Streaming level set algorithm for 3D segmentation of confocal microscopy images

Distance mapping possessing computational advantage largely decides the effectiveness of level sets for image segmentation. This article presents the impact of several distance mapping and level set methods suggested in the literature and provides an effective way of handling it. Different distance metric schemes such as the Euclidean, city-block and chessboard distances have been very much prevalent in the literature. This article highlights the use of an effective, fast and efficient distance-mapping technique, i.e. distance mapping using scanning and filling technique, proposed by the authors in their earlier work. Further, this article emphasises the need of periodic reinitialisation of the level set function to a signed distance function which makes curvature term become redundant. As the curve evolves, the level set function loses its signed distance property, leading to reduction in the evolution speed. Frequent reinitialisation of the level set to signed distance function overcomes this limitation and increases the speed of evolution.

In this paper, we present a robust and accurate algorithm for interactive image segmentation. The level set method is clearly advantageous for image objects with a complex topology and fragmented appearance. Our method integrates discriminative classification models and distance transforms with the level set method to avoid local minima and better snap to true object boundaries. The level set function approximates a transformed version of pixelwise posterior probabilities of being part of a target object. The evolution of its zero level set is driven by three force terms, region force, edge field force, and curvature force. These forces are based on a probabilistic classifier and an unsigned distance transform of salient edges. We further propose a technique that improves the performance of both the probabilistic classifier and the level set method over multiple passes. It makes the final object segmentation less sensitive to user interactions. Experiments and comparisons demonstrate the effectiveness of our method.

Paper presents the concept of applying image segmentation algorithms for precise extraction of cerebrospinal fluid (CSF) from CT brain scans. Accurate segmentation of the CSF from the intracranial brain area is crucial for further reliable analysis and quantitative assessment of hydrocephalus. Presented research was aimed at the comparison of effectiveness of three modern segmentation approaches used for this purpose. Specifically, random walk, level set and min-cut/max-flow algorithms were considered. The visual and numerical comparison of the segmentation results leads to conclusion that the most effective algorithm for the considered problem is level set, although the positive medical verification of the results revealed that either of considered algorithms can be successfully applied in the diagnostic applications.

The morphological analysis of dendritic spines is an important challenge for the neuroscientific community. Most state-of-the-art techniques rely on user-supervised algorithms to segment the spine surface, especially those designed for light microscopy images. Therefore, processing large dendritic branches is costly and time-consuming. Although deep learning (DL) models have become one of the most commonly used tools in image segmentation, they have not yet been successfully applied to this problem. In this article, we study the feasibility of using DL models to automatize spine segmentation from confocal microscopy images. Supervised learning is the most frequently used method for training DL models. This approach requires large data sets of high-quality segmented images (ground truth). As mentioned above, the segmentation of microscopy images is time-consuming and, therefore, in most cases, neuroanatomists only reconstruct relevant branches of the stack. Additionally, some parts of the dendritic shaft and spines are not segmented due to dyeing problems. In the context of this research, we tested the most successful architectures in the DL biomedical segmentation field. To build the ground truth, we used a large and high-quality data set, according to standards in the field. Nevertheless, this data set is not sufficient to train convolutional neural networks for accurate reconstructions. Therefore, we implemented an automatic preprocessing step and several training strategies to deal with the problems mentioned above. As shown by our results, our system produces a high-quality segmentation in most cases. Finally, we integrated several postprocessing user-supervised algorithms in a graphical user interface application to correct any possible artifacts.

This paper proposes a hybrid boundary detection method for image based on a new modified level set method and a fuzzy model. It is applied to a boundary detection problem of coronary plaque. Level set method has been applied widely in image processing. It however does not work well for an intravascular ultrasound (IVUS) image because an image gradient, commonly used for calculating a speed function in the level set method, cannot detect an image boundary well. The level set method and the weighted image separability proposed by the authors in the past were applied for a coronary plaque boundary detection problem. The level set method could not however detect the plaque boundary in several regions. One problem was that the candidates of the plaque boundary detected by the weighted separability were unclear in several regions. The other problem was that the IVUS image often becomes shadowed and it contains no texture information there, which is caused by the presence of the guide wire. To overcome this problem, we propose a new modified level set, and we further propose a hybrid boundary detection method based on the new modified level set and the Takagi Sugeno (T-S) fuzzy model for detecting a coronary plaque boundary. The boundary detection accuracy of the proposed method was significantly better than those of the previous methods we proposed in the past.

Segmenting brain tumors in 3D magnetic resonance imaging (3D-MRI) accurately is critical for easing the diagnostic and treatment processes. In the field of energy functional theory-based methods for image segmentation and analysis, level set methods have emerged as a potent computational approach that has greatly aided in the advancement of the geometric active contour model. An important factor in reducing segmentation error and the number of required iterations when using the level set technique is the choice of the initial contour points, both of which are important when dealing with the wide range of sizes, shapes, and structures that brain tumors may take. To define the velocity function, conventional methods simply use the image gradient, edge strength, and region intensity. This article suggests a clustering method influenced by the Quantum Inspired Dragonfly Algorithm (QDA), a metaheuristic optimizer inspired by the swarming behaviors of dragonflies, to accurately extract initial contour points. The proposed model employs a quantum-inspired computing paradigm to stabilize the trade-off between exploitation and exploration, thereby compensating for any shortcomings of the conventional DA-based clustering method, such as slow convergence or falling into a local optimum. To begin, the quantum rotation gate concept can be used to relocate a colony of agents to a location where they can better achieve the optimum value. The main technique is then given a robust local search capacity by adopting a mutation procedure to enhance the swarm's mutation and realize its variety. After a preliminary phase in which the cranium is disembodied from the brain, tumor contours (edges) are determined with the help of QDA. An initial contour for the MRI series will be derived from these extracted edges. The final step is to use a level set segmentation technique to isolate the tumor area across all volume segments. When applied to 3D-MRI images from the BraTS' 2019 dataset, the proposed technique outperformed state-of-the-art approaches to brain tumor segmentation, as shown by the obtained results.

This paper addresses the problem of intensity correction of fluorescent confocal laser scanning microscope images. Confocal laser scanning microscope images are frequently used in medicine for obtaining 3D information about specimen structures by imaging a set of 2D cross sections and performing 3D volume reconstruction afterwards. However, 2D images acquired from fluorescent confocal laser scanning microscope images demonstrate significant intensity heterogeneity, for example, due to photo-bleaching and fluorescent attenuation in depth. We developed an intensity heterogeneity correction technique that (a) adjusts the intensity heterogeneity of 2D images, (b) preserves fine structural details and (c) enhances image contrast, by performing spatially adaptive mean-weight filtering. Our solution is obtained by formulating an optimization problem, followed by filter design and automated selection of filtering parameters. The proposed filtering method is experimentally compared with several existing techniques by using four quality metrics: contrast, intensity heterogeneity (entropy) in a low frequency domain, intensity distortion in a high frequency domain and saturation. Based on our experiments and the four quality metrics, the developed mean-weight filtering outperforms other intensity correction methods by at least a factor of 1.5 when applied to fluorescent confocal laser scanning microscope images.

To investigate the cytologic findings of superior bulbar conjunctiva in superior limbic keratoconjunctivitis (SLK) using laser scanning confocal microscopy and impression cytology in a prospective controlled study. Twenty-one eyes of 11 SLK patients (9 women, 2 men; mean age, 49.3 +/- 17.9 years) and 18 eyes of 9 control subjects (6 women, 3 men; mean age, 46.4 +/- 8.7 years) underwent tear function tests including vital stainings, Schirmer test, tear clearance test, digital confocal laser scanning microscopy, and conjunctival impression cytology. After confocal microscopy and impression cytology images were obtained, the mean individual epithelial cell area (MIECA), nucleocytoplasmic (N/C) ratio, and inflammatory cell density were analyzed. The correlation between confocal microscopy and impression cytology parameters was investigated. The MIECA of SLK patients and control subjects in confocal microscopy was 786.54 +/- 463.88 microm(2) and 311.50 +/- 78.30 microm(2), respectively. The mean N/C ratio was 0.356 +/- 0.090 and 0.490 +/- 0.038, respectively. The MIECA and N/C ratio in impression cytology showed significant correlation with the corresponding confocal microscopy parameters (MIECA, P = 0.0028; N/C, P = 0.0051). The inflammatory cell density in confocal microscopy significantly correlated with superior bulbar conjunctival Rose-Bengal scores (P = 0.0264). Laser scanning confocal microscopy seems to be an efficient noninvasive tool in the evaluation of phenotypic alterations of the conjunctival epithelium in SLK and may serve as an alternative for impression cytology. N/C ratio and inflammatory cell density appear to be two new promising parameters of in vivo confocal microscopy in the assessment of ocular surface disease in SLK.

To overcome over-segmentation of Watershed transform, a novel improved Watershed algorithm based on adaptive marker-extraction is proposed. The original marker-based Watershed algorithm is improved by considering multiple feature information of local minima and adaptively selecting threshold. The proposed method consists of five steps: 1) Calculating gradient directly with color vectors; 2) Low-pass filtering of gradient image with BTPF; 3) Employing Hminima transform to extract true local minima whose depth is lower than that of threshold H, which is adaptively adjusted according to gradient image's statistical character. 4) Further marker-extraction being based on water basin scale. 5) Imposing the markers on the original gradient image as its minima; finally, Watershed transform is implied to the marked gradient image to segment the image. Experimental results show that, compared with other testing Watershed algorithms, the proposed method can more efficiently reduce over-segmentation and obtain better segmentation performance with lower computational complexity; in addition, it has better anti-noise performance and edge-location capability as well.

The variational level set model for piecewise constant multiphase image segmentation on the plane and the related Split-Augmented-Lagrangian Projection Method (SALPM) are investigated in this paper. On the analysis of the current problems based on the variational level set method for image segmentation, we also design a rapid SALPM method for two-phase image segmentation model, getting the general model of multiple phase level set in order to facilitate the generic design and program. In addition, the concrete formula of the rapid split algorithms for multiphase image segmentation with level set model and the calculation steps are given, and simultaneously taking liver tumor CT image as examples of the multiphase segmentation. Moreover, by comparing with the traditional methods, the experiments show that our algorithm presented in this paper have higher computational efficiency and accuracy, and better the extraction of liver contour.

Image segmentation is one of the most fundamental and important areas in the field of image processing and computer vision. The traditional level set methods need initialize the level set function as a distance function. If the initial contour is selected inappropriate, we may not get the desired ideal segmentation result. In order to solve the problem of level set automation initial, we proposed a new image segmentation algorithm based on level set and marker extraction. First, we extract the internal mark as level set initial curve by using Extended-minima transform. And then, through using the local binary fitting active contour model, we evolve the labeled image to get the final segmentation result. The simulation results show that this method has low computing complexity than the traditional level set method requirements, and can effectively solve the initialization problem of level set.

A difference in refractive index (n) between immersion medium and specimen results in increasing loss of intensity and resolution with increasing focal depth and in an incorrect axial scaling in images of a confocal microscope. Axial thickness measurements of an object on such images are therefore not exact. The present paper describes a simple procedure to determine the correct axial thickness of an object with confocal fluorescence microscopy. We study this procedure for a specimen that has a higher refractive index than the immersion medium and with a thickness up to 100 microm. The measuring method was experimentally tested by comparing the thickness of polymer layers measured on axial images of a confocal microscope in case of a water-polymer mismatch to reference values obtained from an independent technique, i.e. scanning electron microscopy. The case when the specimen has a lower refractive index than the immersion medium is also shown by way of illustration. Measured thickness data of a water layer and an oil layer with the same actual thickness were obtained using an oil-immersion objective lens with confocal microscopy. Good agreement between theory and experiment was found in both cases, consolidating our method.

Wire Arc Additive Manufacturing allows the cost-effective manufacturing of customized, large-scale metal parts. As the post-process quality assurance of large parts is costly and time-consuming, process monitoring is inevitable. In the present study, a context-aware monitoring solution was investigated by integrating machine, temporal, and spatial context in the data analysis. By analyzing the voltage patterns of each cycle in the oscillating cold metal transfer process with a deep neural network, temporal context was included. Spatial context awareness was enabled by building a digital twin of the manufactured part using an Octree as spatial indexing data structure. By means of the spatial context awareness, two quality metrics—the defect expansion and the local anomaly density—were introduced. The defect expansion was tracked in-process by assigning detected defects to the same defect cluster in case of spatial correlation. The local anomaly density was derived by defining a spherical region of interest which enabled the detection of aggregations of anomalies. By means of the context aware monitoring system, defects were detected in-process with a higher sensitivity as common defect detectors for welding applications, showing less false-positives and false-negatives. A quantitative evaluation of defect expansion and densities of various defect types such as pore nests was enabled.

We have been developing a computer-aided diagnosis (CAD) system for automatically recognizing cervical cancer cells from Papanicolaou smear. Considering that pathological changes of cervix can be indicated by the abnormity of the nucleus of intermediate cell, the key task of this system is to find the intermediate cells and segment the nucleus precisely. This paper presents a novel approach for automatic segmentation of microscopic cervical cell images using multispectral imaging techniques. In order to capture images at different wavelengths, a Liquid Crystal Tunable Filter (LCTF) device is used to provide wavelength selection from 400nm to 720nm with an increment of 10nm. Considering the spectral variances of background, nucleus and cytoplasm, background is extracted firstly from the microscopic images by calculating pixel intensity variance at 470nm, 530nm, 570nm, 580nm and 650nm. Then superficial cells are extracted apart from intermediate cells easily at 530nm 650nm because of the different pixel intensity distribution of the two kinds of cells at these two wavelengths. To segment the nucleus from intermediate cells, we adopt two procedures. Firstly, the nuclei are roughly segmented apart by using an iterative maximum deviation between-cluster algorithm. Secondly, a novel rigorous algorithm based on active contour model is adopted to achieve more exact nuclei segmentation. Using the method proposed in this paper, we did experiments on over 300 cervical smears, and the results show that this method is more robust and precise.

For the foreground and background segmentation in Three-Dimensional(3D) video inpainting, the inaccuracy of the foreground extraction will easily affect the quality of the repaired image. In order to solve this problem, this paper combines watershed algorithm with marker for foreground extraction, making full use of the structural information of the depth image. In order to enhance the ability of distinguishing the foreground object, introducing K-means clustering to mark it in the gradient image. The experimental results show that the improved algorithm overcomes the defect of original watershed algorithm in image segmentation, which is likely to occur the over-segmentation phenomenon, and it also can completely extract the texture information of the foreground object, making the repaired image have better visual effect, and the peak signal-to-noise ratio(PSNR) is increased by 1 to 3 dB, compared to the other algorithm.