Search

Early diagnosis of potentially malignant disorders, such as oral epithelial dysplasia, is the most reliable way to prevent oral cancer. Computational algorithms have been used as an auxiliary tool to aid specialists in this process. Usually, experiments are performed on private data, making it difficult to reproduce the results. There are several public datasets of histological images, but studies focused on oral dysplasia images use inaccessible datasets. This prevents the improvement of algorithms aimed at this lesion. This study introduces an annotated public dataset of oral epithelial dysplasia tissue images. The dataset includes 456 images acquired from 30 mouse tongues. The images were categorized among the lesion grades, with nuclear structures manually marked by a trained specialist and validated by a pathologist. Also, experiments were carried out in order to illustrate the potential of the proposed dataset in classification and segmentation processes commonly explored in the literature. Convolutional neural network (CNN) models for semantic and instance segmentation were employed on the images, which were pre-processed with stain normalization methods. Then, the segmented and non-segmented images were classified with CNN architectures and machine learning algorithms. The data obtained through these processes is available in the dataset. The segmentation stage showed the F1-score value of 0.83, obtained with the U-Net model using the ResNet-50 as a backbone. At the classification stage, the most expressive result was achieved with the Random Forest method, with an accuracy value of 94.22%. The results show that the segmentation contributed to the classification results, but studies are needed for the improvement of these stages of automated diagnosis. The original, gold standard, normalized, and segmented images are publicly available and may be used for the improvement of clinical applications of CAD methods on oral epithelial dysplasia tissue images., (© 2024. The Author(s) under exclusive licence to Society for Imaging Informatics in Medicine.)

In this work, a computational scheme is proposed to identify the main combinations of handcrafted descriptors and deep-learned features capable of classifying histological images stained with hematoxylin and eosin. The handcrafted descriptors were those representatives of multiscale and multidimensional fractal techniques (fractal dimension, lacunarity and percolation) applied to quantify the histological images with the corresponding representations via explainable artificial intelligence (xAI) approaches. The deep-learned features were obtained from different convolutional neural networks (DenseNet-121, EfficientNet-b2, Inception-V3, ResNet-50 and VGG-19). The descriptors were investigated through different associations. The most relevant combinations, defined through a ranking algorithm, were analyzed via a heterogeneous ensemble of classifiers with the support vector machine, naive Bayes, random forest and K-nearest neighbors algorithms. The proposed scheme was applied to histological samples representative of breast cancer, colorectal cancer, oral dysplasia and liver tissue. The best results were accuracy rates of 94.83% to 100%, with the identification of pattern ensembles for classifying multiple histological images. The computational scheme indicated solutions exploring a reduced number of features (a maximum of 25 descriptors) and with better performance values than those observed in the literature. The presented information in this study is useful to complement and improve the development of computer-aided diagnosis focused on histological images.

Histological images stained with hematoxylin-eosin are widely used by pathologists for cancer diagnosis. However, these images can have color variations that highly influence the histological image processing techniques. To deal with this potential limitation, normalization methods are useful for color correction. In this paper, a histological image color normalization is presented by considering the biological and hematoxylin-eosin properties. To this end, the stain representation of a reference image was applied in place of the original images representation, allowing the preservation of histological structures. This proposal was evaluated on histological images with great variations of contrast, and both visual and quantitative analyzes yielded promising results., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

Histological samples stained with hematoxylin-eosin (H&E) are commonly used by pathologists in cancer diagnoses. However, the preparation, digitization, and storage of tissue samples can lead to color variations that produce poor performance when using histological image processing techniques. Thus, normalization methods have been proposed to adjust the color of the image. This can be achieved through the use of a spectral matching technique, where it is first necessary to estimate the H&E representation and the stain concentration in the image pixels by means of the RGB model. This study presents an estimation method for H&E stain representation for the normalization of faded histological samples. This application has been explored only to a limited extent in the literature, but has the capacity to expand the use of faded samples. To achieve this, the normalized images must have a coherent color representation of the H&E stain with no introduction of noise, which was realized by applying the methodology described in this proposal. The estimation method presented here aims to normalize histological samples with different degrees of fading using a combination of fuzzy theory and the Cuckoo search algorithm, and dictionary learning with an initialization method for optimization. In visual and quantitative comparisons of estimates of H&E stain representation from the literature, our proposed method achieved very good results, with a high feature similarity between the original and normalized images., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

Non-Hodgkin lymphomas are a health problem that affects over 70,000 people per year in the United States alone. The early diagnosis and the identification of this lymphoma are essential for an effective treatment. The classification of non-Hodgkin lymphomas is a task that continues to rank as one of the main challenges faced by hematologists, pathologists, as well as in the producing of computer vision methods due to its inherent complexity. In this paper, we present a new method to quantify and classify tissue samples of non-Hodgkin lymphomas based on the percolation theory. The method consists of associating multiscale and multidimensional approaches in order to divide the image into smaller regions and then verifying color similarity between pixels. A cluster labeling algorithm was applied to each region of interest to obtain the values for the number of clusters, occurrence of percolation and coverage ratio of the largest cluster. The method was tested on different classifiers aiming to differentiate three different groups of non-Hodgkin lymphomas. The obtained results (AUC rates between 0.940 and 0.993) were compared to those provided by methods consolidated in the Literature, which indicates that the percolation theory is a suitable approach for identifying these three classes of non-Hodgkin lymphomas, those being: mantle cell lymphoma, follicular lymphoma and chronic lymphocytic leukemia., (Copyright © 2017 Elsevier Ltd. All rights reserved.)