships. In this paper, we propose an oil platform detection method in polarimetric synthetic aperture radar (PolSAR) images using level set segmentation of a limited initial region and a convolutional neural network (CNN). Firstly, to reduce the interference of sea clutter, the offshore strong scattering targets were initially detected by the generalized optimization of polarimetric contrast enhancement (GOPCE) detector. Secondly, to accurately locate the contour of targets and eliminate false alarms, the coarse results were refined using an improved level set segmentation method. An algorithm for splitting and merging the smallest enclosing circle (SMSEC) was proposed to cover the coarse results and obtain the initial level set function. Finally, the LeNet-5 CNN model was used to classify the oil platforms and ships. Experimental results using multiple sets of polarimetric SAR data acquired by RADARSAT-2 show that the performance of the proposed method, including the detection rate, the false alarm rate, and the Intersection over Union (IOU) index between the extracted ROI and the ground truth, is better than the performance of a method that combines a GOPCE detector and a support vector machine classifier.

Detection and Segmentation of Lungs Regions Using CNN Combined with Levelset

Lung diseases are among the leaders in ranking diseases that kill the most globally. A quick and accurate diagnosis made by a specialist doctor facilitates the treatment of the disease and can save lives. In recent decades, an area that has gained strength in computing has been the aid to medical diagnosis. Several techniques were created to help health professionals in their work using Computer Vision Techniques and Machine Learning. This work presents a method of lung segmentation based on deep learning and computer vision techniques to aid in the medical diagnosis of lung diseases. The method uses the Detectron2 convolutional neural network for detection, which obtained 99.89% accuracy for detecting the pulmonary region. It was then combined with the LevelSet method for segmentation, which got 99.32% accuracy in segmentation in Lung Computed Tomography images being equivalent in state of the art, surpassing different deep learning models for segmentation.

Differentiation of Mild Cognitive Impairment Conditions in MR Images using Fractional order Jacobi Fourier Moment Features

Abstract Mild Cognitive Impairment (MCI) is the asymptomatic, preclinical transitional stage among aging and Alzheimer’s Disease (AD). Detection of MCI can ensure the timely intervention required to manage the disease’s severity. Morphological alterations of Lateral Ventricle (LV) is considered as a significant biomarker for disease diagnosis. This research aims to analyze the shape alterations of the LV region using Fractional Order Jacobi Fourier Moment (FOJFM) features, which are categorized by their generic nature and capabilities to perform time-frequency analysis. T1-weighted transaxial view brain MR images (HC = 92 and MCI = 63) are obtained from publicly available Open Access Series of Imaging Studies (OASIS) database. The LV region is delineated using Weighted Level Set (WLS) segmentation method and results are compared to Ground Truth (GT) images. FOJFM features are employed to characterize the morphometry of LV region. From this segmented region, 200 features are computed by varying the value of order and fractional parameters. Random Forest (RF) and Support Vector Machine (SVM) classifiers are used to differentiate Healthy Control (HC) and MCI subjects. Results show that WLSE is able to delineate the LV structure. The segmented region shows good correlation with the GT area. FOJFM features are observed to be statistically significant in discriminating HC and MCI subjects with p&lt;0.05. For MCI subjects, the feature values show higher variation as compared with HC brain, which might be due to the surface expansion of ventricular area during disease progression. SVM and RF classifiers show high performance F-measure values of 93.14% and 86.24%, respectively, for differentiating MCI conditions. The proposed moment based FOJFM features are able to capture the morphological changes of LV region related to MCI condition. Hence the proposed pipeline of work can be useful for the automated and early diagnosis of diseased conditions

Image segmentation for neuroscience: lymphatics

A recent discovery in neuroscience prompts the need for innovation in image analysis. Neuroscientists have discovered the existence of meningeal lymphatic vessels in the brain and have shown their importance in preventing cognitive decline in mouse models of Alzheimer’s disease. With age, lymphatic vessels narrow and poorly drain cerebrospinal fluid, leading to plaque accumulation, a marker for Alzheimer’s disease. The detection of vessel boundaries and width are performed by hand in current practice and thereby suffer from high error rates and potential observer bias. The existing vessel segmentation methods are dependent on user-defined initialization, which is time-consuming and difficult to achieve in practice due to high amounts of background clutter and noise. This work proposes a level set segmentation method featuring hierarchical matting, LyMPhi, to predetermine foreground and background regions. The level set force field is modulated by the foreground information computed by matting, while also constraining the segmentation contour to be smooth. Segmentation output from this method has a higher overall Dice coefficient and boundary F1-score compared to that of competing algorithms. The algorithms are tested on real and synthetic data generated by our novel shape deformation based approach. LyMPhi is also shown to be more stable under different initial conditions as compared to existing level set segmentation methods. Finally, statistical analysis on manual segmentation is performed to prove the variation and disagreement between three annotators.

Deep learning based fully automatic segmentation of the left ventricular endocardium and epicardium from cardiac cine MRI.

The segmentation of cardiac medical images is a crucial step for calculating clinical indices such as wall thickness, ventricular volume, and ejection fraction. In this study, we introduce a method named LsUnet that combines multi-channel, fully convolutional neural network, and annular shape level-set methods for efficiently segmenting cardiac cine magnetic resonance (MR) images. In this method, the multi-channel deep learning algorithm is applied to train the segmentation task to extract the left ventricle (LV) endocardial and epicardial contours. Next, the segmentation contours from the multi-channel deep learning method are incorporated into a level-set formulation, which is dedicated explicitly to detecting annular shapes to assure the segmentation's accuracy and robustness. The proposed automatic approach was evaluated on 95 volumes (total 1,076 slices, ~80% as for training datasets, ~20% 2D as for testing datasets). This combined multi-channel deep learning and annular shape level-set segmentation method achieved high accuracy with average Dice values reaching 92.15% and 95.42% for LV endocardium and epicardium delineation, respectively, in comparison to the reference standard (the manual segmentation). A novel method for fully automatic segmentation of the LV endocardium and epicardium from different MRI datasets is presented. The proposed workflow is accurate and robust compared to the reference and other state-of-the-art methods.

An Automatically Initialized Level Set Method for Ochotona Curzoniae Image Segmentation

In order to effectively solve this problem that that the traditional level set method needs to mark the initial contour manually, a novel method that automatic initialization by utilizing deep learning is proposed in this paper. Firstly, a target detection network based on Faster-RCNN is used for target detection of ochotona curzoniae. Secondly, the detected target boxes are regarded as the initial contour of level set models, and ochotona curzoniae images are segmented by level set models, by which the problem of manually plotting the initial contour required by traditional level set models is solved. The experimental results of ochotona curzoniae image segmentation show that there is little difference between the level set segmentation method in which the initial contour of the target is automatically generated by deep learning(LSECA) is used to automatically target initial contour and the level set segmentation method where initial contour needs to be manually plotted. Therefore, the experimental results prove that it is feasible to automatically obtain the target initial contour through deep learning.

Automatic Recognition of Parathyroid Nodules in Ultrasound Images Based on Fused Prior Pathological Knowledge Features

Automation diagnosis of parathyroid nodules is of crucial importance to recognize parathyroid nodules in ultrasound images. Aiming at the different nodule shapes of diverse patients, blurred boundaries, complex backgrounds and inhomogeneous intensity of ultrasound images, we propose a novel hybrid level set model to accurately segment nodules. The adaptive global term weight is determined based on the image local entropy of the region around the evolution contour and two scales are proposed for the local term to drive the evolution contour fast approaching to the boundary in order to avoid large amount of calculation and over-segmentation. We also propose membrane features and relative position features based on prior pathological knowledge to describe the inherent characteristics of parathyroid nodules different from thyroid and other nodules. We fused prior pathological knowledge features, morphology features and texture features of the segmented nodules to recognize parathyroid nodules by the support vector data description(SVDD). The experiment result indicates that the incorporation of the proposed hybrid level set segmentation method and the fused prior pathological knowledge features, morphology features and texture features improve the recognition accuracy and efficiency of parathyroid nodules, which is much higher than that only with morphology and texture features.

3D-MRI Brain Tumor Detection Model Using Modified Version of Level Set Segmentation Based on Dragonfly Algorithm

Accurate brain tumor segmentation from 3D Magnetic Resonance Imaging (3D-MRI) is an important method for obtaining information required for diagnosis and disease therapy planning. Variation in the brain tumor’s size, structure, and form is one of the main challenges in tumor segmentation, and selecting the initial contour plays a significant role in reducing the segmentation error and the number of iterations in the level set method. To overcome this issue, this paper suggests a two-step dragonfly algorithm (DA) clustering technique to extract initial contour points accurately. The brain is extracted from the head in the preprocessing step, then tumor edges are extracted using the two-step DA, and these extracted edges are used as an initial contour for the MRI sequence. Lastly, the tumor region is extracted from all volume slices using a level set segmentation method. The results of applying the proposed technique on 3D-MRI images from the multimodal brain tumor segmentation challenge (BRATS) 2017 dataset show that the proposed method for brain tumor segmentation is comparable to the state-of-the-art methods.

Segmentation of spinal cord from computed tomography images based on level set method with Gaussian kernel

In the human body, organs segmentation is the most imperative issues in therapeutic applications. The challenges are connected with medicinal image segmentation and low complexity between required organ and incorporating tissues. There exist a wide range of methodologies for how a segmentation problem can be comprehended. These methods want to have a spot specific region of individual bones. The particular part remains a test for spinal cord segmentation. As a result of the beforehand expressed downsides of the current spinal cord segmentation procedures, this paper proposes a modified spatial fuzzy C clustering with level set segmentation method to incorporate Neumann Boundary Condition, a third function, called by the level set evolution. Neumann Boundary Condition is utilized to specify the normal derivative of the function present on any surface. The proposed method gives better results of segmentation of the spinal cord organs. The execution of the proposed method proves its superiority in term of accuracy as compared with the other methods.

Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Breast cancer is the most common invasive cancer with the highest cancer occurrence in females. Handheld ultrasound is one of the most efficient ways to identify and diagnose the breast cancer. The area and the shape information of a lesion is very helpful for clinicians to make diagnostic decisions. In this study we propose a new deep-learning scheme, semi-pixel-wise cycle generative adversarial net (SPCGAN) for segmenting the lesion in 2D ultrasound. The method takes the advantage of a fully convolutional neural network (FCN) and a generative adversarial net to segment a lesion by using prior knowledge. We compared the proposed method to a fully connected neural network and the level set segmentation method on a test dataset consisting of 32 malignant lesions and 109 benign lesions. Our proposed method achieved a Dice similarity coefficient (DSC) of 0.92 while FCN and the level set achieved 0.90 and 0.79 respectively. Particularly, for malignant lesions, our method increases the DSC (0.90) of the fully connected neural network to 0.93 significantly (p 0.001). The results show that our SPCGAN can obtain robust segmentation results. The framework of SPCGAN is particularly effective when sufficient training samples are not available compared to FCN. Our proposed method may be used to relieve the radiologists' burden for annotation.

Optimized Level Set Method for Segmentation of SAR Images using Adaptive Fuzzy K Means Clustering

The complications that occurred in remote sensing image information and analysis algorithms growth of a large scale image segmentation haven’t kept a place with the requirement for the methods which to develop the final accuracy of object detection as well as the recognition. Traditional Level set segmentation methods which are Chan-Vese (CV), Image and Vision Computing (IVC) 2010, ACM with SBGFRLS, and Online Region-Based ACM (ORACM) are suffered from more amounts of time complexity, as well as low segmentation accuracy due to large intensity homogeneities and the noise at which the region based segmentation is impossible. So this is the reason, we proposed a navel hybrid methodology called adaptive particle swarm optimization (PSO) based Fuzzy K-Means clustering algorithm. The proposed approach is diversified into two stages; in stage one, pre-processing the input image to improve the clustering efficiency and overcome the obstacles present in traditional methods by using particle swarm optimization (PSO) and Adaptive Fuzzy K-means clustering algorithm. With the help of the PSO algorithm, we getithe “optimum” pixels values are extracted from the input SAR images, these optimum values are automatically acted as clusters centers for Adaptive Fuzzy K-Means Clustering instead of random initialization from the original image. The pre-processing segmentation result improved the clustering efficiency but suffers from few drawbacks such as boundary leakages and outliers even particle Swarm optimization is used. To overcome the above drawbacks post-processing is needed to facilitate the superior segmentation results with the help of the level set method. It utilizes an efficient curve deformation driven by external and internal forces to capture the important structures (usual iedges) in an image. The combined approach of both pre-processing and post-processingwhich is called Particle Swarm Optimization based Adaptive Fuzzy-K-Means (AFKM) clustering via the level set method. The proposed approach is successfully implemented on large

The application of gray-scale level-set method in segmentation of concrete deck delamination using infrared images

Conventional nondestructive delamination detection of concrete pavements through thermography is often based on temperature contrasts between delaminated and sound areas. Non-uniform backgrounds caused by the environmental conditions are often challenging for contrast-based methods to robustly differentiate the delaminated areas from the sound areas. Instead of focusing on the temperature contrast, this study proposes a temperature gradient-based level set method (LSM) to detect boundaries for delamination segmentations. A modified edge indicator function is developed to represent the normalized temperature gradient of a thermal image. The experimental study was conducted to evaluate its applicability and stability for boundary detection in terms of different burial depths and observation time windows. Then the LSM generated edges LSM were used for delamination segmentations. The results demonstrated significant improvements compared to the well-accepted k-mean method in terms of accuracy and stability of the detections. A field implementation of this method using an in-service bridge generated satisfactory outcomes when compared to the ground truth data. Major limitations and future works are also discussed in terms of parameter selections.

An automatic level set method for hippocampus segmentation in MR images

ABSTRACT Hippocampus segmentation in MR images is beneficial for the diagnosis of many diseases and pathologies such as Alzheimer’s disease. Manual segmentation of the hippocampus is highly time-consuming and has low reproducibility; however, automated methods have introduced substantial gains in this regard. In this study, we used a novel level-set method for hippocampus segmentation in combination with the SBGFRLS (Selective Binary and Gaussian Filtering Regularised Level Set) and LAC (Localising Region-Based Active Contours) algorithms. The proposed method avoided the algorithms which required a large database and instead used a more complex level set approach to obtain comparable accuracy. This method was applied to a set of 36 MRI scans provided by the Alzheimer’s Disease Neuroimaging Initiative (ADNI), using the Harmonised Hippocampal Protocol (HarP) as the gold standard. In addition, the results were compared with the outputs of the Freesurfer software package. In regards to the similarity indices, the results of our algorithm (mean Dice = 0.847) were more comparable with the gold standard compared to those of Freesurfer. Classification results for AD vs control and MCI vs control showed a high degree of accuracy (91% and 75%, respectively). Therefore, this method can be an option for accurate and robust segmentation of the hippocampus.