Search

Compact binary representations of histopa-thology images using hashing methods provide efficient approximate nearest neighbor search for direct visual query in large-scale databases. They can be utilized to measure the probability of the abnormality of the query image based on the retrieved similar cases, thereby providing support for medical diagnosis. They also allow for efficient managing of large-scale image databases because of a low storage requirement. However, the effectiveness of binary representations heavily relies on the visual descriptors that represent the semantic information in the histopathological images. Traditional approaches with hand-crafted visual descriptors might fail due to significant variations in image appearance. Recently, deep learning architectures provide promising solutions to address this problem using effective semantic representations. In this paper, we propose a deep convolutional hashing method that can be trained “point-wise” to simultaneously learn both semantic and binary representations of histopathological images. Specifically, we propose a convolutional neural network that introduces a latent binary encoding (LBE) layer for low-dimensional feature embedding to learn binary codes. We design a joint optimization objective function that encourages the network to learn discriminative representations from the label information, and reduce the gap between the real-valued low-dimensional embedded features and desired binary values. The binary encoding for new images can be obtained by forward propagating through the network and quantizing the output of the LBE layer. Experimental results on a large-scale histopathological image dataset demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Highlights • Grain boundary detection in microscopic images is a challenge urged to be solved. • We propose a method to detect grain boundary considering 3D information. • We introduce a local processing method to accelerate the image processing. • FastFine-Cut shows the highest performance in polycrystalline iron experiments. Abstract The inner structure of a material is called its microstructure. It stores the genesis of a material and determines all the physical and chemical properties. However, the microstructure is highly complex and numerous image defects such as vague or missing boundaries formed during sample preparation, which makes it difficult to extract the grain boundaries precisely. In this work, we address the task of grain boundary detection in microscopic image processing and develop a graph-cut based method called Fast-FineCut to solve the problem. Our algorithm makes two key contributions: (1) An improved approach that incorporates 3D information between slices as domain knowledge, which can detect the boundaries precisely, even for the vague and missing boundaries. (2) A local processing method based on overlap-tile strategy, which can not only solve the "chain scission" problem at the edge of images, but also economize on the consumption of computing resources. We conduct experiments on a stack of 296 slices of microscopic images of polycrystalline iron (1600 × 2800) and compare the performance against several state-of-the-art boundary detection methods. We conclude that Fast-FineCut can detect boundaries effectively and efficiently. [ABSTRACT FROM AUTHOR]

Mathematical modelling to compute ground truth from 3D images is an area of research that can strongly benefit from machine learning methods. Deep neural networks (DNNs) are state-of-the-art methods design for solving these kinds of difficulties. Convolutional neural networks (CNNs), as one class of DNNs, can overcome special requirements of quantitative analysis especially when image segmentation is needed. This article presents a system that uses a cascade of CNNs with symmetric blocks of layers in chain, dedicated to 3D image segmentation from microscopic images of 3D nuclei. The system is designed through eight experiments that differ in following aspects: number of training slices and 3D samples for training, usage of pre-trained CNNs and number of slices and 3D samples for validation. CNNs parameters are optimized using linear, brute force, and random combinatorics, followed by voter and median operations. Data augmentation techniques such as reflection, translation and rotation are used in order to produce sufficient training set for CNNs. Optimal CNN parameters are reached by defining 11 standard and two proposed metrics. Finally, benchmarking demonstrates that CNNs improve segmentation accuracy, reliability and increased annotation accuracy, confirming the relevance of CNNs to generate high-throughput mathematical ground truth 3D images.

At present, detection methods for rice microbial indicators are usually based on microbial culture or sensory detection methods, which are time-consuming or require expertise and thus cannot meet the needs of on-site rice testing when the rice is taken out of storage or traded. In order to develop a fast and non-destructive method for detecting rice mildew, in this paper, micro-computer vision technology is used to collect images of mildewed rice samples from 9 image locations. Then, a YOLO-V5 convolutional neural network model is used to detect moldy areas of rice, and the mold coverage area is estimated. The relationship between the moldy areas and the total number of bacterial colonies in the image is obtained. The results show that the precision and the recall of the established YOLO-v5 model in identifying the mildewed areas of rice in the validation set were 82.1% and 86.5%, respectively. Based on the mean mildewed area identified by the YOLO-v5 model, the precision and recall for light mold detection were 100% and 95.3%, respectively. The proposed method based on micro-computer vision and the YOLO convolutional neural network can be applied to the rapid detection of mildew in rice taken out of storage or traded., Competing Interests: All the authors declare that they have no conflict of interest., (© 2023 The Authors. Food Science & Nutrition published by Wiley Periodicals LLC.)

Background: Sperm tail morphology and motility have been demonstrated to be important factors in determining sperm quality for in vitro fertilization. However, many existing computer-aided sperm analysis systems leave the sperm tail out of the analysis, as detecting a few tail pixels is challenging. Moreover, some publicly available datasets for classifying morphological defects contain images limited only to the sperm head. This study focuses on the segmentation of full sperm, which consists of the head and tail parts, and appear alone and in groups., Methods: We re-purpose the Feature Pyramid Network to ensemble an input image with multiple masks from state-of-the-art segmentation algorithms using a scale-specific cross-attention module. We normalize homogeneous backgrounds for improved training. The low field depth of microscopes blurs the images, easily confusing human raters in discerning minuscule sperm from large backgrounds. We thus propose evaluation protocols for scoring segmentation models trained on imbalanced data and noisy ground truth., Results: The neural ensembling of noisy segmentation masks outperforms all single, state-of-the-art segmentation algorithms in full sperm segmentation. Human raters agree more on the head than tail masks. The algorithms also segment the head better than the tail., Conclusions: The extensive evaluation of state-of-the-art segmentation algorithms shows that full sperm segmentation is challenging. We release the SegSperm dataset of images from Intracytoplasmic Sperm Injection procedures to spur further progress on full sperm segmentation with noisy and imbalanced ground truth. The dataset is publicly available at https://doi.org/10.34808/6wm7-1159., Competing Interests: Declaration of competing interest None Declared, (Copyright © 2023 The Author(s). Published by Elsevier Ltd.. All rights reserved.)

Background: A move towards integrated curriculum questions the desirability and application of image sensitive disciplines like histology. It calls for redefining the role of drawing in learning. Objective: To determine the effect of drawing microscopic images on retention of knowledge of histology. Study design, settings and duration: The quasi experimental study was conducted at the histology laboratory of Anatomy department, Shifa College of Medicine (SCM), Shifa Tameer-e-Millat University, lasting from January 2017 to September 2017. Subjects and Methods: The second year MBBS students were selected by purposive sampling technique. The study was conducted during an eight week duration module of endocrine and reproductive system. The class of 100 students was divided into group A and group B, comprising of 50 students each. During the first four weeks, students in group A were required to draw the histological image. The students in group B did not draw the microscopic image. During the next four weeks, the two groups were flipped over by a cross-over design. The group A became the no drawing group and group B became the drawing group. The MCQ test was taken at the end of four and eight weeks respectively. Results: The students scored better when they drew the images compared to when they did not draw. The results are statistically significant (p <0.05). Conclusion: Drawing is an effective teaching strategy for knowledge retention. [ABSTRACT FROM AUTHOR]

When trying to identify a printed forged document, examining digital evidence can prove to be a challenge. In this study, microscopic images are used for printed source identification due to their high magnification properties resulting in detailed texture and structure information. Prior research implemented a scanner as a digitizing technique to resolve very fine printed identification, but this technique provided limited information on the resolution and magnification of the sample. In contrast, the performance of microscopy techniques can retrieve the shape and surface texture of a printed document with differing micro structures among printer sources. To explore the relationship between source printers and images obtained by the microscope, the proposed approach utilizes image processing techniques and data exploration methods to calculate many important statistical features, including: Local Binary Pattern (LBP), Gray Level Co-occurrence Matrix (GLCM), Discrete Wavelet Transform (DWT), Spatial filters, the Wiener filter, the Gabor filter, Haralick, and SFTA features. Among the different set of features, the LBP approach achieves the highest identification rate and is significantly superior to other methods. As a result, the proposed technique using microscopic images achieves a high classification accuracy rate, which shows promising applications for real world digital forensics research. [ABSTRACT FROM AUTHOR]

• A new dataset formed with bone marrow microscopic images from two resources. • MultiPathGAN to translate multi-domain images for data augmentation. • Light-weight MobileViTv2 to diagnose six hematologic malignancies. • Achieving high classification performance compared with state-of-the-art methods. Hematologic malignancies, including the associated multiple subtypes, are critically threatening to human health. The timely detection of malignancies is crucial for their effective treatment. In this regard, the examination of bone marrow smears constitutes a crucial step. Nonetheless, the conventional approach to cell identification and enumeration is laborious and time-intensive. Therefore, the present study aimed to develop a method for the efficient diagnosis of these malignancies directly from bone marrow microscopic images. A deep learning-based framework was developed to facilitate the diagnosis of common hematologic malignancies. First, a total of 2033 microscopic images of bone marrow analysis, including the images for 6 disease types and 1 healthy control, were collected from two Chinese medical websites. Next, the collected images were classified into the training, validation, and test datasets in the ratio of 7:1:2. Subsequently, a method of stain normalization to multi-domains (stain domain augmentation) based on the MultiPathGAN model was developed to equalize the stain styles and expand the image datasets. Afterward, a lightweight hybrid model named MobileViTv2, which integrates the strengths of both CNNs and ViTs, was developed for disease classification. The resulting model was trained and utilized to diagnose patients based on multiple microscopic images of their bone marrow smears, obtained from a cohort of 61 individuals. MobileViTv2 exhibited an average accuracy of 94.28% when applied to the test set, with multiple myeloma, acute lymphocytic leukemia, and lymphoma revealed as the three diseases diagnosed with the highest accuracy values of 98%, 96%, and 96%, respectively. Regarding patient-level prediction, the average accuracy of MobileViTv2 was 96.72%. This model outperformed both CNN and ViT models in terms of accuracy, despite utilizing only 9.8 million parameters. When applied to two public datasets, MobileViTv2 exhibited accuracy values of 99.75% and 99.72%, respectively, and outperformed previous methods. The proposed framework could be applied directly to bone marrow microscopic images with different stain styles to efficiently establish the diagnosis of common hematologic malignancies. [ABSTRACT FROM AUTHOR]

Stomatal traits of leaves are critical for regulating the exchange of gases between plant tissues and the atmosphere, and thus play a crucial role in the physiological activities of plants. The hypothesis of this study is that distinct stomatal features among different species grown in diverse habitats can serve as a potential marker for species identification. Leaf samples were collected from the mangrove forests of Sundarbans and the freshwater swamp forests of Ratargul in Bangladesh. In total, we examined 11 species from eight different families. We used deep convolutional neural network (DCNN) to automatically identify tree species from microscopic stomatal imprints, as there is currently no established protocol for this task. For model training, 80% (866 images) of the data was used for training the models. Our study observed significant variations in stomatal attributes such as length, width, and density among different species, families, and habitats. These variations could help in accurate species identification by machine learning approaches used in the present study. An empirical comparison was conducted among EfficientNetV2, Xception, VGG16, VGG19, MobileNetV2, ResNet50V2, Resnet152, DenseNet201, and NasNetLarge. We propose a novel approach called the "Normalized Leverage Factor" that utilizes accuracy, precision, recall, and f1-score to select the optimal model. This approach eliminates the non-uniformity of the scores. Although MobileNetV2 achieved an accuracy of 99.06%, our findings indicate that EfficientNetV2 is the optimal model for species identification. This is due to its higher normalized leverage factor (1.92) compared to MobileNetV2 (1.88). The findings demonstrate that plants of diverse habitats show a unique footprint of stomata that offers an innovative method of species identification using DCNN. The study would help to develop a stomatal image-based user interface to identify species even without expert taxonomic knowledge and could be particularly useful in fields such as pharmacology, conservation biology, forestry, and environmental science. [Display omitted] • Stomata bears a unique footprint for the species identifications. • Deep learning techniques are used for automated species identifications. • We studied 11 species under eight families from mangrove and freshwater swamp forest. • A new approach formulated to evaluate the performance weight of models. • EfficientNetV2 outperforms the other models while showing no overfitting or performance degradation. [ABSTRACT FROM AUTHOR]

This paper describes the archaeological sites Mediana and Naisus during Late Antiquity. Microscopic images of bricks, and the results and analysis of XRF bricks from these archaeological sites are shown. Based on the results, it can be concluded that a similar brick exterior, and approximately the same chemical composition. One reason is, most likely, a similar chemical composition of the soil, because the archaeological sites are geographically close to each other. Another reason could be the same way bricks were producted, and that the same fuel was used in the kilns.

Background: Differential counting of white blood cells (WBCs or leukocytes) is a common task to diagnose many diseases such as leukemia, and infections. An accurate process for recognizing leukocytes is to evaluate a blood smear under a microscope by an expert. Since, this procedure is manual, time-consuming and tedious, making the procedure automatic would overcome these problems. In an automated CAD (Computer-Aided-Design) system for this purpose, a crucial module is leukocytes recognition. In this paper, we are looking for the best features in order to recognize five types of leukocytes (Monocyte, Lymphocyte, Neutrophil, Eosinophil and Basophil) from microscopic images of blood smear in an automated cell counting system. Methods: In this work, we focus on the texture features and seven categories: GLCM features, Haralick features, Spectral texture features, Waveletbased features, Gabor-based features, CoALBP and RICLBP are analyzed to find the best features for leukocytes detection. The best features of each category are selected using stepwise regression and finally three well-known classifiers called K-NN, LDA and NB are utilized for classification. Results: The proposed system is tested on a self-provided dataset composed of 200 cell images. In our experiments, to evaluate the process, the accuracy of each leukocyte type and the mean accuracy are computed. RICLBP features achieved the best mean accuracy (85.53%) for LDA classifier. Conclusions: In our experiments, although the maximum mean accuracy (85.53%) went with RICLBP features, but the accuracies of all five leukocyte types weren't maximized for RICLBP features. This result directs us to design and develop a system based on multiple features and multiple classifiers to maximize the accuracies even for each individual cell type in our future work. [ABSTRACT FROM AUTHOR]

The research is done to provide appropriate detection technique for skin tumor detection. The work is done by using the image processing toolbox of MATLAB. Skin tumors are unwanted skin growth with different causes and varying extent of malignant cells. It is a syndrome in which skin cells mislay the ability to divide and grow normally. Early detection of tumor is the most important factor affecting the endurance of a patient. Studying the pattern of the skin cells is the fundamental problem in medical image analysis. The study of skin tumor has been of great interest to the researchers. DIP (Digital Image Processing) allows the use of much more complex algorithms for image processing, and hence, can offer both more sophisticated performance at simple task, and the implementation of methods which would be impossibly by analog means. It allows much wider range of algorithms to be applied to the input data and can avoid problems such as build up of noise and signal distortion during processing. The study shows that few works has been done on cellular scale for the images of skin. This research allows few checks for the early detection of skin tumor using microscopic images after testing and observing various algorithms. After analytical evaluation the result has been observed that the proposed checks are time efficient techniques and appropriate for the tumor detection. The algorithm applied provides promising results in lesser time with accuracy. The GUI (Graphical User Interface) that is generated for the algorithm makes the system user friendly. [ABSTRACT FROM AUTHOR]

CAS (Cell Annotation Software) is a novel tool for analysis of microscopic images and selection of the cell soma or nucleus, depending on the research objectives in medicine, biology, bioinformatics, etc. It replaces time-consuming and tiresome manual analysis of single images not only with automatic methods for object segmentation based on the Statistical Dominance Algorithm, but also semi-automatic tools for object selection within a marked region of interest. For each image, a broad set of object parameters is computed, including shape features and optical and topographic characteristics, thus giving additional insight into data. Our solution for cell detection and analysis has been verified by microscopic data and its application in the annotation of the lateral geniculate nucleus has been examined in a case study. [ABSTRACT FROM AUTHOR]

Thermally treating wood improves its dimensional stability and duRability. The chemical changes brought about by a heat treatment also affect the mechanical properties of wood. Consequently, a heat treatment also influences how a wood surface responds to machining. This study examined the impact of heat treatments at 200 °C between 1 h and 6 h on the subsequent surface quality of planed beech wood (Fagus sylvatica L.). The new approach was that surface quality was assessed by following a tested method from previous research regarding the measuring and evaluation recommendations meant to reduce the biasing effect of wood anatomy, Also, a large number of roughness paRameters were used for interpretation of the combined effect of processing and wood anatomy after filtering the data with a robust filter. Among those, Rk is the paRameter that is least biased by wood anatomy and that best expresses the effect of processing alone. Electron microgRaphs were taken to visually assess the resultant surfaces. The results showed a gRadual increase in processing roughness, as distinctively measured by Rk, which increased with longer duRations of the treatment. Vessel cavities were deeper than those caused by processing and that influenced, among other paRameters, Ra, which is most commonly used in liteRature to assess surface quality. The Ray tissue, especially, exhibited both greater pull-out of fibers and a sort of plasticization with increased treatment time. The length of the thermal treatment reduced surface waviness. The results also showed that it was necessary to calculate the roughness paRameters to differentiate between two similar surfaces Rather than relying on visual and tactile assessments alone. [ABSTRACT FROM AUTHOR]

Different modifiers composed of polyethylene glycol, paraformaldehyde, polystyrene, and polyphosphoric acid at different ratios (designated CD-0, CD-I, CD-2. CD-3, and CD-4) were added to coal tar pitch. The resulting modified pitches were prepared and designated control D-0 and D-1, D-2, D-3, and D-4, correspondingly. The objective of this study was to evaluate the properties of the control and modified coal tar pitches by the dynamic shear rheometer (DSR) test and microscopic imaging technology. The complex shear modulus G* and phase angle 6 of the modified coal tar pitches decreased with increased temperature, which indicated a decline in the gradual capacity for resistance to permanent deformation at high temperatures. There was good con-elation between the Christensen-Andersen-Marasteanu (CAM) model and the complex modulus master curve of the modified coal tar pitches. The test results showed that D-4 had a higher complex viscosity rf than that of the control D-0 and the other modified coal tar pitches (D-1. D-2, and D-3), along with a higher glassy complex modulus G*g, cross frequency ƒc, rheological parameter r, and better resistance to rheological deformation. Control D-4 also had a strong capability to resist high-temperature permanent deformation. In addition, the homogeneity of D-4 was better than that of the control D-0 and the other modified pitches, as observed in microscopic images. Therefore, it could be concluded that D-4 had better properties compared with the others. [ABSTRACT FROM AUTHOR]

Evaluating whole slide images of histological and cytological samples is used in pathology for diagnostics, grading and prognosis . It is often necessary to rescale whole slide images of a very large size. Image resizing is one of the most common applications of interpolation. We collect the advantages and drawbacks of nine interpolation methods, and as a result of our analysis, we try to select one interpolation method as the preferred solution. To compare the performance of interpolation methods, test images were scaled and then rescaled to the original size using the same algorithm. The modified image was compared to the original image in various aspects. The time needed for calculations and results of quantification performance on modified images were also compared. For evaluation purposes, we used four general test images and 12 specialized biological immunohistochemically stained tissue sample images. The purpose of this survey is to determine which method of interpolation is the best to resize whole slide images, so they can be further processed using quantification methods. As a result, the interpolation method has to be selected depending on the task involving whole slide images. [ABSTRACT FROM AUTHOR]

This paper presents the METINUS Plus (METhod of Immunohistochemical NUclei Segmentation Plus) method that has been developed for localization and quantification of the regulatory T cells’ nuclei. The proposed methodology performs color separation followed by the extraction and validation of objects. Objects categorized as clusters, based on the area and shape, are divided with locally applied watershed supported by color deconvolution. Objects are validated based on RGB, Lab color space, and color deconvolution data. The chosen validation criteria allow quantification and differentiation between the subpopulations of regulatory T cells and breast cancer cells, which express FOXP3. The evaluation is based on the quantity and quality of nuclei segmentation in comparison with the results of the manual counting, done by four pathologists on a set of 20 images. The results were analyzed in comparison with human evaluation using Kendall's tau-b correlation coefficient and the Wilcoxon signed-rank test. The main achievement of this study is a possibility to find criteria that allow differentiation between two types of immunopositive cells’ nuclei: regulatory T cells with homogeneous stained nuclei and all cells with FOXP3 expression; in both tumor and axillary node biopsies. A major advantage of computer image processing is the reproducibility of achieved results, thus minimizing human intervention, and providing traceable clinical information. [ABSTRACT FROM AUTHOR]

Prevalence of breast cancer ranks first in the structure of women cancer. Expert diagnosis of pathological states based on qualitative performance analysis of histological images, so development processing allows making the correct diagnosis process faster and more accurate for the expert. This paper proposes a fuzzy system diagnosing of precancerous and cancerous breast conditions based on attributes obtained from the doctor-diagnostician. The system can be used in cancer medicine that will allow diagnosing in real time and eliminating subjective diagnosis by a doctor. [ABSTRACT FROM AUTHOR]

The use of Environmental Microorganisms (EMs) offers a highly efficient, low cost and harmless remedy to environmental pollution, by monitoring and decomposing of pollutants. This relies on how the EMs are correctly segmented and identified. With the aim of enhancing the segmentation of weakly visible EM images which are transparent, noisy and have low contrast, a Pairwise Deep Learning Feature Network (PDLF-Net) is proposed in this study. The use of PDLFs enables the network to focus more on the foreground (EMs) by concatenating the pairwise deep learning features of each image to different blocks of the base model SegNet. Leveraging the Shi and Tomas descriptors, we extract each image's deep features on the patches, which are centred at each descriptor using the VGG-16 model. Then, to learn the intermediate characteristics between the descriptors, pairing of the features is performed based on the Delaunay triangulation theorem to form pairwise deep learning features. In this experiment, the PDLF-Net achieves outstanding segmentation results of 89.24%, 63.20%, 77.27%, 35.15%, 89.72%, 91.44% and 89.30% on the accuracy, IoU, Dice, VOE, sensitivity, precision and specificity, respectively. • The proposed method is based on Shi and Tomasi corners and deep learning features. • Features from patches centered at descriptors are paired by Delaunay triangulation. • The pairwise features are concatenate to the SegNet to improve the performance. [ABSTRACT FROM AUTHOR]

Overall proposed architecture. [Display omitted] • Breast cancer is a leading cause of death among women. • Death rate from breast cancer is reduced when it is detected early. • In recent years, cancer-related mortality has risen dramatically. • The goal of machine learning is to make it easier for computers to learn independently. Breast cancer is a leading cause of death among women. Death rate from breast cancer is reduced when it is detected early. Deep learning (DL) methods have become a viable alternative for diagnosis, overcoming the limitations of traditional classification methods. Because biosensors and deep learning are required to identify tumors based on microscopic pictures, automation is needed. The goal of machine learning is to make it easier for computers to learn independently. This research proposed novel technique in breast cancer detection utilizing ensemble DL techniques in classification with features extraction. Input image has been collected based on web of things (WoT) in which the pre-historic data and collected microscopic images have been taken as input dataset. The input image has been processed for noise removal using Gaussian filtering and segmented using active contour convolutional neural networks. Then the classification was carried out using convoluted transfer learning integrated with regional attention mechanism. Compared to existing methods in the domain, the simulation results show that the suggested localization-based cancer classification method is superior. It has reported average classification accuracy of 96%, detection accuracy of 92%, Mean Average Precision (mAP) of 82%, sensitivity of 92%, specificity of 91%, root mean square error (RMSE) of 70% on various breast cancer microscopic image datasets. [ABSTRACT FROM AUTHOR]

Tuberculosis (TB) is an infectious disease caused by the bacteria Mycobacterium tuberculosis. It primarily affects the lungs, but it can also affect other parts of the body. TB remains one of the leading causes of death in developing countries, and its recent resurgences in both developed and developing countries warrant global attention. The number of deaths due to TB is very high (as per the WHO report, 1.5 million died in 2013), although most are preventable if diagnosed early and treated. There are many tools for TB detection, but the most widely used one is sputum smear microscopy. It is done manually and is often time consuming; a laboratory technician is expected to spend at least 15 min per slide, limiting the number of slides that can be screened. Many countries, including India, have a dearth of properly trained technicians, and they often fail to detect TB cases due to the stress of a heavy workload. Automatic methods are generally considered as a solution to this problem. Attempts have been made to develop automatic approaches to identify TB bacteria from microscopic sputum smear images. In this paper, we provide a review of automatic methods based on image processing techniques published between 1998 and 2014. The review shows that the accuracy of algorithms for the automatic detection of TB increased significantly over the years and gladly acknowledges that commercial products based on published works also started appearing in the market. This review could be useful to researchers and practitioners working in the field of TB automation, providing a comprehensive and accessible overview of methods of this field of research. [ABSTRACT FROM AUTHOR]

Accurate image segmentation of cells and tissues is a challenging research area due to its vast applications in medical diagnosis. Seed detection is the basic and most essential step for the automated segmentation of microscopic images. This paper presents a robust, accurate and novel method for detecting cell nuclei which can be efficiently used for cell segmentation. We propose a template matching method using a feature similarity index measure (FSIM) for detecting nuclei positions in the image which can be further used as seeds for segmentation tasks. Initially, a Fuzzy C-Means clustering algorithm is applied on the image for separating the foreground region containing the individual and clustered nuclei regions. FSIM based template matching approach is then used for nuclei detection. FSIM makes use of low level texture features for comparisons and hence gives good results. The performance of the proposed method is evaluated on the gold standard dataset containing 36 images (∼8000 nuclei) of tissue samples and also in vitro cultured cell images of Stromal Fibroblasts (5 images) and Human Macrophage cell line (4 images) using the statistical measures of Precision and Recall. The results are analyzed and compared with other state-of-the-art methods in the literature and software tools to prove its efficiency. Precision is found to be comparable and the Recall rate is found to exceed 92% for the gold standard dataset which shows considerable performance improvement over existing methods. [ABSTRACT FROM AUTHOR]

This article presents an automatic system for assessing water quality based on fish gills microscopic images. As fish gills are a good biomarker for assessing water quality, the proposed system uses fish gills microscopic images in order to detect water pollution. The proposed system consists of three phases; namely pre-processing, feature extraction, and classification phases. Since the shape is the main characteristic of fish gills microscopic images, the proposed system uses shape feature based on edge detection and wavelets transform for classifying the water-quality degree. Furthermore, it implemented Principal Components Analysis (PCA) along with Support Vector Machines (SVMs) algorithms for feature extraction and water quality degree classification. The datasets used for experiments were constructed based on real sample images for fish gills. Training dataset is divided into four classes representing the different histopathological changes and the corresponding water quality degrees. Experimental results showed that the proposed classification system has obtained water quality classification accuracy of 95.41%, using the SVMs linear kernel function and 10-fold cross validation with 37 images per class for training. [ABSTRACT FROM AUTHOR]

• Deep learning based efficient classification of plasmodium vivax life cycle to diagnose malaria disease using thin blood smear microscopic images. • Automatic, real-time and robust to real possible inputs and accuracy achieved up to 90.03% on unseen images. • Approach targets complete lifecycle of plasmodium vivax malaria species. • Analysis of different state-of-the-art deep learning models and vision transformer on microscopic dataset. Microscopic images are an important part for haematologists in diagnosing various diseases in the blood cell. Changes in blood cells are caused by malaria disease, and early diagnosis can prevent the disease from entering its severe stage. In this paper, an automated non-invasive and efficient deep learning-based framework is developed for multi-class plasmodium vivax life cycle classification and malaria diagnosis. A multi-class microscopic blood cell of different plasmodium vivax life cycle stage dataset is analysed, and a diagnostic framework is designed. Several stages of the disease are examined and augmented through various techniques to make the framework robust in real-time. Generative adversarial network is specially designed to generate extended training samples of various life cycle stages to increase robustness of the resulting model. A special transformer-based neural network vision transformer is designed to improve generalisation capabilities. Microscopic images are classified into multi classes of plasmodium vivax life cycle stage, where the keystone transformer layers extract relevant disease features from microscopic colour images, and the extracted relevant features are used to make predictive diagnostic decisions. The capabilities of the vision transformer are computed and analysed by statistical parameters, and the performance of the vision transformer model is compared with baseline architectures, where it was evident that the performance of the vision transformer was significantly better, reaching 90.03% accuracy. A comprehensive comparison of the proposed framework to the state-of-the-art methods proves its efficiency in the classification of plasmodium vivax life cycle for malaria disease identification through thin blood smear microscopic images. [ABSTRACT FROM AUTHOR]

Plant disease classification is quite complex and, in most cases, requires trained plant pathologists and sophisticated labs to accurately determine the cause. Our group for the first time used microscopic images (×30) of tomato plant diseases, for which representative plant samples were diagnostically validated to classify disease symptoms using non-coding deep learning platforms (NCDL). The mean F1 scores (SD) of the NCDL platforms were 98.5 (1.6) for Amazon Rekognition Custom Label, 93.9 (2.5) for Clarifai, 91.6 (3.9) for Teachable Machine, 95.0 (1.9) for Google AutoML Vision, and 97.5 (2.7) for Microsoft Azure Custom Vision. The accuracy of the NCDL platform for Amazon Rekognition Custom Label was 99.8% (0.2), for Clarifai 98.7% (0.5), for Teachable Machine 98.3% (0.4), for Google AutoML Vision 98.9% (0.6), and for Apple CreateML 87.3 (4.3). Upon external validation, the model's accuracy of the tested NCDL platforms dropped no more than 7%. The potential future use for these models includes the development of mobile- and web-based applications for the classification of plant diseases and integration with a disease management advisory system. The NCDL models also have the potential to improve the early triage of symptomatic plant samples into classes that may save time in diagnostic lab sample processing., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision., (Copyright © 2023 Choudhary, Sentil, Jones and Paret.)

Expertise studies in the medical domain often focus on either visual or cognitive aspects of expertise. As a result, characteristics of expert behaviour are often described as either cognitive or visual abilities. This study focuses on both aspects of expertise and analyses them along three overarching constructs: (1) encapsulations, (2) efficiency, and (3) hypothesis testing. This study was carried out among clinical pathologists performing an authentic task: diagnosing microscopic slides. Participants were 13 clinical pathologists (experts), 12 residents in pathology (intermediates), and 13 medical students (novices). They all diagnosed seven cases in a virtual microscope and gave post hoc explanations for their diagnoses. The collected data included eye movements, microscope navigation, and verbal protocols. Results showed that experts used lower magnifications and verbalized their findings as diagnoses. Also, their diagnostic paths were more efficient, including fewer microscope movements and shorter reasoning chains. Experts entered relevant areas later in their diagnostic process, and visited fewer of them. Intermediates used relatively high magnifications and based their diagnoses on specific abnormalities. Also, they took longer to reach their diagnosis and checked more relevant areas. Novices searched in detail, described findings by their appearances, and uttered long reasoning chains. These results indicate that overarching constructs can justly be identified: encapsulations and efficiency are apparent in both visual and cognitive aspects of expertise. [ABSTRACT FROM AUTHOR]

Malaria, one of the deadliest diseases, is responsible for nearly 627,000 deaths every year. It is diagnosed manually by pathologists using a microscope. It is time-consuming and subjected to inconsistency due to human intervention, so computerized image analysis for diagnosis has gained importance. In this article, an edge-based segmentation of erythrocytes infected with malaria parasites using microscopic images has been developed to facilitate the diagnostic process. The color space transformation and Gamma equalization reduce the effects of colors and correct luminance differences of images. Fuzzy C-means clustering is applied to extract infected erythrocytes, which is further processed for the final segmentation. The experimental results showed that the proposed method can gain 98%, 93.3%, 98.65% and 90.33% of sensitivity, specificity, prediction value positive and prediction value negative, respectively. In conclusion, the proposed method provides a consistent and robust method of edge-based segmentation of parasite infected erythrocytes using microscopic images for diagnosis. [ABSTRACT FROM AUTHOR]

Multiple classifiers on the dissimilarity space are proposed to address the problem of forest species recognition from microscopic images. To that end, classical texture-based features such as Gabor filters, local binary patterns (LBP) and local phase quantization (LPQ), as well as two keypoint-based features, the scale-invariant feature transform (SIFT) and the speeded up robust features (SURF), are used to generate a pool of diverse classifiers on the dissimilarity space. A comprehensive set of experiments on a database composed of 2,240 microscopic images from 112 different forest species was used to evaluate the performance of each individual classifier of the generated pool, the combination of all classifiers, and different dynamic selection of classifiers (DSC) methods. The best result (93.03 %) was observed by incorporating probabilistic information in a DSC method based on multiple classifier behavior. [ABSTRACT FROM AUTHOR]

The authentication of natural medical herbs has crucial impact on the clinical curative effect. Much of these herbs have been ground into powders in the market, leading to the difficulty of identification. Microscopic images of these powders contain important evidences for identification. Currently identification based on these microscopic images are mostly conducted by manual observation. Identification aided by computer vision technique is an important research subject recently. These microscopic images usually contain variety of substance, and most of them are noises, thus the target segmentation is necessary for identification. An effective automatic target segmentation algorithm based on texture is proposed in this paper. Our method consists of two steps: 'Preliminary Segmentation' and 'Further Segmentation'. Firstly, gradient transform and image fusion are conducted for image preprocessing, then each pixel is encoded into a feature vector based on texture and clustered into two groups: background and foreground. Secondly, taking the continuity of edge and the locality of target into consideration, energy equations are established, and maximum flow-minimum cut algorithm is applied to solve them. Three groups of images are tested to evaluate our method, and the experimental results show that our method achieves a better segmentation compared with Grab-Cut, and additionally user inter-action is not required in our method. [ABSTRACT FROM AUTHOR]

In our paper, we present a performance evaluation of image segmentation algorithms on microscopic image data. In spite of the existence of many algorithms for image data partitioning, there is no universal and 'the best' method yet. Moreover, images of microscopic samples can be of various character and quality which can negatively influence the performance of image segmentation algorithms. Thus, the issue of selecting suitable method for a given set of image data is of big interest. We carried out a large number of experiments with a variety of segmentation methods to evaluate the behaviour of individual approaches on the testing set of microscopic images (cross-section images taken in three different modalities from the field of art restoration). The segmentation results were assessed by several indices used for measuring the output quality of image segmentation algorithms. In the end, the benefit of segmentation combination approach is studied and applicability of achieved results on another representatives of microscopic data category - biological samples - is shown. [ABSTRACT FROM AUTHOR]

Environmental microorganisms (EMs) are single-celled or multi-cellular microscopic organisms living in the environments. They are crucial to nutrient recycling in ecosystems as they act as decomposers. Occurrence of certain EMs and their species are very informative indicators to evaluate environmental quality. However, the manual recognition of EMs in microbiological laboratories is very time-consuming and expensive. Therefore, in this article an automatic EM classification system based on content-based image analysis (CBIA) techniques is proposed. Our approach starts with image segmentation that determines the region of interest (EM shape). Then, the EM is described by four different shape descriptors, whereas the Internal Structure Histogram (ISH), a new and original shape feature extraction technique introduced in this paper, has turned out to possess the most discriminative properties in this application domain. Afterwards, for each descriptor a support vector machine (SVM) is constructed to distinguish different classes of EMs. At last, results of SVMs trained for all four feature spaces are fused in order to obtain the final classification result. Experimental results certify the effectiveness and practicability of our automatic EM classification system. [ABSTRACT FROM AUTHOR]

Background Recent advances in microscopy enable the acquisition of large numbers of tomographic images from living tissues. Three-dimensional microscope images are often displayed with volume rendering by adjusting the transfer functions. However, because the emissions from fluorescent materials and the optical properties based on point spread functions affect the imaging results, the intensity value can differ locally, even in the same structure. Further, images obtained from brain tissues contain a variety of neural structures such as dendrites and axons with complex crossings and overlapping linear structures. In these cases, the transfer functions previously used fail to optimize image generation, making it difficult to explore the connectivity of these tissues. Results This paper proposes an interactive visual exploration method by which the transfer functions are modified locally and interactively based on multidimensional features in the images. A direct editing interface is also provided to specify both the target region and structures with characteristic features, where all manual operations can be performed on the rendered image. This method is demonstrated using two-photon microscope images acquired from living mice, and is shown to be an effective method for interactive visual exploration of overlapping similar structures. Conclusions An interactive visualization method was introduced for local improvement of visualization by volume rendering in two-photon microscope images containing regions in which linear nerve structures crisscross in a complex manner. The proposed method is characterized by the localized multidimensional transfer function and interface where the parameters can be determined by the user to suit their particular visualization requirements. [ABSTRACT FROM AUTHOR]

Most in situ tool wear monitoring methods during micro end milling rely on signals captured from the machining process to evaluate tool wear behavior; accurate positioning in the tool wear region and direct measurement of the level of wear are difficult to achieve. In this paper, an in situ monitoring system based on machine vision is designed and established to monitor tool wear behavior in micro end milling of titanium alloy Ti6Al4V. Meanwhile, types of tool wear zones during micro end milling are discussed and analyzed to obtain indicators for evaluating wear behavior. Aiming to measure such indicators, this study proposes image processing algorithms. Furthermore, the accuracy and reliability of these algorithms are verified by processing the template image of tool wear gathered during the experiment. Finally, a micro end milling experiment is performed with the verified micro end milling tool and the main wear type of the tool is understood via in-situ tool wear detection. Analyzing the measurement results of evaluation indicators of wear behavior shows the relationship between the level of wear and varying cutting time; it also gives the main influencing reasons that cause the change in each wear evaluation indicator.

The article focuses on a proposed technique based on adaptive discrete wavelet entropy energy and neural network classifier (ADWEENN) for the recognition of urine cells from microscopic images that are independent of rotation and scaling. It mentions the importance of the microscopic examination of urine cells in renal and urinary tract disease evaluation, stages in the ADWEENN algorithm such as preprocessing, feature extraction and classification, and the finding of 97.58% recognition success.