Explore the vast range of titles available.

Transformers have been widely used in many computer vision challenges and have shown the capability of producing better results than convolutional neural networks (CNNs). Taking advantage of capturing long-range contextual information and learning more complex relations in the image data, Transformers have been used and applied to histopathological image processing tasks. In this survey, we make an effort to present a thorough analysis of the uses of Transformers in histopathological image analysis, covering several topics, from the newly built Transformer models to unresolved challenges. To be more precise, we first begin by outlining the fundamental principles of the attention mechanism included in Transformer models and other key frameworks. Second, we analyze Transformer-based applications in the histopathological imaging domain and provide a thorough evaluation of more than 100 research publications across different downstream tasks to cover the most recent innovations, including survival analysis and prediction, segmentation, classification, detection, and representation. Within this survey work, we also compare the performance of CNN-based techniques to Transformers based on recently published papers, highlight major challenges, and provide interesting future research directions. Despite the outstanding performance of the Transformer-based architectures in a number of papers reviewed in this survey, we anticipate that further improvements and exploration of Transformers in the histopathological imaging domain are still required in the future. We hope that this survey paper will give readers in this field of study a thorough understanding of Transformer-based techniques in histopathological image analysis, and an up-to-date paper list summary will be provided at https://github.com/S-domain/Survey-Paper .

Cancer seems to have a vast number of deaths due to its heterogeneity, aggressiveness, and significant propensity for metastasis. The predominant categories of cancer that may affect males and females and occur worldwide are colon and lung cancer. A precise and on-time analysis of this cancer can increase the survival rate and improve the appropriate treatment characteristics. An efficient and effective method for the speedy and accurate recognition of tumours in the colon and lung areas is provided as an alternative to cancer recognition methods. Earlier diagnosis of the disease on the front drastically reduces the chance of death. Machine learning (ML) and deep learning (DL) approaches can accelerate this cancer diagnosis, facilitating researcher workers to study a vast majority of patients in a limited period and at a low cost. This research presents Histopathological Imaging for the Early Detection of Lung and Colon Cancer via Ensemble DL (HIELCC-EDL) model. The HIELCC-EDL technique utilizes histopathological images to identify lung and colon cancer (LCC). To achieve this, the HIELCC-EDL technique uses the Wiener filtering (WF) method for noise elimination. In addition, the HIELCC-EDL model uses the channel attention Residual Network (CA-ResNet50) model for learning complex feature patterns. Moreover, the hyperparameter selection of the CA-ResNet50 model is performed using the tuna swarm optimization (TSO) technique. Finally, the detection of LCC is achieved by using the ensemble of three classifiers such as extreme learning machine (ELM), competitive neural networks (CNNs), and long short-term memory (LSTM). To illustrate the promising performance of the HIELCC-EDL model, a complete set of experimentations was performed on a benchmark dataset. The experimental validation of the HIELCC-EDL model portrayed a superior accuracy value of 99.60% over recent approaches.

Prompt diagnosis of breast malignancy is crucial for treatment and patient survival. Computer-aided diagnosis (CAD) technology can improve efficiency, accuracy, and treatment options. The existing algorithms for classifying breast cancer histopathological images have limitations, including high parameter counts, ineffective extraction of global features, and substantial time costs, which result in the loss of valuable information. This study proposes a robust Multi-Modal Feature Fusion Network for Histopathology (MFF-HistoNet) to address the multi-grading challenges of breast image and significantly boost diagnostic accuracy. MFF-HistoNet combines a CNN and a Quantum Tensor Network (QTN), which reduces model parameters through parameter compression, enabling deeper global features. The data enhancement method ensures a balanced training set and minimizes color interference. The GLCM method is fused with LBP and Gabor filtering to obtain local cell shape characteristics of histopathological images in space and different scales and directions. Leveraging the BreaKHis dataset, MFF-HistoNet differentiates between eight breast cancer subtypes and reduces model complexity while preserving the ability to capture vital spatial relationships, thus enhancing computational efficiency. The MFF-HistoNet algorithms reveal the benchmark performance, achieving impressive accuracy of 98.8% at the image level and 98.4% at the patient level under 100 × magnification and 98.1% and 98.9% under 40 × magnification, outperforming existing models and reducing resource requirements. The Grad-CAM method proves the fusion model's reliability and interoperability, showing its firm resolution and good performance. The proposed model codes are publicly available at: https://github.com/tmsherazi/MFF-HistoNet with DOI: https://doi.org/10.5281/zenodo.14808037 .

Gastric cancer is the fifth most common and fourth deadliest cancer worldwide, with a bleak 5-year survival rate of about 20%. Despite significant research into its pathobiology, prognostic predictability remains insufficient due to pathologists’ heavy workloads and the potential for diagnostic errors. Consequently, there is a pressing need for automated and precise histopathological diagnostic tools. This study leverages Machine Learning and Deep Learning techniques to classify histopathological images into healthy and cancerous categories. By utilizing both handcrafted and deep features and shallow learning classifiers on the GasHisSDB dataset, we conduct a comparative analysis to identify the most effective combinations of features and classifiers for differentiating normal from abnormal histopathological images without employing fine-tuning strategies. Our methodology achieves an accuracy of 95% with the SVM classifier, underscoring the effectiveness of feature fusion strategies. Additionally, cross-magnification experiments produced promising results with accuracies close to 80% and 90% when testing the models on unseen testing images with different resolutions.

Cancer prognosis is of essential interest, and extensive research has been conducted searching for biomarkers with prognostic power. Recent studies have shown that both omics profiles and histopathological imaging features have prognostic power. There are also studies exploring integrating the two types of measurements for prognosis modeling. However, there is a lack of study rigorously examining whether omics measurements have independent prognostic power conditional on histopathological imaging features, and vice versa. In this article, we adopt a rigorous statistical testing framework and test whether an individual gene expression measurement can improve prognosis modeling conditional on high-dimensional imaging features, and a parallel analysis is conducted reversing the roles of gene expressions and imaging features. In the analysis of The Cancer Genome Atlas (TCGA) lung adenocarcinoma and liver hepatocellular carcinoma data, it is found that multiple individual genes, conditional on imaging features, can lead to significant improvement in prognosis modeling; however, individual imaging features, conditional on gene expressions, only offer limited prognostic power. Being among the first to examine the independent prognostic power, this study may assist better understanding the “connectedness” between omics profiles and histopathological imaging features and provide important insights for data integration in cancer modeling.

The use of artificial intelligence has shown good performance in the medical imaging area, in particular the deep learning methods based on convolutional neural networks for classification, detection, and/or segmentation tasks. The task addressed in this research work is the segmentation of amastigote nests from histological microphotographs in the study of Trypanosoma cruzi infection (Chagas disease) implementing a U-Net convolutional network architecture. For the nests’ segmentation, a U-Net architecture was trained on histological images of an acute-stage murine experimental model performing a 5-fold cross-validation, while the final tests were carried out with data unseen by the U-Net from three image groups of different experimental models. During the training stage, the obtained results showed an average accuracy of 98.19 ± 0.01, while in the case of the final tests, an average accuracy of 99.9 ± 0.1 was obtained for the control group, as well as 98.8 ± 0.9 and 99.1 ± 0.8 for two infected groups; in all cases, high sensitivity and specificity were observed in the results. We can conclude that the use of a U-Net architecture proves to be a relevant tool in supporting the diagnosis and analysis of histological images for the study of Chagas disease.

INTRODUCTION: In the field of medical imaging, accurate categorization of lung tissue is essential for timely diagnosis and management of lung-related conditions, including cancer. Deep Learning (DL) methodologies have revolutionized this domain, promising improved precision and effectiveness in diagnosing ailments based on image analysis. This research delves into the application of DL models for classifying lung tissue, particularly focusing on histopathological imagery. OBJECTIVES: The primary objective of this study is to explore the deployment of DL models for the classification of lung tissue, emphasizing histopathological images. The research aims to assess the performance of various DL models in accurately distinguishing between different classes of lung tissue, including benign tissue, lung adenocarcinoma, and lung squamous cell carcinoma. METHODS: A dataset comprising 9,000 histopathological images of lung tissue was utilized, sourced from HIPAA compliant and validated sources. The dataset underwent augmentation to ensure diversity and robustness. The images were categorized into three distinct classes and balanced before being split into training, validation, and testing sets. Six DL models - DenseNet201, EfficientNetB7, EfficientNetB5, Vgg19, Vgg16, and Alexnet - were trained and evaluated on this dataset. Performance assessment was conducted based on precision, recall, F1-score for each class, and overall accuracy. RESULTS: The results revealed varying performance levels among the DL models, with EfficientNetB5 achieving perfect scores across all metrics. This highlights the capability of DL in improving the accuracy of lung tissue classification, which holds promise for enhancing diagnosis and treatment outcomes in lung-related conditions. CONCLUSION: This research significantly contributes to understanding the effective utilization of DL models in medical imaging, particularly for lung tissue classification. It emphasizes the critical role of a diverse and balanced dataset in developing robust and accurate models. The insights gained from this study lay the groundwork for further exploration into refining DL methodologies for medical imaging applications, with a focus on improving diagnostic accuracy and ultimately, patient outcomes.

Histopathological imaging has been routinely conducted in cancer diagnosis and recently used for modeling other cancer outcomes/phenotypes such as prognosis. Clinical/environmental factors have long been extensively used in cancer modeling. However, there is still a lack of study exploring possible interactions of histopathological imaging features and clinical/environmental risk factors in cancer modeling. In this article, we explore such a possibility and conduct both marginal and joint interaction analysis. Novel statistical methods, which are “borrowed” from gene–environment interaction analysis, are employed. Analysis of The Cancer Genome Atlas (TCGA) lung adenocarcinoma (LUAD) data is conducted. More specifically, we examine a biomarker of lung function as well as overall survival. Possible interaction effects are identified. Overall, this study can suggest an alternative way of cancer modeling that innovatively combines histopathological imaging and clinical/environmental data.

Adenocarcinoma, squamous cell carcinoma, and small cell carcinoma represent about 95% of lung tumours. However, the lung is the site of numerous types of tumours that may have an epithelial, mesenchymal, neuroendocrine, or lymphohematopoietic origin. With minor exceptions, both the clinical manifestations and the imaging characteristics are non-specific; many of the low-incidence tumours have common features with the high-incidence tumours. This article presents a group of low-incidence pulmonary tumours that pose multiple difficulties in terms of diagnosis due to non-specific symptomatology and non-specific imaging aspect. This article aims to correlate the histological data with imaging aspects to facilitate diagnostics. Because these tumours are rare and because they present in a variety of forms, problems may occur when establishing a diagnosis and trying to predict their behaviour. It is challenging to differentiate common lung tumours from rare ones based on clinical, radiological, or histological features. Only the presence of the imaging particularities, such as the location of the lesion, the association with certain patterns (appearance of ground glass, the “halo” sign, the presence of calcifications), and the histological/immunohisto-chemical profile can lead to the establishment of a correct diagnosis.

We aimed to evaluate whether white and gray matter microstructure changes observed with magnetic resonance imaging (MRI)-based diffusion tensor imaging (DTI) can be used to reflect the progression of chronic brain trauma. The MRI-DTI parameters, neuropathologic changes, and behavioral performance of adult male Wistar rats that underwent moderate (2.1 atm on day “0”) or repeated mild (1.5 atm on days “0” and “2”) traumatic brain injury (TBI or rmTBI) or sham operation were evaluated at 7 days, 14 days, and 1–9 months after surgery. Neurobehavioral tests showed that TBI causes long-term motor, cognitive and neurological deficits, whereas rmTBI results in more significant deficits in these paradigms. Both histology and MRI show that rmTBI causes more significant changes in brain lesion volumes than TBI. In vivo DTI further reveals that TBI and rmTBI cause persistent microstructural changes in white matter tracts (such as the body of the corpus callosum, splenium of corpus callus, internal capsule and/or angular bundle) of both two hemispheres. Luxol fast blue measurements reveal similar myelin loss (as well as reduction in white matter thickness) in ipsilateral and contralateral hemispheres as observed by DTI analysis in injured rats. These data indicate that the disintegration of microstructural changes in white and gray matter parameters analyzed by MRI-DTI can serve as noninvasive and reliable markers of structural and functional level alterations in chronic TBI.