A Novel Approach for 3D Renal Segmentation Using a Modified GAN Model and Texture Analysis

In this thesis a method for segmenting textured images using Gabor filters is presented. One of the most recent approaches for texture segmentation and analysis is multi-channel filtering. There are several applicable choices as filter banks which are used for textured images. Gaussian filters modulated by exponential or by sinusoidal filters, known as Gabor filters, have been proven to be very usefyl for texture analysis for the images containing specific frequency and orientation characteristics. Resembling the human visual cortical cells, Gabor function is a popular sub-band filter for multi-channel decompositon. Optimum joint spatial/spatial frequency uncertainty principle and its ability to recognize and pass specific frequencies and orientations are attributes of Gabor filter that make it more attractive. Gabor function with these attributes could simulate the task of simple visual cells in the cortex. Gabor function has several parameters that determine the sub-band Gabor filter and must be determined accurately to extract the features precisely for texture discrimination. A wide selection range for each parameter exists and many combinations of these parameters are possible. Accurate selection and combination of values for the parameters are of crucial importance. Hence a difficult goal is minimizing the number of filters. On the other hand a variety of approaches of texture analysis and recognition have been presented in remote sensing applications, including land cover/land use classification and urban scene segmentation. With the avaiability of very high-resolution commercial satellite imagery such as IKONOS, it is possible to obtain detailed information on urban land use and change detection that are of particular interest to urban and regional planners. In this thesis considering the attributes of human visual system, a hybrid algorithm is implemented using multi-channel decomposition by Gabor filter bank for feature extraction in conjunction with Artificial Neural Networks for both feature reduction and texture segmentation. Three approaches are implemented to optimize Gabor filter bank for image segmentation. Eventually the proposed method is successfully applied for segmentation of IKONOS satellite images.

In this thesis a method for segmenting textured images using Gabor filters is presented. One of the most recent approaches for texture segmentation and analysis is multi-channel filtering. There are several applicable choices as filter banks which are used for textured images. Gaussian filters modulated by exponential or by sinusoidal filters, known as Gabor filters, have been proven to be very usefyl for texture analysis for the images containing specific frequency and orientation characteristics. Resembling the human visual cortical cells, Gabor function is a popular sub-band filter for multi-channel decompositon. Optimum joint spatial/spatial frequency uncertainty principle and its ability to recognize and pass specific frequencies and orientations are attributes of Gabor filter that make it more attractive. Gabor function with these attributes could simulate the task of simple visual cells in the cortex. Gabor function has several parameters that determine the sub-band Gabor filter and must be determined accurately to extract the features precisely for texture discrimination. A wide selection range for each parameter exists and many combinations of these parameters are possible. Accurate selection and combination of values for the parameters are of crucial importance. Hence a difficult goal is minimizing the number of filters. On the other hand a variety of approaches of texture analysis and recognition have been presented in remote sensing applications, including land cover/land use classification and urban scene segmentation. With the avaiability of very high-resolution commercial satellite imagery such as IKONOS, it is possible to obtain detailed information on urban land use and change detection that are of particular interest to urban and regional planners. In this thesis considering the attributes of human visual system, a hybrid algorithm is implemented using multi-channel decomposition by Gabor filter bank for feature extraction in conjunction with Artificial Neural Networks for both feature reduction and texture segmentation. Three approaches are implemented to optimize Gabor filter bank for image segmentation. Eventually the proposed method is successfully applied for segmentation of IKONOS satellite images.

Objective: To explore the value of contrast-enhanced CT combined with texture analysis in differentiating pancreatic cancer from mass-forming pancreatitis in pancreatic head. Methods: A retrospective study collected 21 patients with pancreatic head mass-forming pancreatitis confirmed by surgery or biopsy and 47 patients with pancreatic ductal adenocarcinoma confirmed by surgery. The patients visited the Affiliated Hospital of Nanjing University of Chinese Medicine and the First Affiliated Hospital of Wannan Medical College between January 2014 and December 2017. Gender, age and CT findings were collected. The parenchymal phase was selected for texture analysis. The minimum absolute shrinkage and selection operator (LASSO) method was applied for dimensionality reduction.Two independent sample t-tests or Mann-Whitney U test were used for continuous variables based on the Shapiro-Wilks normality test results. Categorical variables were tested by Chi-square or Fisher test. By multivariable regression analysis, CT findings, CT texture analysis, CT findings combined with texture analysis prediction models were established. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic performance of individual indicators and each prediction model. The Delong test was used to compare the area under the curve (AUC) of each model. Results: The CT findings prediction model consisted of CT value of lesion on pancreatic parenchymal phase and pancreatic duct penetrating sign. The texture analysis prediction model consists of root mean square and low grey level run emphasis_angle135. The AUC of them were not statistically different (Z=0.150,P>0.05). The combined predictive model had the better diagnostic performance (AUC 0.944, sensitivity 83.0%, specificity 95.2%, +LR 17.43, -LR 0.18) than CT sign prediction model (Z=2.008, P<0.05) and texture analysis prediction model(Z=2.236, P<0.05) were significantly different. Conclusions: The CT findings model and the texture analysis model have equivalent diagnostic performance in the differentiation of mass-forming pancreatitis and pancreatic cancer. The enhanced CT combined with texture analysis model has the best diagnostic efficiency and can further improve the diagnostic ability.

Amyotrophic lateral sclerosis, a progressive neurodegenerative disease, presents challenges in predicting individual disease trajectories due to its heterogeneous nature. This study explores the application of texture analysis on T1-weighted MRI in patients with amyotrophic lateral sclerosis, stratified by the D50 disease progression model. The D50 model, which offers a more nuanced representation of disease progression than traditional linear metrics, calculates the sigmoidal curve of functional decline and provides independent quantifications of disease aggressiveness and accumulation. In this research, a representative cohort of 116 patients with amyotrophic lateral sclerosis was studied using the D50 model and texture analysis on MRI images. Texture analysis, a technique used for quantifying voxel intensity patterns in MRI images, was employed to discern alterations in brain tissue associated with amyotrophic lateral sclerosis. This study examined alterations of the texture feature autocorrelation across sub-groups of patients based on disease accumulation, aggressiveness and the first site of onset, as well as in direct regressions with accumulation/aggressiveness. The findings revealed distinct patterns of the texture-derived autocorrelation in grey and white matter, increase in bilateral corticospinal tract, right hippocampus and left temporal pole as well as widespread decrease within motor and extra-motor brain regions, of patients stratified based on their disease accumulation. Autocorrelation alterations in grey and white matter, in clusters within the left cingulate gyrus white matter, brainstem, left cerebellar tonsil grey matter and right inferior fronto-occipital fasciculus, were also negatively associated with disease accumulation in regression analysis. Otherwise, disease aggressiveness correlated with only two small clusters, within the right superior temporal gyrus and right posterior division of the cingulate gyrus white matter. The findings suggest that texture analysis could serve as a potential biomarker for disease stage in amyotrophic lateral sclerosis, with potential for quick assessment based on using T1-weighted images.

Image processing is a very rich and important research area, which provides efficient solutions to many real and industrial problems. Texture analysis is one of the most interesting fields in image processing and pattern recognition. It became a very attractive research area these last years, especially after the growth and the advancement of technologies. This paper deals with texture analysis and unsupervised texture segmentation problem. The goal of this study is to develop a new segmentation method based on the textural features of the images. The proposed system is composed of different steps. First, the image is analyzed in each pixel using the Gray Level Co-occurrence Matrix (GLCM) feature extraction method. Four Haralick parameters (Haralick Proc IEEE 67(5):786–804, 16) are calculated and represented in four matrix. After that, we applied the gradient to detect edges from the extracted images. In order to localize the area of the discontinuity of the texture, we proposed a new method for joining the edge and region growing. The proposed system is applied on several textured images and the obtained results are shown in the experimentation section. A number of experiments have been done with randomly generated textured images. The experiments have shown the efficiency of the proposed method compared to other existing methods and its robustness to enhance the segmentation precision of textured images.

Practical applications of neural networks in texture analysis

Computer-based texture analysis provides objective data that can be extracted from medical images, including ultrasound images. One popular methodology involves the generation of a gray-level co-occurrence matrix (GLCM) from the image, and from that matrix, texture fractures can be extracted. We performed texture analysis on 280 ultrasound testicular images obtained from 70 dogs and explored the resulting texture data, by means of principal component analysis (PCA). Various abnormal lesions were identified subjectively in 35 of the 280 cropped images. In 16 images, pinpoint-to-small, well-defined, hyperechoic foci were identified without acoustic shadowing. These latter images were classified as having "microliths." The remaining 19 images with other lesions and areas of non-homogeneous testicular parenchyma were classified as "other." In the PCA scores plot, most of the images with lesions were clustered. These clustered images represented by those scores had higher values for the texture features entropy, dissimilarity, and contrast, and lower values for the angular second moment and energy in the first principal component. Other data relating to the dogs, including age and history of treatment for prostatomegaly or chemical castration, did not show clustering on the PCA. This study illustrates that objective texture analysis in testicular ultrasound correlates to some of the visual features used in subjective interpretation and provides quantitative data for parameters that are highly subjective by human observer analysis. The study demonstrated a potential for texture analysis in prediction models in dogs with testicular abnormalities.

Magnetic resonance imaging findings often do not distinguish between people with and without low back pain (LBP). However, there are still a large number of people who undergo magnetic resonance imaging to help determine the etiology of their back pain. Texture analysis shows promise for the classification of tissues that look similar, and machine learning can minimize the number of comparisons. This study aimed to determine if texture features from lumbar spine magnetic resonance imaging differ between people with and without LBP. In total, 14 participants with chronic LBP were matched for age, weight, and gender with 14 healthy volunteers. A custom texture analysis software was used to construct a gray-levelco-occurrence matrix with one to four pixels offset in 0° direction for the discand superior and inferior endplate regions. The Random Forests Algorithm was used to select the most promising classifiers. The linear mixed-effect model analysis was used to compare groups (pain vs. pain-free) at each level controlling for age. The Random Forest Algorithm recommended focusing on intervertebral discs and endplate zones at L4-5 and L5-S1. Differences were observed between groups for L5-S1 superior endplate contrast, homogeneity, and energy (p = .02). Differences were observed for L5-S1 disc contrast and homogeneity (p < .01), as well as for the inferior endplates contrast, homogeneity, and energy (p < .03). Magnetic resonance imaging textural features may have potential in identifying structures that may be the target of further investigations about the reasons for LBP.

A segmentation algorithm for image content analysis is presented. We assume that the textures in a video scene can be labeled subjectively relevant or irrelevant. Relevant textures are defined as containing subjectively meaningful details, while irrelevant textures can be seen as image content with less important subjective details. We apply this idea to video coding using a texture analyzer and a texture synthesizer. The texture analyzer (encoder side) identifies the texture regions with unimportant subjective details and generates side information for the texture synthesizer (decoder side), which in turn inserts synthetic textures at the specified locations. The focus of this paper is the texture analyzer, which uses multiple MPEG-7 descriptors simultaneously for similarity estimation. The texture analyzer is based on a split and merge segmentation approach. Its current implementation yields an identification rate of up to 96% and an average gain of up to 10% compared to single descriptor usage.

Learning techniques have shown high efficacy rates when applied to similar clinical problems (Jabarouti et al. 2011; Mazurowski et al. 2008; Nassar et al. 2007; Said et al. 2006; Lisboa 2002). Recent studies over biological tissue classification or similar applications subjects achieve tissue recognition using different segmentation approaches such as mean shift clustering, region growing, watersheds, and histogram thresholds (Koutsouri et al. 2013; Anuradha et al. 2012; Kang et al. 2010; Perez et al. 2001; Stelt et al. 1985). Furthermore, texture analysis can be used to separate different significant regions in digital images of human tissues or organs (Veredas et al. 2010).

It is unclear whether changes in liver texture in patients with colorectal cancer are caused by diffuse (e.g., perfusional) changes throughout the liver or rather based on focal changes (e.g., presence of occult metastases). The aim of this study is to compare a whole-liver approach to a segmental (Couinaud) approach for measuring the CT texture at the time of primary staging in patients who later develop metachronous metastases and evaluate whether assessing CT texture on a segmental level is of added benefit. 46 Patients were included: 27 patients without metastases (follow-up>2years) and 19 patients who developed metachronous metastases within 24months after diagnosis. Volumes of interest covering the whole liver were drawn on primary staging portal-phase CT. In addition, each liver segment was delineated separately. Mean gray-level intensity, entropy (E), and uniformity (U) were derived with different filters (σ0.5-2.5). Patients/segments without metastases and patients/segments that later developed metachronous metastases were compared using independent samples t tests. Absolute differences in entropy and uniformity between the group without metastases and the group with metachronous metastases group were consistently smaller for the segmental approach compared to the whole-liver approach. No statistically significant differences were found in the texture measurements between both groups. In this small patient cohort, we could not demonstrate a clear predictive value to identify patients at risk of developing metachronous metastases within 2years. Segmental CT texture analysis of the liver probably has no additional benefit over whole-liver texture analysis.

Breast Cancer Treatment Response Monitoring Using Quantitative Ultrasound and Texture Analysis: Comparative Analysis of Analytical Models

Recent advances in medical image processing for the evaluation of chronic kidney disease

Texture analysis of multiple sclerosis: a comparative study

Clinical evidence suggests that the response to immune checkpoint blockade depends on the immune status in the tumor microenvironment. This study aims to predict the immunophenotyping (IP) and overall survival (OS) of intrahepatic cholangiocarcinoma (ICC) patients using preoperative magnetic resonance imaging (MRI) texture analysis. A total of 78 ICC patients were included and divided into inflamed (n = 26) or non-inflamed (n = 52) immunophenotyping based on the density of CD8+ T cells. The enhanced T1-weighted MRI in the arterial phase was employed with texture analysis. The logistic regression analysis was applied to select the significant features related to IP. The OS-related feature was determined by Cox proportional-hazards model and Kaplan-Meier analysis. IP and OS predictive models were developed using the selected features, respectively. Three wavelets and one 3D feature have favorable ability to discriminate IP, a combination of which performed best with an AUC of 0.919. The inflamed immunophenotyping had a better prognosis than the non-inflamed one. The 5-year survival rates of the two groups were 48.5% and 25.3%, respectively (p < 0.05). The only wavelet-HLH_firstorder_Median feature was associated with OS and used to build the OS predictive model with a C-index of 0.70 (95% CI, 0.57, 0.82), which could well stratify ICC patients into high- and low-risk groups. The 1-, 3-, and 5-year survival probabilities of the stratified groups were 62.5%, 30.0%, and 24.2%, and 89.5%, 62.2%, and 42.1%, respectively (p < 0.05). The MRI texture signature could serve as a potential predictive biomarker for the IP and OS of ICC patients. • The MRI texture signature, including three wavelets and one 3D feature, showed significant associations with immunophenotyping of ICC, and all have favorable ability to discriminate immunophenotyping; a combination of the above features performed best with an AUC of 0.919. • The only wavelet-HLH_firstorder_Median feature was associated with the OS of ICC and used to build the OS predictive model, which could well stratify ICC patients into high- and low-risk groups.