Low Magnification Images Research Articles

Interpretable laryngeal tumor grading of histopathological images via depth domain adaptive network with integration gradient CAM and priori experience-guided attention

Tumor grading and interpretability of laryngeal cancer is a key yet challenging task in the clinical diagnosis, mainly because of the commonly used low-magnification pathological images lack fine cellular structure information and accurate localization, the diagnosis results of pathologists are different from those of attentional convolutional network -based methods, and the gradient-weighted class activation mapping method cannot be optimized to create the best visualization map. To address this problem, we propose an end-to-end depth domain adaptive network (DDANet) with integration gradient CAM and priori experience-guided attention to improve the tumor grading performance and interpretability by introducing the pathologist's a priori experience in high-magnification into the depth model. Specifically, a novel priori experience-guided attention (PE-GA) method is developed to solve the traditional unsupervised attention optimization problem. Besides, a novel integration gradient CAM is proposed to mitigate overfitting, information redundancies and low sparsity of the Grad-CAM graphs generated by the PE-GA method. Furthermore, we establish a set of quantitative evaluation metric systems for model visual interpretation. Extensive experimental results show that compared with the state-of-the-art methods, the average grading accuracy is increased to 88.43% (↑4.04%), the effective interpretable rate is increased to 52.73% (↑11.45%). Additionally, it effectively reduces the difference between CV-based method and pathology in diagnosis results. Importantly, the visualized interpretive maps are closer to the region of interest of concern by pathologists, and our model outperforms pathologists with different levels of experience.

Statistical Determination of Atomic-Scale Characteristics of Au Nanocrystals Based on Correlative Multiscale Transmission Electron Microscopy

Abstract Atomic-scale characteristics of individual nanocrystals (NCs), such as the crystallographic orientation of their facets and the kind and density of crystal structure defects, play a tremendous role for the functionality and performance of the whole NC population. However, these features are usually quantified only for a small number of individual particles, and thus with limited statistical relevance. In the present work, we developed the multiscale approach available in transmission electron microscopy (TEM) further, and applied it to describe features of different types of Au NCs in a statistical and scale-bridging manner. This approach combines high-resolution TEM, which is capable of describing the characteristics of NCs on atomic scale, with a semi-automatic analysis of low-magnification high-angle annular dark-field scanning TEM images, which reveals the nanoscopic morphological attributes of NCs with good statistics. The results of these complementary techniques are combined and correlated. The potential of this multiscale approach is illustrated on two examples. In the first one, the habitus of Au NCs was classified and assigned to multiply twinned nanoparticles and nanoplates. These classes were quantified and related to different stacking fault densities. The second example demonstrates the statistical determination of crystallographic orientations and configurations of facets in Au nanorods.

LBSUN129 Adipose Tissue Analysis Toolkit (at-at): An Imagej Plugin For Automated Analysis Of Adipocyte Number, Size, And Pericellular Fibrosis In Histological Sections

Abstract Obesity-induced pathological expansion of white adipose tissue (WAT) involves adipocyte hypertrophy that occurs in tandem with the development of pericellular fibrosis. Other pathophysiological states, such as cancer and HIV, are also associated with fibrotic transformation of WAT depots. Aging, obesity, inactivity, and estrogen deficiency are also associated with an increase in bone marrow adiposity. Disease-associated changes in WAT morphology are typically assessed using histochemical analysis of WAT sections. However, manual analysis of large histological sections of WAT or bone marrow adipose tissue can be very time-consuming, while digital tools for the analysis of adipocyte size and pericellular fibrosis are limited. In response to this need, we have developed the Adipose Tissue Analysis Toolkit (AT-AT), a plugin for the open-source image analysis software ImageJ. AT-AT employs both local and global differentials in pixel intensity to threshold background apart from adipocyte-free tissue without user input. The AT-AT accurately captures adipocyte perimeter, maximum diameter, minimum diameter, and density per square millimeter of tissue in histological sections stained with common histological dyes, and performs automatic analysis of adipocyte cross-sectional area as well as pericellular fibrosis. AT-AT enables fully-automated analysis of batches of histological images using either default or user-defined adipocyte detection parameters including minimum circularity as well as minimum and maximum cross-sectional area. Likewise, users may set parameters for the thresholding algorithms, including the number of localized thresholds to be calculated in each image. AT-AT's thresholding techniques confer uniquely high resistances to both false-negative and false-positive detection events by comparison to other open-source adipocyte analysis software. They also enable superior measurement of adipocyte morphology, especially in the case of low-magnification images. AT-AT features several quality-of-life features, including automatic summarization of batch-collected data, the implementation of its thresholding algorithms as independent ImageJ commands, and an optional histogram builder. AT-AT can enhance the rigor and reproducibility of adipose tissue research. AT-AT will become available as a plugin for ImageJ in the fall of 2022. Presentation: Sunday, June 12, 2022 12:30 p.m. - 2:30 p.m.

Semantic segmentation of low magnification effusion cytology images: A semi-supervised approach

Cytopathologists examine microscopic images obtained at various magnifications to identify malignancy in effusions. They locate the malignant cell clusters at a low magnification and then zoom in to investigate cell-level features at a high magnification. This study predicts the malignancy at low magnification levels such as 4X and 10X in effusion cytology images to reduce scanning time. However, the most challenging problem is annotating the low magnification images, particularly the 4X images. This paper extends two semi-supervised learning (SSL) models, MixMatch and FixMatch, for semantic segmentation. The original FixMatch and MixMatch algorithms are designed for classification tasks. While performing image augmentation, the generated pseudo labels are spatially altered. We introduce reverse augmentation to compensate for the effect of the spatial alterations. The extended models are trained using labelled 10X and unlabelled 4X images. The average F-score of benign and malignant pixels on the predictions of 4X images is improved approximately by 9% for both Extended MixMatch and Extended FixMatch respectively compared with the baseline model. In the Extended MixMatch, 62% sub-regions of low magnification images are eliminated from scanning at a higher magnification, thereby saving scanning time.

Aedes Larva Detection Using Ensemble Learning to Prevent Dengue Endemic

Dengue endemicity has become regular in recent times across the world. The numbers of cases and deaths have been alarmingly increasing over the years. In addition to this, there are no direct medications or vaccines to treat this viral infection. Thus, monitoring and controlling the carriers of this virus which are the Aedes mosquitoes become specially demanding to combat the endemicity, as killing all the mosquitoes regardless of their species would destroy ecosystems. The current approach requires collecting a larva sample from the hatching sites and, then, an expert entomologist manually examining it using a microscope in the laboratory to identify the Aedes vector. This is time-consuming, labor-intensive, subjective, and impractical. Several automated Aedes larvae detection systems have been proposed previously, but failed to achieve sufficient accuracy and reliability. We propose an automated system utilizing ensemble learning, which detects Aedes larvae effectively from a low-magnification image with an accuracy of over 99%. The proposed system outperformed all the previous methods with respect to accuracy. The practical usability of the system is also demonstrated.

New application of endocytoscope for histopathological diagnosis of colorectal lesions.

BACKGROUNDThe endocytoscope with ultra-high magnification (x 520) allows us to observe the cellular structure of the colon epithelium during colonoscopy, known as virtual histopathology. We hypothesized that the endocytoscope could directly observe colorectal histopathological specimens and store them as endocyto-pathological images by the endoscopists without a microscope, potentially saving the burden on histopathologists.AIMTo assess the feasibility of endocyto-pathological images taken by an endoscopist as adequate materials for histopathological diagnosis.METHODSThree gastrointestinal pathologists were invited and asked to diagnose 40 cases of endocyto-pathological images of colorectal specimens. Each case contained seven endocyto-pathological images taken by an endoscopist, consisting of one loupe image, three low-magnification images, and three ultra-high magnification images. The participants chose hyperplastic polyp or low-grade adenoma for 20 cases of endocyto-pathological images (10 hyperplastic polyps, and 10 Low-grade adenomas in conventional histopathology) in study 1 and high-grade adenoma/ shallow invasive cancer or deep invasive cancer for 20 cases [10 tumor in situ/T1a and 10 T1b] in study 2. We investigated the agreement between the histopathological diagnosis using the endocyto-pathological images and conventional histopathological diagnosis. RESULTSAgreement between the endocyto-pathological and conventional histopathological diagnosis by the three gastrointestinal pathologists was 100% (95%CI: 94.0%–100%) in studies 1 and 2. The interobserver agreement among the three gastrointestinal pathologists was 100%, and the κ coefficient was 1.00 in both studies. CONCLUSIONEndocyto-pathological images were adequate and reliable materials for histopathological diagnosis.

Low-magnification narrow-band imaging for small gastric neoplasm detection on screening endoscopy.

Background and AimsMicrosurface patterns of the gastric mucosa can be observed using magnifying narrow-band imaging (M-NBI). However, the efficacy of M-NBI at low-magnification (LM-NBI) screening for detecting small gastric neoplasms is unclear.MethodsThis prospective study was conducted at a single institution. LM-NBI, defined as minimal magnification that could reveal the microsurface pattern of the gastric mucosa, was performed after routine white-light imaging (WLI) observation of the stomach. Depending on the phase in which the neoplastic lesions were initially found, they were divided into the WLI group and the LM-NBI group, and the characteristics of these neoplastic lesions were investigated accordingly.ResultsSixty-five epithelial lesions (adenomas or noninvasive carcinomas) of 20 mm or less in diameter were identified in this study. Sixteen lesions were detected only with LM-NBI. Smaller lesions were detected using LM-NBI (P = .01). WLI took about 160 to 260 seconds, while LM-NBI required about 70 to 80 seconds. All lesions in the LM-NBI group had a background of map-like redness (n = 5) or atrophic/metaplastic mucosa (n = 11).ConclusionsLM-NBI was able to detect lesions overlooked by WLI, especially those in areas of map-like redness or atrophic/metaplastic mucosa of the stomach. Approximately one-quarter of newly diagnosed neoplasms were retrieved on routine examination during an extra 1.5 minutes.

High-Magnification Super-Resolution Reconstruction of Image with Multi-Task Learning

Single-image super-resolution technology has made great progress with the development of the convolutional neural network, but most of the current super-resolution methods do not attempt high-magnification image super-resolution reconstruction; only reconstruction with ×2, ×3, ×4 magnification is carried out for low-magnification down-sampled images without serious degradation. Based on this, this paper proposed a single-image high-magnification super-resolution method, which extends the scale factor of image super-resolution to high magnification. By introducing the idea of multi-task learning, the process of the high-magnification image super-resolution process is decomposed into different super-resolution tasks. Different tasks are trained with different data, and network models for different tasks can be obtained. Through the cascade reconstruction of different task network models, a low-resolution image accumulates reconstruction advantages layer by layer, and we obtain the final high-magnification super-resolution reconstruction results. The proposed method shows better performance in quantitative and qualitative comparison on the benchmark dataset than other super-resolution methods.

Generative-adversarial-network-based dimensional measurement of optical waveguides.

We propose a high-throughput and precise waveguide-dimensional-measurement method consisting of a generative adversarial network (GAN) and curve-fitting-based dimensional calculator using sidewall functions. The GAN can learn the differences between low-magnification (LM) and high-magnification (HM) optical microscope images taken with different objective lenses at different magnifications over the same area. The LM and HM images of the waveguides are captured using an optical microscope at magnifications of 500× and 2000×, respectively. We obtained a standard deviation of the waveguide widths of approximately 0.8 pixels (∼ 42 nm), and confirmed precise width measurement using super-resolution images at the same imaging throughput as with an LM microscope.

Efficient and Accurate Synapse Detection With Selective Structured Illumination Microscopy on the Putative Regions of Interest of Ultrathin Serial Sections.

Critical determinants of synaptic functions include subcellular locations, input sources, and specific molecular characteristics. However, there is not yet a reliable and efficient method that can detect synapses. Electron microscopy is a gold-standard method to detect synapses due to its exceedingly high spatial resolution. However, it requires laborious and time-consuming sample preparation and lengthy imaging time with limited labeling methods. Recent advances in various fluorescence microscopy methods have highlighted fluorescence microscopy as a substitute for electron microscopy in reliable synapse detection in a large volume of neural circuits. In particular, array tomography has been verified as a useful tool for neural circuit reconstruction. To further improve array tomography, we developed a novel imaging method, called “structured illumination microscopy on the putative region of interest on ultrathin sections”, which enables efficient and accurate detection of synapses-of-interest. Briefly, based on low-magnification conventional fluorescence microscopy images, synapse candidacy was determined. Subsequently, the coordinates of the regions with candidate synapses were imaged using super-resolution structured illumination microscopy. Using this system, synapses from the high-order thalamic nucleus, the posterior medial nucleus in the barrel cortex were rapidly and accurately imaged.

Improved efficiency in automated acquisition of ultra-high-resolution electron holograms using automated target detection.

An automated hologram acquisition system for big-data analysis and for improving the statistical precision of phase analysis has been upgraded with automated particle detection technology. The coordinates of objects in low-magnification images are automatically detected using zero-mean normalized cross-correlation with preselected reference images. In contrast with the conventional scanning acquisitions from the whole area of a microgrid and/or a thin specimen, the new method allows efficient data collections only from the desired fields of view including the particles. The acquisition time of the cubic/triangular nanoparticles that were observed was shortened by about one-fifty eighth that of the conventional scanning acquisition method because of efficient data collections. The developed technology can improve statistical precision in electron holography with shorter acquisition time and is applicable to the analysis of electromagnetic fields for various kinds of nanoparticles.

A simple preparation method for CLEM using pre-embedding immunohistochemistry with a novel fluorescent probe and stable embedding resin.

Correlative light and electron microscopy (CLEM) is an excellent approach for examining the cellular localization of biomolecules. Here, we developed a simple method for CLEM by combining pre-embedding immunohistochemistry with a novel fluorescent probe, namely Fluolid NS Orange, and an embedding resin called 'Durcupan™'. Specimens were embedded in Durcupan™ or LR White after immunolabeling and post-fixation using glutaraldehyde and osmium tetroxide. Next, ultrathin sections were prepared on a finder grid with navigation markers. The section of the specimen embedded in Durcupan™ was found to be more stable against electron beam irradiation than specimens embedded in LR White. A fluorescence light microscopy image and a transmission electron microscopy (TEM) image, at wide-field, and low magnification, were independently obtained with the same ultrathin section. Using the three corners between finder grid bars as landmarks, fluorescence light microscopy images were superimposed with wide-field, low-magnification TEM images to identify the region of interest, which was subsequently enlarged to ascertain cellular structures localized beneath fluorescent signals. However, the enlarged TEM images appeared blurred, and fluorescence signals had a hazy appearance. To resolve this, the enlarged TEM images were replaced by high-resolution TEM images focused directly on the region of interest, thereby facilitating the collection of high-resolution CLEM images. The simple sample processing method for CLEM using osmium-resistant Fluolid NS Orange and electron beam damage-resistant Durcupan™ allowed the determination of the precise localization of fluorescence signals at subcellular levels.

Self-assembly of superstructures at all scales

Summary Controllable assembly of molecular and nanoscale building blocks into uniform superstructures up to bulk dimensions remains a key challenge in the next phase of nanotechnology development. Here, we report the self-assembly of superstructures at all scales by taking advantage of the partial miscibility of water and 1-butanol to generate transient aqueous emulsion droplets that can encapsulate the target materials and introduce them into template holes. Further diffusion of water into 1-butanol depletes the emulsion droplets, assembling the building blocks into one well-defined superstructure in each hole. Superstructuring of various types and shapes of nanoparticles, biomolecules, and inorganic compounds could be achieved without the need for surfactants and chemical modifications. The versatility, scalability, low-cost, and accurate positioning are crucial advantages for the future development of advanced precision manufacturing.

Practical fluorescence reconstruction microscopy for large samples and low-magnification imaging.

Fluorescence reconstruction microscopy (FRM) describes a class of techniques where transmitted light images are passed into a convolutional neural network that then outputs predicted epifluorescence images. This approach enables many benefits including reduced phototoxicity, freeing up of fluorescence channels, simplified sample preparation, and the ability to re-process legacy data for new insights. However, FRM can be complex to implement, and current FRM benchmarks are abstractions that are difficult to relate to how valuable or trustworthy a reconstruction is. Here, we relate the conventional benchmarks and demonstrations to practical and familiar cell biology analyses to demonstrate that FRM should be judged in context. We further demonstrate that it performs remarkably well even with lower-magnification microscopy data, as are often collected in screening and high content imaging. Specifically, we present promising results for nuclei, cell-cell junctions, and fine feature reconstruction; provide data-driven experimental design guidelines; and provide researcher-friendly code, complete sample data, and a researcher manual to enable more widespread adoption of FRM.

Understanding the mechanical behaviour of 718 and 625 + nickel based super-alloys under cathodic polarization

The effect of hydrogen on the mechanical behaviour of UNS N07718 (alloy 718) and UNS N07716 (alloy 625 + ) nickel based super-alloys was studied. Atomic hydrogen was introduced into the metal during Slow Strain Rate Tests (SSRTs) via an electrochemical technique, following the method described in [1], this is a standard used in the oil and gas industry to assess the performance of precipitation hardened nickel based alloys towards hydrogen embrittlement (HE). Different heat treatment conditions, used to produce different microstructures, were selected to highlight the effect of specific microstructural phases/ features on the HE susceptibility, including: a Solution Annealed (SA) microstructure, a microstructure with a high volume fraction of delta phase (δ), a microstructure containing low volume fraction of gamma prime (γ’) and gamma double-prime (γ’’), the As Received material (AR) and a Coarse Grain Size (CGZ) microstructure. For each microstructural condition, the mechanical performance under electrochemical hydrogenating conditions was compared against that obtained in an inert (glycerol) environment to evaluate and quantify the susceptibility to HE. In addition, an analysis of fracture surfaces was carried out to assess the preferential mode of fracture. Results revealed that in low magnification images the Hydrogen charged samples showed smooth featureless areas on the fracture surface. HE susceptibility was seen to increase as the volume fraction of the precipitates increased whilst grain size had negligible effect. 625 + displayed higher susceptibility to HE compared to 718.

Influence of Vertical and Radial Distances of Nano-Resistors in SSI-LEDs on Light Intensity Distributions

Broad-band white light emission phenomena from a novel type of solid-state incandescent light emitting device (SSI-LED) was reported by Kuo's group [1-4]. The device is made from the dielectric breakdown of a metal-oxide-semiconductor (MOS) capacitor with a high-k gate dielectric deposited over a p-type Si wafer [1-4]. The device emits light from the thermal excitation of the nano-sized conductive paths in the high-k stack upon the passage of a large current. This blackbody radiation phenomenon is similar to that of the incandescent light bulb [4]. In this paper, the light intensity distributions from nano-resistors with various vertical and radial distances are investigated by simulation.Equation 1, also known as Stefan-Boltzmann law [5, 6], quantifies the total energy radiated per unit surface area of a black body across all wavelengths per unit time j* that is directly proportional to the fourth power of the black body's thermodynamic temperature T, i.e., j* = eσT4 (1)where e is the emissivity of the object (e = 1 for perfect black body) and σ is the Stefan–Boltzmann constant. Therefore, the total power P radiated from an object of surface area A can be given by the following equation P = Aj* = eσT4 (2)Assuming that the nano-resistors formed in the high-k stack behave as point light sources, we can calculate the intensity at a given vertical distance from the nano-resistors. The following equation, i.e., the intensity of light emitted from a point source I is inversely proportional to the square of the distance r, is used in the calculation I = P/ 4πr2 (3)The resultant intensity due to all the nano-resistors at any given vertical distance can then be simply obtained by scalar addition of intensity from each of the individual nano-resistors. In order to simulate the light intensity distribution from the SSI-LED, authors made some assumptions based on experimental observations. For instance, nano-resistors with cylindrical geometry are assumed to be formed and distributed randomly across the high-k gate stack [1]. Further it is assumed that all nano-resistors in a SSI-LED are of the same diameter, i.e., 100 nm. In actual, nano-resistors formed on the same device are of different sizes, i.e., the number decreases with the increase of the diameter size [4]. Previously, it has been reported that light emitted from the SSI-LED is similar to that from the conductive paths at the temperature of 4,000 K [7]. The same temperature of 4,000 K was used to calculate the total radiated power P in equation 1. Although it was reported earlier that the temperature across the cross-section of a nano-resistor could vary with its diameter size [7], here we assume it to be constant. Moreover, the effect of thin film ITO gate on intensity of emitted light is neglected for simplifying current calculations. The Beer-Lambert Law could be used to account for the transmittance of light across ITO thin film to further improve the intensity distribution estimation.Figure 1(a) shows the light intensity distributions of two 100 nm diameter nano-resistor sources observed at varying distance above the nano-resistor/ITO gate interface. Figure 1(b) shows the low and high magnification images of a SSI-LED with the ZrHfO gate dielectric. It can be clearly seen that at the low magnification, the whole SSI-LED appears as a single light dot, i.e., individual nano-resistors are unresolvable [1]. At the high magnification, which is similar to the short distance case, peaks of the two light dots are observable. This study will be extended in 3-D to better understand the light distribution. The result will be discussed in detail. Moreover, the effect of the radial separation distance of the nano-resistors on the light intensity distribution will also be presented and discussed.[1] Y. Kuo and C.-C. Lin, Appl. Phys. Lett., 102, 031117 (2013).[2] Y. Kuo and C.-C. Lin, ECS Solid-State Lett. , 2, Q59 (2013).[3] Y. Kuo and C.-C. Lin, Solid-State Electron., 89, 120 (2013).[4] C.-C. Lin and Y. Kuo, Appl. Phys. Lett., 106, 121107 (2015).[5] Stefan J, Sitzber. Math.-Nat.wiss. Cl. kaiserlichen Akad. Wiss., 79, 391 (1879).[6] Boltzmann L, Ann. Phys. Chem., 22, 291 (1884)[7] A. Shukla and Y. Kuo, ECS Trans., Submitted, April 2020. Figure 1

Boosting CO2 Conversion with Terminal Alkynes by Molecular Architecture of Graphene Oxide-Supported Ag Nanoparticles

Summary Carboxylation of terminal alkynes with CO2 via direct C–H bond activation is a highly appealing way to capture and convert CO2 but faces great challenges and difficulty. Here, we report an efficient catalyst that is prepared simply through introduction of p-tert-butylaniline into graphite oxide powders followed by localized growth of Ag nanoparticles (NPs). The bulky nature of p-tert-butylaniline facilitates the powder exfoliation, giving rise to mass production of uniform modified graphene oxide (tert-GO) nanosheets of 4 nm in thickness. Notably, Ag/tert-GO shows excellent catalytic activity for converting alkynes regardless of containing electron-donating or electron-withdrawing groups under mild reaction conditions. The synergy of the amide linkages on tert-GO nanosheets and the supported small-sized Ag NPs is recognized as the vital role in promoting adsorption of CO2 and subsequent conversion.

Reproducibility of AJCC Criteria for Classifying Deeply Invasive Colon Cancers Is Suboptimal for Consistent Cancer Staging.

The eighth edition of the American Joint Committee on Cancer (AJCC) Staging Manual attempts to address ambiguity in the pT category assignment for colon cancer from prior editions. Despite modifications, the distinction between the pT3 and pT4a categories continues to be a source of diagnostic confusion. In this study, we assessed interobserver agreement among pathologists from different institutions in the application of AJCC eighth edition criteria for categorizing deeply invasive colonic adenocarcinomas. We identified morphologic patterns that produce diagnostic confusion. We assessed 47 colon cancers that closely approached the serosal surface. Six pathologists with interest in gastrointestinal pathology and 4 focused in other subspecialties classified each case as pT3 or pT4a, based on examination of low-magnification and high-magnification images of the most deeply invasive area. Interobserver agreement was assessed using Fleiss' κ. Cases displayed 3 morphologic patterns at the advancing tumor edge, namely, (1) continuous invasion through an inflammatory focus, (2) pushing border, and (3) infiltrative glands and cell clusters with serosal reaction. Gastrointestinal pathologists achieved slight (κ=0.21) or moderate (κ=0.46) and (κ=0.51) agreement in each category, whereas agreement among nongastrointestinal pathologist was fair (0.31) and (0.39), or moderate (0.57) for each category, respectively. In 10 (21%) cases, the distinction between pT3 and pT4a would have changed the overall clinical stage. We conclude that histologic criteria for serosal penetration is a persistent source of diagnostic ambiguity for gastrointestinal and general pathologists in the pT categorization of colon cancers. Clarification of these criteria will help ensure uniform reporting of pathologic and clinical stage.

3D Microscopy Keeps Getting Faster, Smarter, Leaner

3D Microscopy Keeps Getting Faster, Smarter, Leaner

CoMHisP: A Novel Feature Extractor for Histopathological Image Classification Based on Fuzzy SVM With Within-Class Relative Density

Machine learning (ML) has emerged as a powerful tool for pattern recognition. Traditional ML algorithms have limited ability to reveal the most sophisticated features of cancer histopathological images, but their robustness and fault tolerance can be enhanced by using fuzzy modeling to capture the uncertainty in image data. Therefore, this article proposes a novel CoMHisP framework based on a fuzzy support vector machine with within-class density information (FSVM-WD). It utilizes a novel feature extraction technique by optimizing the block size to extract image micropatterns and computing center of mass (CoM) for each pixel to extract feature vectors. The performance of the proposed framework is evaluated using a CMTHis dataset comprising histopathological images of canine mammary tumor (CMT), a prevalent neoplastic disease in female dogs, and an established model for human breast cancer. Data analysis reveals that stain normalization and magnification influence the performance of the CoMHisP framework, with the best results achieved at lower magnifications after stain normalization. The proposed framework achieves a classification accuracy of 97.25% ( $\pm$ 1.80%) using a FSVM-WD classifier, outperforming both traditional ML and deep FE-VGGNET16-based feature descriptors. To the best of our knowledge, this is the first time a CoM-based feature descriptor has been proposed for histopathological image analysis of CMTs and its performance was evaluated using a fuzzy SVM-based classifier. The proposed method performs well with datasets of limited size and low-magnification images and, therefore, has the potential to provide rapid and accurate diagnosis in low-cost clinical settings.