Deep learning (DL) and convolutional neural networks (CNNs) have achieved state-of-the-art performance in many medical image analysis tasks. Histopathological images contain valuable information that can be used to diagnose diseases and create treatment plans. Therefore, the application of DL for the classification of histological images is a rapidly expanding field of research. The popularity of CNNs has led to a rapid growth in the number of works related to CNNs in histopathology. This paper aims to provide a clear overview for better navigation. In this paper, recent DL-based classification studies in histopathology using strongly annotated data have been reviewed. All the works have been categorized from two points of view. First, the studies have been categorized into three groups according to the training approach and model construction: 1. fine-tuning of pre-trained networks for one-stage classification, 2. training networks from scratch for one-stage classification, and 3. multi-stage classification. Second, the papers summarized in this study cover a wide range of applications (e.g., breast, lung, colon, brain, kidney). To help navigate through the studies, the classification of reviewed works into tissue classification, tissue grading, and biomarker identification was used.

In this work, we describe a dataset suitable for analyzing the extent to which hematoxylin–eosin (HE)-stained tissue contains information about the expression of Ki67 in immunohistochemistry staining. The dataset provides images of corresponding pairs of HE and Ki67 stainings and is complemented by algorithms for computing the Ki67 index. We introduce a dataset of high-resolution histological images of testicular seminoma tissue. The dataset comprises digitized histology slides from 77 conventional testicular seminoma patients, obtained via surgical resection. For each patient, two physically adjacent tissue sections are stained: one with hematoxylin and eosin, and one with Ki67 immunohistochemistry staining. This results in a total of 154 high-resolution images. The images are provided in PNG format, facilitating ease of use for image analysis compared to the original scanner output formats. Each image contains enough tissue to generate thousands of non-overlapping 224 × 224 pixel patches. This shows the potential to generate more than 50,000 pairs of patches, one with HE staining and a corresponding Ki67 patch that depicts a very similar part of the tissue. Finally, we present the results of applying a ResNet neural network for the classification of HE patches into categories according to their Ki67 label.

Hematoxylin and Eosin (HE) staining is the gold standard in histopathological examination of cancer tissue, representing the first step towards cancer diagnosis. The second step is a series of immunohistochemical stainings, including cell proliferation markers called the Ki67 index. Deep learning models offer promising solutions for improving medical diagnostics, while generative models provide additional explainability of predictive models that is essential for their adoption in clinical practice. Our previous work introduced a novel approach that utilises a conditional StyleGAN model for generating HE-stained images conditioned on the Ki67 index. This study proposes to employ this model for generating sequences of HE-stained images reflecting varying Ki67 index values. Sequences enable exploration of hidden relationships between HE and Ki67 staining and can enhance the explainability of predictive models, e.g., by generating counterfactual examples. While our previous research focused on assessing the quality of generated HE images, this study extends that work by evaluating the model’s ability to capture Ki67-related variations in HE-stained images. Additionally, expert pathologists evaluated generated sequences and proposed criteria for assessing their relevance. Our findings demonstrate the potential of the conditional StyleGAN model as part of an explainable framework for analysing and predicting immunohistochemical information from HE-stained images. Results highlight the relevance of generative models in histopathology and their potential applications in cancer progression analysis.

Hematoxylin and Eosin (HE) staining is gold standard in histopathological examination of cancer tissue, representing first step towards cancer diagnosis. Second step is series of immunohistochemical stainings, including cell proliferation marker called Ki67 index. Deep learning models offer promising solutions for improving medical diagnostics, while generative models provide additional explainability of predictive models, which is essential for their adoption in clinical practice. Our previous work introduced novel approach that utilises conditional StyleGAN model for generating HE-stained images conditioned on Ki67 index. This study proposes to employ this model for generating sequences of HE-stained images reflecting varying Ki67 index values. Sequences enable exploration of hidden relationships between HE and Ki67 staining and can enhance explainability of predictive models, e.g., by generating counterfactual examples. While our previous research focused on assessing quality of generated HE images, this study extends that work by evaluating model’s ability to capture Ki67-related variations in HE-stained images. Additionally, expert pathologists evaluated generated sequences, proposing criteria for assessing their relevance. Our findings demonstrate potential of conditional StyleGAN model as part of explainable framework for analysing and predicting immunohistochemical information from HE-stained images. Results highlight relevance of generative models in histopathology and their potential applications in cancer progression analysis.