Thin-section Images Research Articles

Quantification of enamel decussation in capuchin monkeys

Introduction & Objective – An intricate suite of highly complex enamel has led robust capuchin monkeys to be uniquely capable in breaking down hard-object foods as compared to other cebid monkeys, and in routinely breaking down extremely mechanically challenging foods as compared to other capuchins. The goal of this study is to compare dental enamel complexity in images of dental thin sections from a sample of robust vs. gracile capuchins using Image Compression Ratio (ICR) analysis. We hypothesize that: 1) robust capuchins will exhibit an overall pattern of higher ICR, and therefore more complex enamel, compared to gracile capuchins, and 2) the difference in ICR between the two groups will be greater for anterior as compared to posterior teeth. Materials and Methods – 113 images of thin sections from three gracile capuchin species and from unidentified species of robust capuchins, Sapajus spp were captured using a Leica-Leitz DMRX/E Universal Microscope configured with a Marzhauser motorized stage, phase contrast, and circularly polarizing filters. The images were then digitally processed to delete non-enamel components from the images and any other factor that can affect the compression ratio. After various compressions, the ICR values were compared across our images using standard non-parametric Kruskal-Wallis and Mann Whitney tests. Results – As for the first hypothesis, we found no significant difference in ICR between robust and gracile capuchins across all teeth (p = 0.35) using Mann-Whitney U tests. Likewise, we found no significant difference in ICR for all tooth categories when all gracile capuchin species were examined separately rather than assessed as one category (Kruskal Wallis, p = 0.48), nor was there a significant difference among tooth types (canines, premolars, and molars) in the total dataset (p = 0.36). As for the second hypothesis, we separately compared anterior (canines) and posterior teeth (premolars and molars) across the same taxonomic groupings. We once again found no difference in ICR between robust and gracile capuchins for the posterior dentition (p = 0.28). However, the anterior tooth sample did display a significant difference between the two groups (p = 0.04). Conclusion – Our ICR results indicate that premolars and molars among gracile and robust capuchins exhibit similar degrees of enamel complexity, but that canines in robust capuchins display increased complexity reflecting high specialization in this group for hard object feeding in the anterior masticatory apparatus. Significance/Implication – These findings provide further evidence to the hypothesis that the need to break apart obdurate foodstuffs with the anterior – as opposed to posterior – teeth has driven much of the specialized craniodental anatomy seen in robust capuchins. Future work including further sampling may help assess the degree in which different species of robust capuchins differ from one another. The findings of this research may also lead us to assess differences among the robust species, considering previous studies suggesting that tool use in some species may lessen the need for generation of high anterior bite forces compare capuchins to other primate hard object feeders such as the pitheciines and Paranthropus, the latter of which would be particularly useful for assessing patterns of hard object feeding in hominin evolution.

Application of deep learning for semantic segmentation of sandstone thin sections

Sedimentary petrology is the basis for most mineral and textural identification in sandstones. Automating mineralogical interpretation of an entire thin section image has many practical applications, including improved geological understanding, input of spatial distribution of mineralogy and grain size for petrophysical evaluations, and integration with 3D imaging modalities (micro-CT, nano-CT). We investigate the application of Convolutional Neural Network (CNN) based supervised semantic segmentation methods for predicting pixel-scale mineralogy using 2D RGB images of sandstones acquired by transmission light microscopy. Models were trained to interpret a simple binary pore-mineral (grain) segmentation and a 10-class segmentation (porosity, quartz, feldspar, rock fragments, carbonate grains, opaque grains, quartz cement, carbonate cement, clay cement, and hydrocarbons filling pores). For the 2-class classification framework to distinguish between pores and minerals, most models lead to satisfactory results with acceptable accuracy. For the 10-class classification framework, models trained with Deeplab V3+ Resnet-18 network yield more continuous results compared to those based on VGG networks. We conclude that the effectiveness of the models, in predicting a petrology class in a thin section, strongly correlates with the amount of labeled data available to train the model to interpret the class in question. Semantic segmentation models, based on CNNs, can produce encouraging results for a 10-class petrological classification framework of an entire thin section image and thus provide a complete scene understanding which is difficult to produce manually.

A microscale simulation methodology for subsurface heterogeneity quantification using petrographic image analysis with multiscale mixed finite element method

A microscale heterogeneity quantification methodology is presented, which uses petrographic image analysis with multiscale simulation. Petrographic images of micro-porous carbonate rocks are converted into microscale models with help of an image analysis workflow, which helps in assigning specific porosity, permeability and petro-physical properties to the microscale model. Thin section images are discretized into classes of specific porosity. Each specific porosity class is then assigned a permeability using the Kozeny-Carman equation for permeability. Doing so, the thin section image is transformed into a 2D porosity and permeability grid preserving the true heterogeneity at microscale. A multi-scale flow simulation using mixed finite element formulation is then used to perform fast flow simulation on the reconstructed grid representing the microscale model. Although researcher have used pore-network or Lattice-Boltzmann modeling techniques to understand impact of fine scale heterogeneities on fluid flow in porous media, these methodologies are computationally very expensive and impractical for large scale application. Authors present a pragmatic and time-efficient approach for simulating flow experiments at the microscale providing valuable insights into the impact of petrographic components on flow characteristics. A methodology which uses multi-scale methods flow simulations at microscale level has not been presented before in literature.

Improving permeability estimation of carbonate rocks using extracted pore network parameters: a gas field case study

Despite numerous studies carried out on permeability estimation from either 2D/3D images or models, a precise evaluation of the permeability for carbonate rocks is still a challenging issue. In this study, the capability and advantages of pore network parameters extracted from 2D thin-section images as inputs of intelligent methods for permeability estimation of carbonate rocks are explored. Pore network extraction in image processing is an effective approach for microstructure analysis. A physically practical pore network is not just a portrayal of the pore space in the context of both morphology and topology, but also a valuable instrument for predicting transport properties precisely. In the current research, a comprehensive workflow was first presented to extract the pore network parameters from a set of core thin-section microscopic images from the carbonate reservoir rock of the South Pars gas field located in the southern borders of Iran. Subsequently, an artificial neural network (ANN) model was designed to predict the permeability of the considered samples using the extracted pore network parameters. To highlight the efficiency of the proposed approach, the second ANN model was implemented to estimate the permeability of the samples using the conventional well log data. The quantitative comparison of the obtained results using both ANN-based models reveals a significant enhancement in the predicted permeability through the extracted pore network parameters.

Automatic classification of volcanic rocks from thin section images using transfer learning networks

In this study, efficient deep transfer learning models are proposed to classify six types of volcanic rocks, and this paper has a novelty in classifying volcanic rock types for the first time using thin section images. Convolutional neural network-based DenseNet121 and ResNet50 networks, which are transfer learning methods, are used to extract the features from thin section images of rocks, and the classification process is carried out with a single-layer fully connected neural network. The proposed models are trained and tested on 1200 thin section images using four different optimizers (Adadelta, ADAM, RMSprop, SGD). AUC, accuracy, precision, recall and f1-score are used as performance metrics. Proposed models are run 10 times for each optimizer. DenseNet121 classifies volcanic rock types using RMSprop with an average accuracy of 99.50% and a maximum of 100.00%, and ResNet50 classifies using ADAM with an average accuracy of 98.80% and a maximum of 99.72%. Thus, the applied deep transfer learning is promising in geosciences and can be used to identify rock types quickly and accurately.

The Cellular Organization of the Mammary Gland: Insights From Microscopy.

Despite rapid advances in our knowledge of the cellular heterogeneity and molecular regulation of the mammary gland, how these relate to 3D cellular organization remains unclear. In addition to hormonal regulation, mammary gland development and function is directed by para- and juxtacrine signaling among diverse cell-types, particularly the immune and mesenchymal populations. Precise mapping of the cellular landscape of the breast will help to decipher this complex coordination. Imaging of thin tissue sections hasprovided foundational information about cell positioning in the mammary gland and now technological advances in tissue clearing and subcellular-resolution 3D imaging are painting a more complete picture. In particular, confocal, light-sheet and multiphoton microscopy applied to intact tissue can fully capture cell morphology, position and interactions, and have the power to identify spatially rare events. This review will summarize our current understanding of mammary gland cellular organization as revealed by microscopy. We focus on the mouse mammary gland and cover a broad range of immune and stromal cell types at major developmental stages and give insights into important tissue niches and cellular interactions.

Quantification of enamel decussation in gracile and robust capuchins (Cebus, Sapajus, Cebidae, Platyrrhini).

Multiple behavioral and biomechanical analyses have demonstrated that capuchin monkeys (Cebus and Sapajus) are specialized for breaking down hard-object foods as compared to other cebid monkeys. In addition to a complex suite of craniodental adaptations, it has specifically been demonstrated that capuchins possess highly complex dental enamel, with extensive Hunter-Schreger banding and other decussation, that likely serve as an adaptation to resist crack propagation during hard-object feeding. Furthermore, it has been demonstrated that robust capuchins (Sapajus spp., formerly Cebus apella) demonstrate further adaptation for hard-object feeding than other capuchins, routinely breaking down extremely mechanically challenging foods. However, there has been no comparison of dental enamel complexity in robust versus gracile capuchins, to assess whether the dental enamel in Sapajus follows this same pattern of further specialization. Therefore, this study compares dental enamel complexity in images of dental thin sections from a sample of robust versus gracile capuchins using image compression ratio (ICR) analysis. ICR is a variable that correlates with enamel complexity, such that higher ICR values are indicative of increased complexity in the form of enamel decussation. We found no significant difference between robust and gracile capuchins when assessing all teeth in our sample together, however, we did find that robust capuchins have significantly higher ICR values than gracile capuchins for canine teeth, specifically. Our results support prior studies suggesting that robust capuchins are specialized to generate increased masticatory loads with their anterior dentition, specifically, as compared to gracile species.

A method to extract the content, radius and specific surface area of maceral compositions in coal reservoirs based on image modeling

Macerals are the basic components of coal reservoirs. Identification of macerals in coal reservoirs is of great practical significance. In this paper, ten coal samples and six typical maceral images, including megaspores, resinite, cutinite, barkinite, microspores and alginate, from Permian coal samples of the Qingshui Basin are taken as examples for quantitative image-algorithm analysis. Combined with the functions (e.g., “bwlabel”, “ginput”, “Regionprops”, and “bwboundaries”), algorithms and the “Color Thresholder” application in MATLAB, the contents, radii and specific surface areas of different types of maceral structures are identified and calculated. Binarization of thin-section images by means of the RGB and HSV modes in the "Color Threshold" in MATLAB is the first step of automatic recognition. Various functions and algorithms are used to capture pixel distributions in binary images with maceral distributions. The contents of six representative macerals were 58.747%, 11.302%, 20.563%, 48.293%, 24.044% and 11.469%, with corresponding average radii of 150.6976 μm, 93.5590 μm, 88.3547 μm, 93.5590 μm, 114.8465 μm and 84.3030 μm, respectively; the specific surface areas of the macerals were 0.1172 m2/g, 0.6375 m2/g, 0.7646 m2/g, 0.4169 m2/g, 0.1261 m2/g and 0.7374 m2/g, respectively. To determine the accuracy of the quantitative analysis process, we compared the results of traditional recognition methods with the results of the image-function identification method. The results show that there is a good linear relationship between the two experiments for exinite, inertinite and vitrinite contents in coal, with R2 values of 0.9275, 0.9907 and 0.9930, respectively. Due to the influence of microscopy, artificial parameters in the imaging process, and the influence of impurities, inclusions or dark minerals in the original coal samples, some optical noise was present inside and outside the maceral components. We establish a noise-reduction system that is based on the "imerode" and "imdilate" functions to eliminate noisy areas in nonmaceral features and repair damaged spaces in the components caused by noise.

SCRUM-Japan genesis virtual sequencing (VSQ) project: A novel algorithm combining deep learning (DL) with pathological diagnostics to enable the prediction of BRAF mutations and microsatellite instability (MSI) in advanced colorectal cancer (CRC).

112 Background: Numerous genetic and epigenetic abnormalities may lead to various morphologies of cancer. However, exactly which gene abnormality causes which morphology is unknown. The VSQ Project aims at investigating a novel algorithm by synergistically fusing DL technology and pathological diagnostics for the prediction of cancer genome abnormalities. This was achieved by elucidating the association between the morphological findings and genetic abnormalities, including BRAF V600E mutations and MSI status directly linked to the therapeutic strategies for advanced CRC patients (pts). Methods: Clinicopathological-genomic integrated DB derived from SCRUM-Japan GI-SCREEN, a nation-wide cancer genome screening project including CRC, were used. A total of 1,657 images of thin sections (one representative image per pt) cut from formalin-fixed and paraffin-embedded (FFPE) tissue specimens from primary or metastatic tumors with genetic abnormalities confirmed by next-generation sequencing (NGS) were investigated; 1,234 and 423 images (one per pt) were used for training and validation cohorts, respectively. First, we developed image-prediction models based on the morphological features precisely annotated by the single central pathologist, and then constructed the DL algorithms (gene-prediction models) that enabled the prediction of gene abnormalities by using images filtered by the image-prediction models. Results: We achieved high accuracy of AUC > 0.90 for 12 features among the 33 morphological features analyzed. Next, we created several DL algorithms that enabled the prediction of BRAF mutations and MSI. The prediction level reached a high accuracy of AUC = 0.955 for the BRAF mutations and AUC = 0.857 for MSI in the training cohort. We verified the AUCs in the validation cohort and achieved AUC = 0.831 and 0.883 for BRAF mutations and MSI, respectively. Conclusions: Our findings suggest that VSQ can appropriately predict BRAF mutation and MSI status in advanced CRC, potentially without performing NGS tests. VSQ may also enable prompt initiation of systemic treatments in CRC patients as well as establish an unprecedented next-generation pathology in the near future.

Major, trace-elemental and sedimentological characterization of the upper Ordovician Wufeng-lower Silurian Longmaxi formations, Sichuan Basin, south China: Insights into the effect of relative sea-level fluctuations on organic matter accumulation in shales

The relative sea-level fluctuations play a vital role in the organic matter enrichment of black shales across the world. The Wufeng-Lower Longmaxi shales deposited during the Late Ordovician to Early Silurian stage are accompanied by the global transgression and regression, whereas the role of relative sea-level changes on organic matter accumulation has been rarely discussed. In this study, we integrated TOC content, X-ray diffraction (XRD) mineralogical composition, thin section and scanning electron microscope (SEM) images, elemental geochemical data, and petrographic features of pyrite framboids to investigate the varying paleoenvironmental conditions and the effect of relative sea-level changes on organic matter accumulation under a robust sequence stratigraphic framework.Based on detailed petrographic observations and mineralogical analysis, six major lithofacies are identified in the Wufeng-Lower Longmaxi formations, including siliceous mudstone, clay-rich siliceous mudstone, siliceous-argillaceous mixed shale, silty mudstone, siltstone, and bentonite. TOC content is generally higher in the transgressive systems tracts (relative sea-level rise stage) than in the highstand systems tracts and falling stage systems tract (relative sea-level fall stage). Multiple geochemical proxies indicate contrasting paleoenvironmental changes between relative sea-level rise and fall stage. Relative sea-level rise resulted in elevated productivity, predominantly anoxic conditions, low detrital flux, and moderate water mass restriction, as indicated by high biogenic Si content, high Mo-EF (enrichment factor of element Mo), small-sized pyrite framboids, low Al content and moderate Mo/TOC ratio. On the contrary, relative sea-level fall generally resulted in decreased productivity, dysoxic conditions, increased detrital flux, and strongly restricted conditions. The correlations between TOC content and geochemical proxies suggest that organic matter accumulation was a combined effect of paleoproductivity, paleoredox conditions, detrital influx, and water mass restriction. The variation of these factors was strongly influenced by relative sea-level fluctuations related to Hirnantian glaciation and regional tectonic movements during the Late Ordovician-Early Silurian transitional stage.

The Influence of Quantitative Parameters Measured on CT with Different Slice Thicknesses on Evaluating the Growth of Pulmonary Nodules

We did this study to investigate the effect of thick (5mm) and thin (1 or 0.625 mm) slice thickness of CT images on evaluating pulmonary nodules' growth to improve their diagnostic accuracy. The clinical and CT data of 251 patients with lung nodules and two follow-up CTs from October 2016 to October 2019 were analyzed retrospectively. Malignant nodules were confirmed by pathology, and benign nodules were confirmed by pathology or follow-up. Two radiologists double-blindly assessed the CT features (density, shape, lobes, border), maximum diameter, and volume of nodules on the thick (5MM) and thin (≤1MM) images of two follow-up CTs. We use One-way analysis of variance for quantitative data; the X2 test or FISHER exact probability method was used for qualitative data; and the ROC curve was used to analyze the diagnostic power of nodule size, volume, and change in differentiating benign and malignant lesions. Among 251 pulmonary nodules, 117 (46.6%) benign nodules and 134 (53.3%) malignant nodules. During the CT follow-up, the volume measured on the thick-section image, the diameter, and the volume measured on the thin-section image were statistically different in benign and malignant lung nodules (P<0.001). In contrast, the diameter measured on the thick-section image was similar between these two groups (P=0.328). For benign and malignant pulmonary nodules, the diameter, volume, and change measured on the thin-section image were significantly larger than the thick-section image's data (P<0.001). The ROC curve showed that the diagnostic efficiency of volume was higher compared to the diameter. There were significant differences in nodule type, density change, shape, lobulation, and pleural retraction between benign and malignant nodules for CT features. Accurately assessing the volume changes combined with CT characteristics will help improve lung nodules' diagnosis accuracy. Volume measured on thin-section (1mm) CT images is the best quantitative parameter for assessing the change of pulmonary nodules. Combining Volume change with CT characteristics would help to improve the diagnostic accuracy.

Optimization of negative stage bias potential for faster imaging in large-scale electron microscopy

Large-scale electron microscopy (EM) allows analysis of both tissues and macromolecules in a semi-automated manner, but acquisition rate forms a bottleneck. We reasoned that a negative bias potential may be used to enhance signal collection, allowing shorter dwell times and thus increasing imaging speed. Negative bias potential has previously been used to tune penetration depth in block-face imaging. However, optimization of negative bias potential for application in thin section imaging will be needed prior to routine use and application in large-scale EM. Here, we present negative bias potential optimized through a combination of simulations and empirical measurements. We find that the use of a negative bias potential generally results in improvement of image quality and signal-to-noise ratio (SNR). The extent of these improvements depends on the presence and strength of a magnetic immersion field. Maintaining other imaging conditions and aiming for the same image quality and SNR, the use of a negative stage bias can allow for a 20-fold decrease in dwell time, thus reducing the time for a week long acquisition to less than 8 h. We further show that negative bias potential can be applied in an integrated correlative light electron microscopy (CLEM) application, allowing fast acquisition of a high precision overlaid LM-EM dataset. Application of negative stage bias potential will thus help to solve the current bottleneck of image acquisition of large fields of view at high resolution in large-scale microscopy.

Role of MRI in Evaluation of Focal Liver Lesions: A Cross-sectional Study

Introduction: The non-invasive diagnosis of liver lesions is usually achieved with contrast material-enhanced computed tomography and Magnetic Resonance (MR) imaging. Dynamic three dimensional gradient-recalled-echo MR imaging provides dynamic contrast-enhanced thin-section images with fat saturation and a high signal-to-noise ratio and is excellent for the evaluation of various focal hepatic lesions. Aim: To explore the effectiveness and hence the clinical utility of plain and contrast MR characterisation of focal hepatic lesions in patients and their correlation with the histopathological findings. Materials and Methods: The present study was a multicenteric study done on both Inpatients and Outpatients of Department of Radiology of Tertiary care hospitals in Bengaluru, from January 2020 to August 2020. A total of 42 patients diagnosed with focal liver lesions by Ultrasonography were included in this study. Plain MRI study was done by using spin echo technique and contrast MRI study was done after rapid bolus injection (0.1 mmol/kg body weight) of gadolinium dimeglumine. Pathological and post-operative histopathological findings were noted. Microsoft excel software was used for data analysis and tables were prepared for comparison of collected data. Specificity and sensitivity were calculated for diagnostic accuracy of MRI. Results: This study included forty-two patients in total, twenty-five male and seventeen female patients, in age group of 15 to 84 years with a mean age of 48 years. Metastasis and Hepatocellular Carcinoma (HCC) were most common malignant lesions in liver while Hemangiomas were most common benign lesions followed by simple and complex cysts and hepatic adenoma. The present study calculated the specificity and sensitivity of MRI of focal hepatic lesions and correlated the findings clinically, pathologically and histopathologically. The specificity of MRI for characterisation of focal liver disease was high (90.4%). Specificity of MRI for HCC was found to be 96.7 % and that for metastasis was 97% while sensitivity for Haemangiomas is found to be 100%.MRI is superior compared to other modalities like ultrasound and CT in having high spatial resolution. Multiple sequences like T1, T2 and fat suppression techniques like IN PHASE, OUT OF PHASE helps in demonstrating lesion characters like architecture, margin, haemorrhage, fat component and vascular invasion of the lesions. Conclusion: MRI is valuable for the characterization of focal liver lesions detected on Ultrasonography. Basic MRI sequences along with chemical shift imaging sequences and dynamic contrast imaging has significant role in characterising the focal liver lesions. Despite that, It should be the modality choice for the lesions which are undetected or doubtful on ultrasound and CT examinations.

An innovative image-based workflow for quantitative pore structure characterization in reservoir rocks

Pore connectivity is a key parameter to affect reservoir quality (e.g. permeability). Routinely, this parameter is assessed with special core analysis (i.e. mercury injection capillary pressure analysis). In this paper, an innovative image-based pore structure workflow is proposed to quantitatively delineate the connectivity of pore space of reservoir rocks by using widely available thin section images. By borrowing the concept of mercury injection capillary pressure (MICP), this workflow is able to generate pseudo MICP curves that can be used to quantitatively characterize pore throat/body size distribution and pore connectivity of pore networks. This workflow consists of four steps: (1) HSV threshold-based pore identification; (2) K-D tree structure-based pore segmentation; (3) enhanced watershed-driven pore throat detection, and (4) pore throat distribution generation derived with a media-axis-maximal-circle algorithm. The pseudo MICP curves generated by the workflow can be used to quantitatively assess the variations in pore size, pore throat size, pore geometrical complexity as well as the density of pores of reservoir rocks. Two connectivity indices derived from pseudo MICP curves are introduced to evaluate the connectivity of pore network of reservoir rocks. This workflow provides an innovative and efficient tool for geologists and reservoir engineers to perform quantitative reservoir rock quality evaluation by utilizing widely-available thin section images.

Single isotropic 3D fast spin echo sequence compared with conventional 2D sequences for detecting meniscal and cruciate ligament tears in the knee

BackgroundThis work was conducted to assess the diagnostic efficiency of isotropic three-dimensional VISTA-fast spin echo versus standard two-dimensional fast spin echo at 1.5 T MRI, in the assessment of internal knee derangement in symptomatic patients, aiming to obtain similar diagnostic accuracy in a shorter time span, with reduction of partial volume artifacts by thin continuous sections.ResultsThis was a non-randomized control study including 39 patients (32 male and 7 females, mean age 37 years old). A correlative study was done utilizing MRI standard 2D FSE (protocol A) versus 3D-VISTA-FSE (protocol B) for medial meniscus (MM) and lateral meniscus (LM), as well as anterior cruciate ligament ACL lesions, comparing the MRI results with the findings of arthroscopy as the gold standard. Both protocols depicted medial meniscus lesions with accuracy, specificity, and sensitivity (97.44%, 96.30%, and 100% respectively), lateral meniscus lesions with accuracy, specificity, and sensitivity (97.44%, 100%, and 50% respectively), and ACL lesions with accuracy, specificity, and sensitivity (100%, 100%, and 100% respectively), while there were no PCL lesions depicted through the study population. Comparing the time factor between both protocols revealed protocol A to consume 13.7 min, while protocol B consumed 6.6 min.ConclusionThree-dimensional isotropic VISTA-FSE sequence, although having similar accuracy in diagnosing cruciate and meniscal lesions as the standard sequences, facilitates thin-section data acquisition and multi-planar image reformation in standard and non-standard planes, without intersection gaps that are crucial for the detection and dissection of compound structures; also, it allows a shorter time span, which is more advantageous for patients, particularly the traumatized and emergency patients.

Distribution and cementation characteristics of beachrocks along southern, southwestern and western coast of Sri Lanka

Beachrocks are sedimentary structures where gravelly or sandy beaches have been transformed into rock outcrops formed through precipitation of connective cements amid their interstices. They are well-noted coastal features along the coastal belt of Sri Lanka due to the prevalent tropical climate. This study was aimed at gathering data on surface nature and cementation characteristics of beachrock occurrences along a part of Sri Lankan shoreline through field observations and a series of analyses including X-ray diffraction, X-ray fluorescence, scanning electron microscopy (SEM) techniques and petrographic thin-section analysis. The combined research findings from different techniques are also employed as a preliminary step to determine the formation mechanism of the studied beachrocks. The seaward-inclined low-angle beds running parallel to present shoreline are composed mostly of sandstone with occasional conglomerate. Almost all the beachrocks are made of quartz grains amalgamated by cement. One remarkable feature of Sri Lankan beachrocks is the presence of heavy minerals generally in thin lamina form. The cementing agents are predominantly composed of metastable carbonate phases, high magnesium calcite (HMC) and aragonite (Ar) with varying microfabrics and textures. From SEM examinations and thin-section images, main morphologies identified are acicular Ar, scalenohedral magnesium calcites along with bridge cements and micritic coatings which are typical of a marine-phreatic precipitation with the exception of occasional meniscus cements. Further, the presence of evidences of living organisms may be an indication of influence from the biological aspects which can be confirmed by more detailed analyses.

The image quality of deep-learning image reconstruction of chest CT images on a mediastinal window setting

To assess the image quality of deep-learning image reconstruction (DLIR) of chest computed tomography (CT) images on a mediastinal window setting in comparison to an adaptive statistical iterative reconstruction (ASiR-V). Thirty-six patients were evaluated retrospectively. All patients underwent contrast-enhanced chest CT and thin-section images were reconstructed using filtered back projection (FBP); ASiR-V (60% and 100% blending setting); and DLIR (low, medium, and high settings). Image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were evaluated objectively. Two independent radiologists evaluated ASiR-V 60% and DLIR subjectively, in comparison with FBP, on a five-point scale in terms of noise, streak artefact, lymph nodes, small vessels, and overall image quality on a mediastinal window setting (width 400 HU, level 60 HU). In addition, image texture of ASiR-Vs (60% and 100%) and DLIR-high was analysed subjectively. Compared with ASiR-V 60%, DLIR-med and DLIR-high showed significantly less noise, higher SNR, and higher CNR (p<0.0001). DLIR-high and ASiR-V 100% were not significantly different regarding noise (p=0.2918) and CNR (p=0.0642). At a higher DLIR setting, noise was lower and SNR and CNR were higher (p<0.0001). DLIR-high showed the best subjective scores for noise, streak artefact, and overall image quality (p<0.0001). Compared with ASiR-V 60%, DLIR-med and DLIR-high scored worse in the assessment of small vessels (p<0.0001). The image texture of DLIR-high was significantly finer than that of ASIR-Vs (p<0.0001). DLIR-high improved the objective parameters and subjective image quality by reducing noise and streak artefacts and providing finer image texture.

Structure-dependent amplification for denoising and background correction in Fourier ptychographic microscopy.

Fourier Ptychographic Microscopy (FPM) allows high resolution imaging using iterative phase retrieval to recover an estimate of the complex object from a series of images captured under oblique illumination. FPM is particularly sensitive to noise and uncorrected background signals as it relies on combining information from brightfield and noisy darkfield (DF) images. In this article we consider the impact of different noise sources in FPM and show that inadequate removal of the DF background signal and associated noise are the predominant cause of artefacts in reconstructed images. We propose a simple solution to FPM background correction and denoising that outperforms existing methods in terms of image quality, speed and simplicity, whilst maintaining high spatial resolution and sharpness of the reconstructed image. Our method takes advantage of the data redundancy in real space within the acquired dataset to boost the signal-to-background ratio in the captured DF images, before optimally suppressing background signal. By incorporating differentially denoised images within the classic FPM iterative phase retrieval algorithm, we show that it is possible to achieve efficient removal of background artefacts without suppression of high frequency information. The method is tested using simulated data and experimental images of thin blood films, bone marrow and liver tissue sections. Our approach is non-parametric, requires no prior knowledge of the noise distribution and can be directly applied to other hardware platforms and reconstruction algorithms making it widely applicable in FPM.

Machine-Learning Image Recognition Enhances Rock Classification

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 196657, “Machine Learning for 3D Image Recognition To Determine Porosity and Lithology of Heterogeneous Carbonate Rock,” by Omar Al-Farisi, SPE, Hongtao Zhang, and Aikifa Raza, Khalifa University of Science and Technology, prepared for the 2019 SPE Reservoir Characterization and Simulation Conference and Exhibition, Abu Dhabi, 17-19 September. The paper has not been peer reviewed. Automated image-processing algorithms can improve the quality and speed in classifying the morphology of heterogeneous carbonate rock. Several commercial products have produced petrophysical properties from 2D images and, to a lesser extent, from 3D images. Images are mainly microcomputed tomography (μCT), optical images of thin sections, or magnetic resonance images (MRI). However, most successful work is from homogeneous and clastic rocks. In the complete paper, the authors have demonstrated a machine-learning-assisted image-recognition (MLIR) approach to determine the porosity and lithology of heterogeneous carbonate rock by analyzing 3D images from µCT and MRI. Introduction The authors’ literature review has revealed the pressing need to perform 3D image processing instead of 2D. Achieving this goal requires an interdisciplinary approach. This study deployed new analysis and verification approaches, including 3D micromodels (3DMM) with various micropore sizes and uses 3DMM as an image-processing calibration reference. Additionally, a new image-resolution enhancement for quality segmentation is developed. Porosity was determined mainly using two methods. The first is a standalone image processing, in which image-information extraction was successful. The second is MLIR. The difference between image processing and image analysis is important. If image processing enables extraction of meaningful data, then image analysis is the ability to interpret this data through numerical analysis.

DIGITAL ROCK ANALYSIS: AN ALTERNATIVE METHOD TO PREDICT PETROPHYSICAL PROPERTIES, CASE STUDY FROM MISHRIF FORMATION

The digital core analysis of petrophysical properties replace the use of conventional core analysis by reducing the required time for investigation. Also, the ability to capture pore geometries and fluid behavior at the pore-scale improves the understanding of complex reservoir structures. In this work, 53 samples of 2D thin section petrographic images were used for analyses from the core plugs taken from the Buzurgan oil field. Each sample was impregnated with blue-dyed epoxy, thin sectioned and then was stained for discrimination of carbonate minerals. Each thin section has been described in detail and illustrated by photomicrographs. The studied samples include a variety of rock types. Packstone is the most common rock type observed followed by grainstone and packstone – wackestone. Floatstone and dolostone are noted rarely in the studied interval. However, the samples of thin section images are processed and digitized, utilizing MATLAB programming and image analysis software. The entire workflow of digital core analysis from image segmentation to petrophysical rock properties determination was performed. A focused has been made on determining effective and total porosity, absolute permeability, and irreducible water saturation. Absolute permeability is estimated with the Kozeny-Carman permeability correlation model and Timur-Coates permeability correlation model. Irreducible water saturation simply is derived from total and effective porosity. Also, some pore void characteristics, such as area and perimeter, were calculated. The results of Digital 2D image analysis have been compared to laboratory core measurements to investigate the reliability and restrictions of the digital image interpretation techniques.