CLASS-M: Adaptive stain separation-based contrastive learning with pseudo-labeling for histopathological image classification.

MF-OMKT: Model fusion based on online mutual knowledge transfer for breast cancer histopathological image classification

Histopathological image contains rich pathological information that is valued for the aided diagnosis of many diseases such as cancer. An important issue in histopathological image classification is how to learn a high-quality discriminative dictionary due to diverse tissue pattern, a variety of texture, and different morphologies structure. In this paper, we propose a discriminative dictionary learning algorithm with pairwise local constraints (PLCDDL) for histopathological image classification. Inspired by the one-to-one mapping between dictionary atom and profile, we learn a pair of discriminative graph Laplacian matrices that are less sensitive to noise or outliers to capture the locality and discriminating information of data manifold by utilizing the local geometry information of category-specific dictionaries rather than input data. Furthermore, graph-based pairwise local constraints are designed and incorporated into the original dictionary learning model to effectively encode the locality consistency with intra-class samples and the locality inconsistency with inter-class samples. Specifically, we learn the discriminative localities for representations by jointly optimizing both the intra-class locality and inter-class locality, which can significantly improve the discriminability and robustness of dictionary. Extensive experiments on the challenging datasets verify that the proposed PLCDDL algorithm can achieve a better classification accuracy and powerful robustness compared with the state-of-the-art dictionary learning methods. Graphical abstract The proposed PLCDDL algorithm. 1) A pair of graph Laplacian matrices are first learned based on the class-specific dictionaries. 2) Graph-based pairwise local constraints are designed to transfer the locality for coding coefficients. 3) Class-specific dictionaries can be further updated.

The field of histopathological image analysis has evolved significantly with the advent of digital pathology, leading to the development of automated models capable of classifying tissues and structures within diverse pathological images. Artificial intelligence algorithms, such as convolutional neural networks, have shown remarkable capabilities in pathology image analysis tasks, including tumor identification, metastasis detection, and patient prognosis assessment. However, traditional manual analysis methods have generally shown low accuracy in diagnosing colorectal cancer using histopathological images. This study investigates the use of AI in image classification and image analytics using histopathological images using the histogram of oriented gradients method. The study develops an AI-based architecture for image classification using histopathological images, aiming to achieve high performance with less complexity through specific parameters and layers. In this study, we investigate the complicated state of histopathological image classification, explicitly focusing on categorizing nine distinct tissue types. Our research used open-source multi-centered image datasets that included records of 100.000 non-overlapping images from 86 patients for training and 7180 non-overlapping images from 50 patients for testing. The study compares two distinct approaches, training artificial intelligence-based algorithms and manual machine learning models, to automate tissue classification. This research comprises two primary classification tasks: binary classification, distinguishing between normal and tumor tissues, and multi-classification, encompassing nine tissue types, including adipose, background, debris, stroma, lymphocytes, mucus, smooth muscle, normal colon mucosa, and tumor. Our findings show that artificial intelligence-based systems can achieve 0.91 and 0.97 accuracy in binary and multi-class classifications. In comparison, the histogram of directed gradient features and the Random Forest classifier achieved accuracy rates of 0.75 and 0.44 in binary and multi-class classifications, respectively. Our artificial intelligence-based methods are generalizable, allowing them to be integrated into histopathology diagnostics procedures and improve diagnostic accuracy and efficiency. The CNN model outperforms existing machine learning techniques, demonstrating its potential to improve the precision and effectiveness of histopathology image analysis. This research emphasizes the importance of maintaining data consistency and applying normalization methods during the data preparation stage for analysis. It particularly highlights the potential of artificial intelligence to assess histopathological images.

Automated classification of histopathological images is an important research problem in medical imaging. Digital histopathology exhibits two principally distinct characteristics: 1) invariably histopathological images are multi-channel (color) with key geometric information spread across the color channels instead of being captured by luminance alone, and 2) the richness of geometric structures in such tissue imagery makes feature extraction for classification very demanding. Inspired by recent work in the use of sparsity for single channel image classification, we propose a new simultaneous Sparsity model for multi-channel Histopathological Image Representation and Classification (SHIRC). Essentially, we represent a multi-channel histopathological image as a sparse linear combination of training examples under suitable channel-wise constraints and classification is performed by solving a newly formulated simultaneous sparsity-based optimization problem. Experiments on two challenging real-world image databases: 1) provided by pathologists of the Animal Diagnostics Lab (ADL) at Pennsylvania State University, and 2) histopathological images corresponding to intraductal breast lesions [1], reveal the merits of the proposed SHIRC model over state of the art alternatives.

Accurate histopathological image classification plays a crucial role in cancer detection and diagnosis. In automated cancer detection methods, extraction of histological features of malignant and benign tissues is a challenging task. This paper presents a modified approach on octave convolution to extract high and low-frequency features which help to provide a comprehensive representation of histopathological images. Proposed octave convolution model is used to perform histopathological image classification using three different optimization strategies. Firstly, an optimal alpha value of 0.5 is used to give equal importance to both high-frequency and low-frequency feature maps. This balanced approach ensures that the model effectively considers critical high-frequency features as well as low-frequency features of cancerous tissues. Secondly, high-frequency and low-frequency feature maps are extracted and down sampled into half the spatial dimension size to reduce the computational cost compared to standard CNN. Thirdly, training and validation was conducted using ReLU, PReLU, LeakyReLU, ELU, GELU and Swish activation functions. From the experiment, it was concluded that PReLU is the best activation function for capturing intricate patterns inherent in cancer-related histopathological images. Combining all these optimization strategies, the proposed method proved to provide a classification accuracy of 93% and also to reduce the computational cost by 50%. Performance validation against pre-trained models, CNN variants and vision transformer-based models has also been conducted, which proved superior performance of the proposed model.

Deep neural network incorporating domain and resolution transformations model for histopathological image classification

Breast cancer is a big concern for women due to its higher mortality compared to other cancers. Objective and accurate early diagnosis is primordial for the treatment and survival improvement of patients. Histopathological image classification is considered the gold standard and is usually the last and most dependent diagnosis approach for doctors to make patient treatment proposals. In particular, recent deep learning-based methods provide remarkable classification results. However, these methods ignore rationale or logical explanation that is important for diagnosis reliability and human-level understanding. This paper proposes a multi-instance classification network (MICNet) with the mechanism of visual explanation to achieve the explainable classification of histopathological breast cancer images. The method uses a simple two-dimensional convolution kernel to generate explanation maps (i.e., visual explanation) through features coming from the end of the feature extractor in the VGG11 model pre-trained by ImageNet. Multiple instance learning (MIL) based on mirror padding and overlap cropping is adopted to improve the network’s classification performance. We also design a weighted average pooling method to encourage the network to learn more accurate visual explanation. Experiments on BreakHis and Camelyon16 patch-based datasets demonstrate that our MICNet outperforms other CNN models in classification and is able to provide a logical visual explanation that supports the network’s prediction.

The application of machine learning techniques to histopathology images enables advances in the field, providing valuable tools that can speed up and facilitate the diagnosis process. The classification of these images is a relevant aid for physicians who have to process a large number of images in long and repetitive tasks. This work proposes the adoption of metric learning that, beyond the task of classifying images, can provide additional information able to support the decision of the classification system. In particular, triplet networks have been employed to create a representation in the embedding space that gathers together images of the same class while tending to separate images with different labels. The obtained representation shows an evident separation of the classes with the possibility of evaluating the similarity and the dissimilarity among input images according to distance criteria. The model has been tested on the BreakHis dataset, a reference and largely used dataset that collects breast cancer images with eight pathology labels and four magnification levels. Our proposed classification model achieves relevant performance on the patient level, with the advantage of providing interpretable information for the obtained results, which represent a specific feature missed by the all the recent methodologies proposed for the same purpose.

Techniques of medical image processing and analysis play a crucial role in many clinical scenarios, including in diagnosis and treatment planning. However, immense quantities of data and high complexity of the algorithms often used are computationally demanding. As a result, there now exists a wide range of techniques of medical image processing and analysis that require the application of high-performance computing solutions in order to reduce the required runtime. The main purpose of this review is to provide a comprehensive reference source of techniques of medical image processing and analysis that have been accelerated by high-performance computing solutions. With this in mind, the articles available in the Scopus and Web of Science electronic repositories were searched. Subsequently, the most relevant articles found were individually analyzed in order to identify: (a) the metrics used to evaluate computing performance, (b) the high-performance computing solution used, (c) the parallel design adopted, and (d) the task of medical image processing and analysis involved. Hence, the techniques of medical image processing and analysis found were identified, reviewed, and discussed, particularly in terms of computational performance. Consequently, the techniques reviewed herein present the progress made so far in reducing the computational runtime involved, and the difficulties and challenges that remain to be overcome.

BackgroundHistopathology image analysis is a gold standard for cancer recognition and diagnosis. Automatic analysis of histopathology images can help pathologists diagnose tumor and cancer subtypes, alleviating the workload of pathologists. There are two basic types of tasks in digital histopathology image analysis: image classification and image segmentation. Typical problems with histopathology images that hamper automatic analysis include complex clinical representations, limited quantities of training images in a dataset, and the extremely large size of singular images (usually up to gigapixels). The property of extremely large size for a single image also makes a histopathology image dataset be considered large-scale, even if the number of images in the dataset is limited.ResultsIn this paper, we propose leveraging deep convolutional neural network (CNN) activation features to perform classification, segmentation and visualization in large-scale tissue histopathology images. Our framework transfers features extracted from CNNs trained by a large natural image database, ImageNet, to histopathology images. We also explore the characteristics of CNN features by visualizing the response of individual neuron components in the last hidden layer. Some of these characteristics reveal biological insights that have been verified by pathologists. According to our experiments, the framework proposed has shown state-of-the-art performance on a brain tumor dataset from the MICCAI 2014 Brain Tumor Digital Pathology Challenge and a colon cancer histopathology image dataset.ConclusionsThe framework proposed is a simple, efficient and effective system for histopathology image automatic analysis. We successfully transfer ImageNet knowledge as deep convolutional activation features to the classification and segmentation of histopathology images with little training data. CNN features are significantly more powerful than expert-designed features.

Single-field-of-view (FoV) histopathological image classification is vital for colorectal cancer (CRC) diagnosis in mid- to low-tier hospitals lacking whole-slide imaging (WSI) scanners and storage, yet suffers from severe class imbalance and degraded performance. To address this, we propose a dual-level imbalance mitigation (DIM) framework integrating data-level and algorithm-level approaches. Specifically: (1) A global context generative adversarial network (GCGAN) generates realistic minority-class images for augmentation to balance the dataset. (2) A frequency-aware adaptive focal loss (FAFL) applies a frequency-aware offset and adaptive modulation to better separate overlapping classes. (3) A lightweight receptive field-based convolutional neural network (LRF-CNN) is trained under DIM to leverage both augmentation and loss modulation for improved classification. Extensive experiments on the single-FoV colorectal histopathology dataset demonstrate that DIM-equipped LRF-CNN outperforms five state-of-the-art models (SOTA) across multiple metrics. Furthermore, each DIM component enhances performance when applied individually to those SOTA models, and additional validation on six single-FoV histopathological datasets confirms the generalizability and effectiveness of the proposed DIM framework. Our code is available at https://github.com/Lingling-Yuan/DIM.

Fusion strategies for deep convolutional neural network representations in histopathological image classification

Breast cancer is currently one of the most diagnosed oncological diseases worldwide, with thousands of new cases per year. Early detection and identifying its progression are key to overcoming the mortality rate. A recurrent test, to determine how far the disease has spread throughout the patient’s body, is the histological analysis of the sentinel lymph node near the breast. Although an expert pathologist performs this, it is usually an exhausting and time-consuming task, with a high possibility of error. This work presents a method to detect breast cancer metastasis through histological imaging of sentinel lymph nodes using convolutional neural networks. In this study, the performance of three models DenseNet-121, DenseNet-169 and DenseNet-201 are tested and compared. Experimental results indicated that the accuracy, precision, sensitivity and specificity (97.93%, 97.4%, 97.48% and 98.24%) of DenseNet-201 could reduce pathologist errors during the diagnostic process or serve as a second opinion tool.

Histopathological images are widely used for the analysis of diseased (tumor) tissues and patient treatment selection. While the majority of microscopy image processing was previously done manually by pathologists, recent advances in computer vision allow for accurate recognition of lesion regions with deep learning-based solutions. Such models, however, usually require extensive annotated datasets for training, which is often not the case in the considered task, where the number of available patient data samples is very limited. To deal with this problem, we propose a novel DeepCMorph model pre-trained to learn cell morphology and identify a large number of different cancer types. The model consists of two modules: the first one performs cell nuclei segmentation and annotates each cell type, and is trained on a combination of 8 publicly available datasets to ensure its high generalizability and robustness. The second module combines the obtained segmentation map with the original microscopy image and is trained for the downstream task. We pre-trained this module on the Pan-Cancer TCGA dataset consisting of over 270K tissue patches extracted from 8736 diagnostic slides from 7175 patients. The proposed solution achieved a new state-ofthe-art performance on the dataset under consideration, detecting 32 cancer types with over 82% accuracy and outperforming all previously proposed solutions by more than 4%. We demonstrate that the resulting pre-trained model can be easily fine-tuned on smaller microscopy datasets, yielding superior results compared to the current top solutions and models initialized with ImageNet weights. The codes and pre-trained models presented in this paper are available at: https://github.com/aiff22/DeepCMorph

The computer-assisted classification of breast cancer histopathological image in the future is an essential method for the improvement of the diagnostic performance, thus reducing breast cancer deaths. Although deep convolutional neural networks (DCNNs) have made dramatic breakthroughs, many image classification tasks still remain challenging due to the insufficiency of training data and the lack of the ability to focus on improving classification efficiency. To address these issues, we share a densely connected convolutional network (DenseNet) model that is composed of 3 layers DenseNet, pooling layer, and a classification layer for breast cancer classification in microscopic images. Each layer of DenseNet contains 4 dense blocks. Each dense block jointly uses the dense connection and a novel attention learning mechanisms to increase its ability for discriminative representation. Meanwhile, the transfer learning algorithm is applied to determine the model parameters to extract the features of the patient image that is performed. In order to ensure sufficient data volume, a data enhancement method based on the quad-tree principle is proposed for high-resolution images. On the other hand, the classification probability of each part after dicing is fused by three algorithms of addition, product, and maximum. We evaluated our DenseNet model on the BreastKHis dataset. Our results indicate that the DenseNet model and data enhancement method we adopted can adaptively focus on the study of breast cancer histopathological image classification, thus achieving the state-of-the-art performance in breast cancer classification. The results of the experiments are in terms of patient-level and image-level accuracy. The best recognition accuracy increased to 90.9%-92.5% and 89.3%-91.8%, respectively, compared with previous studies.