Basis Of Texture Features Research Articles

Application of artificial intelligence in ophthalmology for solving the problem of semantic segmentation of fundus images

The paper presents main aspects of the application of artificial intelligence in ophthalmology for the diagnosis and treatment of eye diseases, considering the problem of semantic segmentation of fundus images as an example. The classic approach to semantic segmentation on the basis of textural features is compared to the proposed approach based on neural networks. Basic problems of using the neural network approach in biomedicine are formulated. We propose a new method for selecting an optimal zone of laser exposure for laser coagulation based on two neural networks. The first network is used for detecting anatomical objects in the fundus and the second one is used for selecting the area of macular edema. The region of interest is formed from the edema area while taking into account the location of anatomical objects in it. A comparative analysis of sev-eral architectures of neural networks for solving the problem of selecting the edema area is carried out. The best results in the selection of the edema area are shown by the neural network architecture of Unet++.

Abstract P3-13-02: A genome-wide association study of mammographic texture variation

Abstract Background: Breast parenchymal texture features capture the patterns of texture variation on a mammogram. One such texture feature is called V, which reflects the gray scale variation on the image. Recent studies have shown that texture features, including V, are associated with breast cancer risk, independent of mammographic density (MD). However, our knowledge on the genetic basis of texture features and the shared genetic components between these features, MD, and breast cancer is limited. Methods: We conducted a genome-wide association study (GWAS) of breast texture variation, including four different assessments of V, in 7,040 European-ancestry women within the Nurses’ Health Studies and Mayo Mammography Health Study cohorts. Genotype data were imputed to the 1000 Genomes Phase 3 version 5 reference panel. The four V assessments, differ by amount of breast edge erosion and image resolution, were generated from digitized film mammograms using fully automated methods. We used linear regression to test the single-nucleotide polymorphism (SNP)-phenotype associations adjusting for age, body mass index (BMI), MD phenotypes (percent density, dense area, and nondense area), and the top four principal components. Multivariate phenotype association tests were performed to estimate summary test statistics of the four V assessments. We compared our results to GWAS of breast cancer and MD phenotypes, and identified shared susceptibility loci. We calculated genetic correlations of V measures with MD phenotypes, breast cancer, overall and by estrogen receptor status, BMI, childhood body fatness, age at menarche, and age at natural menopause. Aggregate tests were performed to assess the relationship between V measures and the above traits using genome-wide significant SNPs for these traits. Results: We identified three novel genome-wide significant loci that are associated with V measures: rs79670367 (8q24.22) in LINC01591, rs113174754 (12q22) near PGAM1P5, and rs138141444 (6q24.1) in ECT2L. 12q22 is a known locus for both MD (nondense area) and breast cancer. 8q24.22 and 6q24.1 have not previously been associated with MD or breast cancer. We identified four additional loci that are associated with MD or breast cancer, or both and are significantly associated with the V measures at the Bonferroni-corrected thresholds accounting for the number of MD or breast cancer SNPs tested: rs335189 (5q23.2) in PRDM6, rs13256025 (8p21.2) in EBF2, rs11836164 (12p12.1) in LOC105369705, and rs17817449 (16q12.2) in FTO, the directions of association are consistent with MD or breast cancer for all four loci. We observed significant genetic correlations between V measures and dense area (rg = 0.79, p = 6.0 × 10-5), percent density (rg = 0.73, p = 1.0 × 10-4), and BMI (rg = -0.36, p = 3.9 × 10-7). Additional significant relationships were observed for nondense area (z = -4.14, p = 3.5 × 10-5) and childhood body fatness (z = -4.91, p = 9.1 × 10-7) from the aggregate tests. Conclusions: This is the first GWAS of breast parenchymal texture features. These findings provide new insights into the genetic basis of texture variation and their associations with MD, breast cancer risk, and other breast cancer risk factors. Citation Format: Yuxi Liu, Xia Jiang, Constance Turman, Celine M Vachon, Rulla Tamimi, Peter Kraft. A genome-wide association study of mammographic texture variation [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-13-02.

The origin of the carbonate-hosted Huize Zn–Pb–Ag deposit, Yunnan province, SW China: constraints from the trace element and sulfur isotopic compositions of pyrite

The Huize Zn–Pb–Ag deposit in SW China has an ore reserve in excess of 5 Mt. with an ore grade of ≥25 wt% Zn + Pb and 46–100 g/t Ag. This deposit is hosted in Carboniferous dolostone and limestone. Sulfide mineralization is dominated by sphalerite, galena, and pyrite. Four types of pyrite (Py1 to Py4) are temporally and spatially related to mineralization distinguished on the basis of textural features and mineral associations. Pyrite 1 to 3 corresponds to the pyrite-sphalerite sub-stage, whereas Py4 corresponds to sphalerite-galena-pyrite sub-stage. Pyrite 1 shows zoned texture composed of an inclusion-rich core and an inclusion-free rim, whereas Py2, Py3, and Py4 show replacement relic or overgrowth textures. The zoned texture in Py1 was formed by multiple stages of ore fluids, whereas replacement relic texture in Py2 to Py4 was formed by replacement of pyrite by late Pb-Zn-rich fluids. Trace element variation in pyrite results from a combination of mineral inclusions, co-precipitating minerals, and various fluid compositions. Sphalerite, pyrite, and galena have δ34S values of 10.4–23.5‰, suggesting that sulfur was probably derived from the thermochemical reduction of marine sulfates. The Huize pyrite has Co and Ni concentrations (0.02–9.5 ppm and 0.08–143 ppm, respectively) and Co/Ni ratios (~0.01–2.63) similar to pyrite from sedimentary exhalative deposits, submarine hydrothermal vents, and sedimentary pyrite, which may be due to pyrite precipitation from low-temperature (<~250 °C), basin brines or seawater. The Co/Ni ratios of the Huize pyrite are lower than those (~0.2–7.2) of pyrite from Mississippi Valley-type Zn-Pb deposits. The Huize pyrite has Co and Ni contents lower than those associated with volcanogenic massive sulfide, iron oxide copper gold, and porphyry Cu deposits, arguing against the involvement of a magmatic/volcanic component and a direct genetic link between Permian Emeishan basalts and Pb-Zn mineralization. Combining with fluid inclusion temperatures and silver grade, we define the Huize deposit as a transitional type between Mississippi Valley-type and high-temperature carbonate-replacement Zn-Pb deposits.

EXTRACTION OF TEXTURE FEATURES BY USING GABOR FILTER IN WHEAT CROP DISEASE DETECTION

Like country India, there are so many people depending upon agriculture. In this area, many farmers don’t know about new diseases which are impacting on their farm. As the disease changes, the disease control policy also changes. So many farmers have very sharp observation on crop diseases, but whenever there is new diseases fall on crops then problems occur. Climate also changes instantly many of times, because of such reasons farmers unable to understand various diseases. If farmer unable to predict that diseases quickly then it will affect life of crops. Indirectly it gets affects on total productivity of farm. As we are well known about that world facing lot of problems due rapid growth in population. So our goal is to increase agricultural productivity using image processing technology which can help farmer in great extent [7]. In this research work, we are trying that crop disease using Artificial neural network (ANN) which work very effectively. First of all, we have provided an digital image which is taken by digital camera. That image given to Gaussian filter firstly then transferred to adaptive median filter to filter out noise present inside image. Gaussian filter removes Gaussian noise which is present inside image. Adaptive noise filter removes impulsive noise which is present inside image. Also it will reduce distortions which are present inside images. Then image transferred to segmentation part. In image segmentation we have choose CIELAB color space method to extract color components properly. For segmentation we have used Gabor filter. After this we distinguish crop diseases on the basis of texture features which are extracted by Gabor filter [6].

Computer-Aided Breast Cancer Diagnosis Based on the Analysis of Cytological Images of Fine Needle Biopsies

The effectiveness of the treatment of breast cancer depends on its timely detection. An early step in the diagnosis is the cytological examination of breast material obtained directly from the tumor. This work reports on advances in computer-aided breast cancer diagnosis based on the analysis of cytological images of fine needle biopsies to characterize these biopsies as either benign or malignant. Instead of relying on the accurate segmentation of cell nuclei, the nuclei are estimated by circles using the circular Hough transform. The resulting circles are then filtered to keep only high-quality estimations for further analysis by a support vector machine which classifies detected circles as correct or incorrect on the basis of texture features and the percentage of nuclei pixels according to a nuclei mask obtained using Otsu's thresholding method. A set of 25 features of the nuclei is used in the classification of the biopsies by four different classifiers. The complete diagnostic procedure was tested on 737 microscopic images of fine needle biopsies obtained from patients and achieved 98.51% effectiveness. The results presented in this paper demonstrate that a computerized medical diagnosis system based on our method would be effective, providing valuable, accurate diagnostic information.

PHOSPHATE ASSEMBLAGES FROM THE BRISSAGO GRANITIC PEGMATITE, WESTERN SOUTHERN ALPS, SWITZERLAND

We describe a first significant occurrence of graftonite–triphylite-bearing LCT-type granitic pegmatite in the Alps. In 1999, in the Brissago area, in the western Southern Alps, Canton Ticino, Switzerland, we found a boulder containing an almost complete section of a phosphate-bearing pegmatite dike. The zoned pegmatite is composed by four main units: 1) an aplite unit, 2) a plagioclase + quartz graphic unit, 3) a coarse-grained plagioclase unit, and 4) a quartz core containing several phosphate masses up to 10 cm in diameter. On the basis of textural features and results of chemical analyses, we distinguish two phosphate associations. Association I, of magmatic origin, is represented by lamellar intergrowths of graftonite and triphylite. The leaching of Li and oxidation of Fe have led to the formation of ferrisicklerite and heterosite in replacement of triphylite, whereas the origin of kryzhanovskite and stanekite remains unclear. Association II, of metamorphic origin, consists of a polycrystalline aggregate of graftonite and triphylite. Exceptionally, primary maghagendorfite grains are present. Non-oxidizing low-temperature processes led to the formation of ludlamite and vivianite in replacement of triphylite. Locally, at the rim of phosphate masses, in close association with apatite-(CaOH), jahnsite and mitridatite formed at the expenses of graftonite. The chemical characteristics of the phosphate assemblages and associated mafic silicates indicate that this phosphate-bearing pegmatite dike belongs to a geochemically primitive system. Our U/Pb dating of zircon (242 ± 2.8 Ma) allows us to relate the formation of this dike to the Mid-Triassic tectonomagmatic cycle, characterized by the evolution of a magmatic arc well represented in the Southern Alps. In this geodynamic environment, a pegmatite-forming melt formed after partial melting of metapelites belonging to the Kinzigitic Complex of the Ivrea–Verbano zone.

Retrieving forest structure variables based on image texture analysis and IKONOS-2 imagery

Remote sensing techniques have been seen as valuable and low-cost tools for frequent forest inventory purposes. However, estimation errors of relevant forest structure variables remain too high for operational use of high spatial resolution satellite imagery, such as Landsat TM/ETM and SPOT HRV, in temperate forests. Very high spatial resolution images that have been acquired by new commercial satellites, such as IKONOS-2 or QuickBird, are expected to reduce estimation errors to a level that is acceptable by foresters. This study assessed the capability of 1-m resolution IKONOS-2 imagery to estimate the five main forest variables—age, top height, circumference, stand density and basal area—in even-aged common spruce stands. They were estimated on the basis of texture features that were derived from the grey-level co-occurrence matrix (GLCM). The coefficients of determination, R 2, of the best models ranged from 0.76 to 0.82 for top height, circumference, stand density and age variables. Basal area was found to be weakly correlated to texture variables ( R 2 = 0.35). Relative prediction errors of four out of the five studied forest variables were comparable to the usual sampling inventory errors (top height: 10%; circumference: 15%; basal area: 16%; age: 18%), but the stand density estimation error (29%) remained too high for use in forest planning. The sensitivity analysis to the GLCM parameters showed that the most important parameters were the texture feature, the displacement and the window size. The orientation parameter had minimal effects on the R 2 values, even if it influenced the values of the texture features.

THE COMPOSITION OF Co-Ni-Fe SULFARSENIDES, DIARSENIDES AND TRIARSENIDES FROM THE SAN JUAN DE PLAN DEPOSIT, CENTRAL PYRENEES, SPAIN

The San Juan de Plan deposit, in the central Pyrenees of Spain, is situated in an ankeritic horizon in Silurian black shales. The main association consists of pyrite, sulfarsenides, diarsenides, and Co triarsenides (skutterudite), as well as a late stage consisting of cobaltite, Fe, Cu and Bi sulfides and native Bi. The sulfarsenides are arsenopyrite, alloclasite, members of the gersdorffite‐ cobaltite solid-solution series, arsenic-rich gersdorffite and cobaltite, whereas the diarsenides range from rammelsbergite to safflorite. Rammelsbergite is characterized by a composition within the range (Ni0.71‐0.97Co0.02‐0.29Fe0.00‐0.02)As1.75‐2.01S0.00‐0.26, exceeding the theoretical value (As1.93S0.07) given by Yund (1962). Two generations of the gersdorffite‐cobaltite solid-solution series have been found on the basis of textural features and chemical composition; the first, (Co0.10‐0.77Ni0.07‐0.80Fe0.03‐0.58)As0.95‐ 1.27S0.75‐1.06, shows a positive correlation between S and Fe contents, and between Co and As contents, whereas the second one, (Co0.02‐0.91Ni0.02‐0.96Fe0.00‐0.28)As0.99‐1.38 S0.67‐1.01, exhibits a considerable compositional field in the Fe-poor region of the system CoAsS‐NiAsS‐FeAsS , providing evidence of extensive substitution of Co by Ni, with a positive correlation between As and Ni contents. Arsenic-rich gersdorffite displays a broad range of As and S mutual substitution, (Ni0.65‐0.98Co0.02‐0.27Fe0.00‐0.06)As1.37‐ 1.83S0.22‐0.64, exceeding the experimental value of NiAs1.77S0.23 reported by Yund (1962). The presence of alloclasite and the extent of the solid-solution fields of the cobaltite‐gersdorffite series suggest that the main assemblage of minerals at the San Juan de Plan deposit formed at temperatures as high as 600°C. Preliminary geothermometric data obtained on graphite support such a high temperature.

北海道中央部旭川市西方の神居古たん変成岩類に含まれるアルカリ角閃石の産状と変成史

Two stages of metamorphism, initial and main stages, are distinguished on the basis of textural features and chemical compositions of sodic amphiboles and associated minerals in the Kamuikotan high-pressure and low-temperature metabasites. The initial stage is detected by sodic amphiboles rimmed by actinolite and enclosed by epidote porphyroblast in the metabasites. These sodic amphiboles are optically and chemically heterogeneous, and range in chemical composition from crossite to glaucophane (Al2O3 content: 2.6-10.5 wt.%). The main stage is characterized by the occurrence of epidote, pumpellyite, chlorite, calcite, hematite and jadeite component poor sodic pyroxene in the metabasites. Amphiboles coexisting with these minerals are actinolite, winchite and magnesioriebeckite (Al2O3 content: 0-2.9 wt.%). These variations of modes of occurrence and chemical compositions of sodic amphiboles suggest that some studied metabasites initially had been metamorphosed under low grade condition with a relatively high ratio of pressure/temperature and were subsequently converted to assemblages of the main stage, correspond to the pumpellyite-actinolite facies, by decline in the pressure/temperature ratio with local effect of deformation and metamorphic fluid.

Evidence for the supergene origin of alunite in sediment-hosted micron gold deposits, Nevada

On the basis of textural features and spatial distributions, both supergene and hypogene origins have been proposed for alunite in sediment-hosted micron gold deposits. If alunite is indeed primary, its presence provides important constraints on the chemistry of the ore fluid. Thus, determination of the genesis of alunite is critical to the understanding of the geologic history of these ore deposits.At the Post deposit in Nevada K/Ar ages of alunite are much younger than the age of mineralization. Sulfur isotope compositions of alunite from several sediment-hosted micron gold deposits are virtually identical to those of sulfides, suggesting that alunite sulfur came from quantitative (closed-system) oxidation of primary sulfide minerals during weathering. Sulfur and oxygen isotope compositions of barite are markedly different from those of alunite. The oxygen and hydrogen isotope compositions of alunite fall in the supergene alunite fields; data for cogenetic kaolinitc fall on the kaolinitc line, whereas those for hypogene kaolinite do not. From all available petrographic and geochemical data, it would appear that alunite in the micron gold deposits of Nevada is entirely of secondary supergene origin.

The Honeymoon Well nickel sulfide deposits, Western Australia

The Honeymoon Well Ni-Fe sulfide deposits occur in a thick assemblage of altered intrusive dunires and extrusive komatiites in an Archean greenstone belt near Wiluna in the eastern Yilgarn Block, Western Australia. The sulfides, host rocks, and surrounding intermediate to basic volcanics have undergone greenschist facies metamorphism in a low-strain environment; ultramafic rocks have been totally serpentinized and/or carbonated, but igneous textures are commonly well preserved. The coarse-grained dunitic rocks are interpreted as slowly cooled igneous cumulates on the basis of textural features and mineralogical and cryptic layering. Spinilex textures within the ultramafic assemblage, and rare interbedded black shale, suggest an extrusive origin for some of the rocks, and the dunite bodies are possibly consanguineous subvolcanic sills.Disseminated Ni-Fe sulfide mineralization occurs in two zones up to 100 m wide near the western and eastern contacts of the ultramarie assemblage. Pentlandite is the dominant sulfide in lizardite serpentinites and occurs in an intercumulus position, moulded around former olivine grains. In talc-carbonate rocks locally developed adjacent to the ultramarie contacts, pyrrhotite is also common. Exploration to date has outlined two major zones of Ni-Fe sulfide mineralization with mean grades ranging from 0.5 to 1 percent Ni.Honeymoon Well sulfide mineralogy is Ni rich in comparison with most other intrusive dunite-associated deposits in the region, and regular differences in S/Ni, Fe/Ni, and Ni/Cu ratios appear to correlate with the metamorphic setting and host rock mineralogy.