Histopathological images (HIs) are the gold standard for evaluating some types of tumors for cancer diagnosis. The analysis of such images is time and resource-consuming and very challenging even for experienced pathologists, resulting in inter-observer and intra-observer disagreements. One of the ways of accelerating such an analysis is to use computer-aided diagnosis (CAD) systems. This paper presents a review on machine learning methods for histopathological image analysis, including shallow and deep learning methods. We also cover the most common tasks in HI analysis, such as segmentation and feature extraction. Besides, we present a list of publicly available and private datasets that have been used in HI research.

Information from an image occurs over multiple and distinct spatial scales. Image pyramid multiresolution representations are a useful data structure for image analysis and manipulation over a spectrum of spatial scales. This paper employs the Gaussian–Laplacian pyramid to separately treat different spatial frequency bands of a texture. First, we generate three images corresponding to three levels of the Gaussian–Laplacian pyramid for an input image to capture intrinsic details. Then, we aggregate features extracted from gray and color texture images using bioinspired texture descriptors, information-theoretic measures, gray-level co-occurrence matrix feature descriptors, and Haralick statistical feature descriptors into a single feature vector. Such an aggregation aims at producing features that characterize textures to their maximum extent, unlike employing each descriptor separately, which may lose some relevant textural information and reduce the classification performance. The experimental results on texture and histopathologic image datasets have shown the advantages of the proposed method compared to state-of-the-art approaches. Such findings emphasize the importance of multiscale image analysis and corroborate that the descriptors mentioned above are complementary.

Age-related macular degeneration (AMD) is a degenerative ocular disease that develops by the formation of drusen in the macula region leading to blindness. This condition can be detected automatically by automated image processing techniques applied in spectral domain optical coherence tomography (SD-OCT) volumes. The most common approach is the individualized analysis of each slice (B-Scan) of the SD-OCT volumes. However, it ends up losing the correlation between pixels of neighboring slices. The retina representation by topographic maps reveals the similarity of these structures with geographic relief maps, which can be represented by geostatistical descriptors. In this paper, we present a methodology based on geostatistical functions for the automatic diagnosis of AMD in SD-OCT. The proposed methodology is based on the construction of a topographic map of the macular region. Over the topographic map, we compute geostatistical features using semivariogram and semimadogram functions as texture descriptors. The extracted descriptors are then used as input for a Support Vector Machine classifier. For training of the classifier and tests, a database composed of 384 OCT exams (269 volumes of eyes exhibiting AMD and 115 control volumes) with layers segmented and validated by specialists were used. The best classification model, validated with cross-validation k-fold, achieved an accuracy of 95.2% and an AUROC of 0.989. The presented methodology exclusively uses geostatistical descriptors for the diagnosis of AMD in SD-OCT images of the macular region. The results are promising and the methodology is competitive considering previous results published in literature.