Automatic recognition of different tissue types in histological images is an essential part in the digital pathology toolbox. Texture analysis is commonly used to address this problem; mainly in the context of estimating the tumour/stroma ratio on histological samples. However, although histological images typically contain more than two tissue types, only few studies have addressed the multi-class problem. For colorectal cancer, one of the most prevalent tumour types, there are in fact no published results on multiclass texture separation. In this paper we present a new dataset of 5,000 histological images of human colorectal cancer including eight different types of tissue. We used this set to assess the classification performance of a wide range of texture descriptors and classifiers. As a result, we found an optimal classification strategy that markedly outperformed traditional methods, improving the state of the art for tumour-stroma separation from 96.9% to 98.6% accuracy and setting a new standard for multiclass tissue separation (87.4% accuracy for eight classes). We make our dataset of histological images publicly available under a Creative Commons license and encourage other researchers to use it as a benchmark for their studies.
Color texture classification has been an area of intensive research activity. From the very onset, approaches to combining color and texture have been the subject of much discussion, and in particular, whether they should be considered joint or separately. We present a comprehensive comparison of the most prominent approaches both from a theoretical and experimental standpoint. The main contributions of our work are: (i) the establishment of a generic and extensible framework to classify methods for color texture classification on a mathematical basis, and (ii) a theoretical and experimental comparison of the most salient existing methods. Starting from an extensive set of experiments based on the Outex dataset, we highlight those texture descriptors that provide good accuracy along with low dimensionality. The results suggest that separate color and texture processing is the best practice when one seeks for optimal compromise between accuracy and limited number of features. We believe that our work may serve as a guide for those who need to choose the appropriate method for a specific application, as well as a basis for the development of new methods.
The aim of this paper is to describe a general framework for texture analysis which we refer to as the HEP (histograms of equivalent patterns). The HEP, of which we give a clear and unambiguous mathematical definition, is based on partitioning the feature space associated to image patches of predefined shape and size. This task is approached by defining, a priori, suitable local or global functions of the pixels' intensities. In a comprehensive survey we show that diverse texture descriptors, such as co-occurrence matrices, gray-level differences and local binary patterns, can be seen all to be examples of the HEP. In the experimental part we comparatively evaluate a comprehensive set of these descriptors on an extensive texture classification task. Within the class of HEP schemes, improved local ternary patterns (ILTP) and completed local binary patterns (CLBP) emerge as the best of parametric and nonparametric methods, respectively. The results also show the following patterns: (1) higher effectiveness of multi-level discretization in comparison with binarization; (2) higher F. Bianconi performed this work as a visiting researcher in the accuracy of parametric methods when compared to nonparametric ones; (3) a general trend of increasing performance with increasing dimensionality; and (4) better performance of point-to-average thresholding against point-topoint thresholding.
Convolutional Neural Networks (CNN) have brought spectacular improvements in several fields of machine vision including object, scene and face recognition. Nonetheless, the impact of this new paradigm on the classification of fine-grained images—such as colour textures—is still controversial. In this work, we evaluate the effectiveness of traditional, hand-crafted descriptors against off-the-shelf CNN-based features for the classification of different types of colour textures under a range of imaging conditions. The study covers 68 image descriptors (35 hand-crafted and 33 CNN-based) and 46 compilations of 23 colour texture datasets divided into 10 experimental conditions. On average, the results indicate a marked superiority of deep networks, particularly with non-stationary textures and in the presence of multiple changes in the acquisition conditions. By contrast, hand-crafted descriptors were better at discriminating stationary textures under steady imaging conditions and proved more robust than CNN-based features to image rotation.
In this paper, we investigate the role of shape and texture features from 18F-FDG PET/CT to discriminate between benign and malignant solitary pulmonary nodules. To this end, we retrospectively evaluated cross-sectional data from 111 patients (64 males, 47 females, age = 67.5 ± 11.0) all with histologically confirmed benign (n=39) or malignant (n=72) solitary pulmonary nodules. Eighteen three-dimensional imaging features, including conventional, texture, and shape features from PET and CT were tested for significant differences (Wilcoxon-Mann-Withney) between the benign and malignant groups. Prediction models based on different feature sets and three classification strategies (Classification Tree, k-Nearest Neighbours, and Naïve Bayes) were also evaluated to assess the potential benefit of shape and texture features compared with conventional imaging features alone. Eight features from CT and 15 from PET were significantly different between the benign and malignant groups. Adding shape and texture features increased the performance of both the CT-based and PET-based prediction models with overall accuracy gain being 3.4–11.2 pp and 2.2–10.2 pp, respectively. In conclusion, we found that shape and texture features from 18F-FDG PET/CT can lead to a better discrimination between benign and malignant lung nodules by increasing the accuracy of the prediction models by an appreciable margin.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.