Cancer Histopathology Research Articles

Histopathological domain adaptation with generative adversarial networks: Bridging the domain gap between thyroid cancer histopathology datasets.

Deep learning techniques are increasingly being used to classify medical imaging data with high accuracy. Despite this, due to often limited training data, these models can lack sufficient generalizability to predict unseen test data, produced in different domains, with comparable performance. This study focuses on thyroid histopathology image classification and investigates whether a Generative Adversarial Network [GAN], trained with just 156 patient samples, can produce high quality synthetic images to sufficiently augment training data and improve overall model generalizability. Utilizing a StyleGAN2 approach, the generative network produced images with an Fréchet Inception Distance (FID) score of 5.05, matching state-of-the-art GAN results in non-medical domains with comparable dataset sizes. Augmenting the training data with these GAN-generated images increased model generalizability when tested on external data sourced from three separate domains, improving overall precision and AUC by 7.45% and 7.20% respectively compared with a baseline model. Most importantly, this performance improvement was observed on minority class images, tumour subtypes which are known to suffer from high levels of inter-observer variability when classified by trained pathologists.

Cardiotoxicity in Breast Cancer: Impact of Clinical Classifications and Treatment on Heart Health.

Cardio-oncology has become essential in addressing cardiovascular complications from cancer therapies. While advancements in treatments have improved survival rates, they also increase cardiovascular risks. This study evaluates the cardiotoxic effects of cytostatic treatments, examining the relationship between tumor characteristics, such as histopathology and TNM classification, and cardiovascular complications, aiming to improve cardiotoxicity prevention and management in oncology patients. We conducted a retrospective analysis of cancer patients treated with anthracyclines, HER2-targeted therapies, and radiotherapy. Cardiac function was monitored via echocardiography, focusing on global longitudinal strain and left ventricular ejection fraction (LVEF). Cardiac troponins and natriuretic peptides were measured to detect subclinical cardiotoxicity, with patients stratified by TNM cancer stage and histopathology. Our analysis identified a significant association between certain cytostatic treatments, such as anthracyclines and HER2-targeted therapies, and a reduction in LVEF, particularly in patients with advanced-stage cancer. Biomarker assessments indicated early signs of cardiotoxicity in patients before clinical symptoms emerged. The findings also demonstrated a higher prevalence of cardiovascular complications in patients with pre-existing risk factors. This study highlights the importance of personalized treatment protocols in minimizing cardiotoxicity and improving the quality of life for oncology patients. Regular cardiac monitoring, combined with the use of biomarkers, can help identify high-risk patients early, allowing for timely interventions. Future research should focus on optimizing cardioprotective strategies to mitigate the cardiovascular risks associated with modern cancer therapies. N/A (retrospective study).

A Novel Approach for Lung Cancer Segmentation and Classification Using MSF-Customized ResNet152

This study presents a novel Multi-Scale Fusion-based Customized ResNet152 (MSF-Customized ResNet152) model for lung cancer segmentation and classification, designed to address challenges in class imbalance and feature extraction. The model combines multi-scale feature fusion with depth-wise separable convolution to capture fine and broad image details, enhancing segmentation accuracy. To address the minority representation of cancerous regions, a weighted recall loss was integrated with cross-entropy loss, optimized via Grid Search, to increase sensitivity. Additionally, data augmentation techniques were applied to expand the cancerous image set, further improving detection robustness. The model was evaluated on two datasets: a chest CT-scan dataset and a lung and colon cancer histopathology dataset. Using a 5-fold cross-validation approach, the MSF-Customized ResNet152 achieved a classification accuracy of 98.5%, precision of 98.2%, specificity of 97.2%, F1-score of 97.4%, and Dice coefficient of 97.3%. Comparative analysis demonstrated the superiority of the combined recall and cross-entropy loss function over standard methods, validating the model's effectiveness in handling class imbalance.

Developed improved lion optimization for breast cancer classification using histopathology images

Breast cancer, a prevalent kind of cancer, is a major health problem among women. Researchers recently achieved categorization effectiveness of breast cancer (BC) detection in histopathology picture database using convolutional neural networks (CNNs) of medical image processing. Although CNN method parameter settings were complex, employing breast cancer histopathological database (BCHD) data for categorization was valued as expensive. This research used uniform experimental design (UED) to solve these issues and improved lion optimization (ILO) breast cancer histopathology image categorization. To optimize the variables at UED-ILO, a regression method was employed. According to the experimental data, the proposed approach of UED-ILO (uniform experimental design based improved lion optimization) variable optimization provided a categorization accuracy rate of 84.41%. Finally, the proposed approach can effectively increase classification accuracy, with results that outperform others of an equivalent nature.

In-context learning enables multimodal large language models to classify cancer pathology images

Medical image classification requires labeled, task-specific datasets which are used to train deep learning networks de novo, or to fine-tune foundation models. However, this process is computationally and technically demanding. In language processing, in-context learning provides an alternative, where models learn from within prompts, bypassing the need for parameter updates. Yet, in-context learning remains underexplored in medical image analysis. Here, we systematically evaluate the model Generative Pretrained Transformer 4 with Vision capabilities (GPT-4V) on cancer image processing with in-context learning on three cancer histopathology tasks of high importance: Classification of tissue subtypes in colorectal cancer, colon polyp subtyping and breast tumor detection in lymph node sections. Our results show that in-context learning is sufficient to match or even outperform specialized neural networks trained for particular tasks, while only requiring a minimal number of samples. In summary, this study demonstrates that large vision language models trained on non-domain specific data can be applied out-of-the box to solve medical image-processing tasks in histopathology. This democratizes access of generalist AI models to medical experts without technical background especially for areas where annotated data is scarce.

A moment-based pooling approach in convolutional neural networks for breast cancer histopathology image classification

A moment-based pooling approach in convolutional neural networks for breast cancer histopathology image classification

A Hybrid Deep Learning and Machine Learning Approach with Mobile-EfficientNet and Grey Wolf Optimizer for Lung and Colon Cancer Histopathology Classification.

Lung and colon cancers are among the most prevalent and lethal malignancies worldwide, underscoring the urgent need for advanced diagnostic methodologies. This study aims to develop a hybrid deep learning and machine learning framework for the classification of Colon Adenocarcinoma, Colon Benign Tissue, Lung Adenocarcinoma, Lung Benign Tissue, and Lung Squamous Cell Carcinoma from histopathological images. Current approaches primarily rely on the LC25000 dataset, which, due to image augmentation, lacks the generalizability required for real-time clinical applications. To address this, Contrast Limited Adaptive Histogram Equalization (CLAHE) was applied to enhance image quality, and 1000 new images from the National Cancer Institute GDC Data Portal were introduced into the Colon Adenocarcinoma, Lung Adenocarcinoma, and Lung Squamous Cell Carcinoma classes, replacing augmented images to increase dataset diversity. A hybrid feature extraction model combining MobileNetV2 and EfficientNetB3 was optimized using the Grey Wolf Optimizer (GWO), resulting in the Lung and Colon histopathological classification technique (MEGWO-LCCHC). Cross-validation and hyperparameter tuning with Optuna were performed on various machine learning models, including XGBoost, LightGBM, and CatBoost. The MEGWO-LCCHC technique achieved high classification accuracy, with the lightweight DNN model reaching 94.8%, LightGBM at 93.9%, XGBoost at 93.5%, and CatBoost at 93.3% on the test set. The findings suggest that our approach enhances classification performance and offers improved generalizability for real-world clinical applications. The proposed MEGWO-LCCHC framework shows promise as a robust tool in cancer diagnostics, advancing the application of AI in oncology.

Optimizing Deep Learning Models with Custom ReLU for Breast Cancer Histopathology Image Classification

Purpose: The prompt identification of breast cancer is crucial in preventing the considerable damage inflicted by this dangerous form of cancer, which is widely distributed across the globe. This study seeks to refine the efficacy of a deep learning-driven approach for the precise diagnosis of breast cancer by employing diverse bespoke Rectified Linear Units (ReLU) to improve the model's performance and reduce inaccuracies within the system. Method: This study focuses on analyzing a deep learning approach utilizing the BreakHis dataset with 7,909 images, incorporating changes to the ReLU activation function across different pre-trained CNN models. It then evaluates performance through measures such as accuracy, precision, recall, and F1-Score. Result: Based on our experiment results, show that the DenseNet201 models with a custom LeakyReLU excel beyond the typical ReLU, achieving the highest accuracy, recall, and F1-Score at 99.21%, 99.21%, and 99.11%, respectively. Simultaneously, ResNet152, utilizing LessNegativeReLU ( ), achieved the highest precision at 99.11%. The VGG11 model exhibited the most notable performance enhancement, with improvements ranging from 1.39% to 1.59%. Novelty: The research is original in optimizing a model for accurate breast cancer diagnosis. The proposed model is superior to the model utilizing the default activation function. This finding indicates that the study significantly enhances performance while effectively minimizing errors, thereby necessitating further exploration into the effectiveness of the customized activation function when applied to other medical imaging modalities.

Unsupervised Histopathological Sub-Image Analysis for Breast Cancer Diagnosis Using Variational Autoencoders, Clustering, and Supervised Learning

This paper presents an integrated approach to breast cancer diagnosis that combines unsupervised and supervised learning techniques. The method involves using a pre-trained VGG19 model to process sub-images from the BreaKHis dataset, divided into nine parts for comprehensive analysis. This will be followed by a complete description of the architecture and workings of the variational Autoencoder (VAE) used for unsupervised Learning. The encoder network maps the input features to lower dimensions, capturing the most essential information. VAE learns a compressed representation of sub-images, facilitating a more profound understanding of underlying patterns and structures. For this reason, we then employ k-means clustering on the encoded representation to find naturally occurring clusters in our data set comprising a histopathological image. Every single sub-image is later fed into the VGG19-SVM model for classification purposes. During magnification at 100x, this model has attained a fantastic accuracy rate of 98.56%. Combining unsupervised analysis with VAE/k-means clustering and supervised classification with VGG19/SVM can integrate information from both methods, thereby improving the accuracy and robustness of such a task as sub-image classification in breast cancer histopathology.

Classification of breast cancer histopathology images using a modified supervised contrastive learning method.

Deep neural networks have reached remarkable achievements in medical image processing tasks, specifically in classifying and detecting various diseases. However, when confronted with limited data, these networks face a critical vulnerability, often succumbing to overfitting by excessively memorizing the limited information available. This work addresses the challenge mentioned above by improving the supervised contrastive learning method leveraging both image-level labels and domain-specific augmentations to enhance model robustness. This approach integrates self-supervised pre-training with a two-stage supervised contrastive learning strategy. In the first stage, we employ a modified supervised contrastive loss that not only focuses on reducing false negatives but also introduces an elimination effect to address false positives. In the second stage, a relaxing mechanism is introduced that refines positive and negative pairs based on similarity, ensuring that only relevant image representations are aligned. We evaluate our method on the BreakHis dataset, which consists of breast cancer histopathology images, and demonstrate an increase in classification accuracy by 1.45% in the image level, compared to the state-of-the-art method. This improvement corresponds to 93.63% absolute accuracy, highlighting the effectiveness of our approach in leveraging properties of data to learn more appropriate representation space. The code implementation of this study is accessible on GitHub https://github.com/matinamehdizadeh/Breast-Cancer-Detection .

Data set for the reporting of lung cancer: recommendations from the International Collaboration on Cancer Reporting (ICCR).

Lung cancer is the leading cause of cancer related deaths worldwide, although some patients with early-stage disease can be cured with surgical resection. Standardised reporting of all clinically relevant pathological parameters is essential for best patient care and is also important for ongoing data collection and refinement of important pathological features that impact patient prognosis, staging and clinical care. Using the established International Collaboration on Cancer Reporting (ICCR) procedure, a representative international expert panel of nine lung pathologists as well as an oncologist was convened. Essential core elements and suggested non-core elements were identified for inclusion in the resected lung cancer pathology data set based on predetermined levels of evidence as well as consensus expert opinion. A lung cancer histopathology reporting guide was developed that includes relevant clinical, macroscopic, microscopic and ancillary testing. Critical review and discussion of current evidence was incorporated into the new data set including changes from the 2021 World Health Organisation (WHO) Classification of Thoracic Tumours, fifth edition, new requirements for grading invasive non-mucinous adenocarcinomas, assessment of response to neoadjuvant therapy and requirements for molecular testing in early-stage resected lung carcinomas. This ICCR data set represents incorporation of all relevant parameters for histology reporting of lung cancer resection specimens. Routine use of this data set is recommended for all pathology reporting of resected lung cancer and it is freely available worldwide on the ICCR website (https://www.iccr-cancer.org/datasets/published-datasets/). Widespread implementation will help to ensure consistent and comprehensive pathology reporting and data collection essential for lung cancer patient care, clinical trials and other research.

Adjuvant Therapy after Esophagectomy for Esophageal Cancer: Who Needs It?: Multi-institution Worldwide Observational Study.

Based on current practice guidelines, we hypothesized that most patients with esophageal cancer, particularly those with locally advanced cancer, would benefit from adjuvant therapy after esophagectomy versus esophagectomy alone. We sought to obtain a granular estimate of patient-level risk-adjusted survival for each therapeutic option by cancer histopathology and stage. Although esophagectomy alone is now an uncommon therapy for treating locally advanced esophageal cancer, the value of adjuvant therapy after esophagectomy is unknown. From 1970 to 2014, 22,123 consecutive patients from 33 centers on 6 continents (Worldwide Esophageal Cancer Collaboration) were diagnosed with biopsy-proven adenocarcinoma (n = 7526) or squamous cell carcinoma (n = 5625), of whom 10,873 received esophagectomy alone and 2278 additional adjuvant therapy. Random forests for survival and virtual-twin analyses were performed for all-cause mortality. For adenocarcinoma, adjuvant therapy was beneficial only in pT4NanyM0 cancers (6-8 month survival benefit) and in pTanyN3M0 cancers (4-8 month benefit); a survival decrement was observed in pT1-3N0M0 cancers, with no effect on TanyN1-2M0 cancers. In squamous cell carcinoma, there was a 4 to 21 month survival benefit for pT3-4N0M0 cancers and a 4 to 15 month survival benefit for pT2-4N1-3M0 cancers. Adjuvant therapy after esophagectomy appears to benefit most patients with node-positive squamous cell carcinoma, but for adenocarcinoma, its value is limited to deep cancers and to those with substantial nodal burden. Future studies of the role of adjuvant therapies should treat these 2 cancers differently, with guidelines reflecting the histopathologic-appropriate survival value of adjuvant therapy.

Performance analysis of breast cancer histopathology image classification using transfer learning models

Performance analysis of breast cancer histopathology image classification using transfer learning models

Few-shot learning based histopathological image classification of colorectal cancer

Background Colorectal cancer is a prevalent and deadly disease worldwide, posing significant diagnostic challenges. Traditional histopathologic images classification is often inefficient and subjective. Although some histopathologists use computer-aided diagnosis (CAD) to improve efficiency, these methods depend heavily on extensive data and specific annotations, limiting their development. To address these challenges, this paper proposes a method based on few-shot learning. Methods This study introduces a few-shot learning approach that combines transfer learning and contrastive learning to classify colorectal cancer histopathology images into benign and malignant categories. The model comprises modules for feature extraction, dimensionality reduction, and classification, trained using a combination of contrast loss and cross-entropy loss. In this paper, we detail the setup of hyperparameters:n-way, k-shot, β and the creation of support, query, and test datasets. Results Our method achieved over 98% accuracy on a query dataset with 35 samples per category using only 10 training samples per category. We documented the model’s loss, accuracy, and the confusion matrix of the results. Additionally, we employed the t-SNE algorithm to analyze and assess the model’s classification performance. Conclusion The proposed model demonstrates significant advantages in accuracy and minimal data dependency, performing robustly across all tested n-way, k-shot scenarios. It consistently achieved over 93% accuracy on comprehensive test datasets, including 1916 samples, confirming its high classification accuracy and strong generalization capability. Our research advances the use of few-shot learning in medical diagnostics, and also lays the groundwork for extending it to deal with rare, difficult-to-diagnose cases.

Colorectal cancer histopathology image analysis: A comparative study of prognostic values of automatically extracted morphometric nuclear features in multispectral and red-blue-green imagery.

Multispectral imaging (MSI) has been utilized to predict the prognosis of colorectal cancer (CRC) patients, however, our understanding of the prognostic value of nuclear morphological parameters of bright-field MSI in CRC is still limited. This study was designed to compare the efficiency of MSI and standard red-green-blue (RGB) images in predicting the prognosis of CRC. We compared the efficiency of MS and conventional RGB images on the quantitative assessment of hematoxylin-eosin (HE) stained histopathology images. A pipeline was developed using a pixel-wise support vector machine (SVM) classifier for gland-stroma segmentation, and a marker-controlled watershed algorithm was used for nuclei segmentation. The correlation between extracted morphological parameters and the five-year disease-free survival (5-DFS) was analyzed. Forty-seven nuclear morphological parameters were extracted in total. Based on Kaplan-Meier analysis, eight features derived from MS images and seven featured derived from RGB images were significantly associated with 5-DFS, respectively. Compared with RGB images, MSI showed higher accuracy, precision, and Dice index in nuclei segmentation. Multivariate analysis indicated that both integrated parameters 1 (factors negatively correlated with CRC prognosis including nuclear number, circularity, eccentricity, major axis length) and 2 (factors positively correlated with CRC prognosis including nuclear average area, area perimeter, total area/total perimeter ratio, average area/perimeter ratio) in MS images were independent prognostic factors of 5-DFS, in contrast with only integrated parameter 1 (P<0.001) in RGB images. More importantly, the quantification of HE-stained MS images displayed higher accuracy in predicting 5-DFS compared with RGB images (76.9% vs 70.9%). Quantitative evaluation of HE-stained MS images could yield more information and better predictive performance for CRC prognosis than conventional RGB images, thereby contributing to precision oncology.

Artificial Intelligence Algorithms and Their Current Role in the Identification and Comparison of Gleason Patterns in Prostate Cancer Histopathology: A Comprehensive Review.

The development of the Gleason grading system has proven to be an irreplaceable tool in prostate cancer diagnostics within urology. Despite the advancements and developments in diagnostics, there remains a discrepancy in the grading process among even the most experienced pathologists. AI algorithms have demonstrated potential in detecting cancer and assigning Gleason grades, offering a solution to the issue of significant variability among pathologists' evaluations. Our paper explores the evolving role of AI in prostate cancer histopathology, with a key focus on outcomes and the reliability of various AI algorithms for Gleason pattern assessment. We conducted a non-systematic review of the published literature to examine the role of artificial intelligence in Gleason pattern diagnostics. The PubMed and Google Scholar databases were searched to gather pertinent information about recent advancements in artificial intelligence and their impact on Gleason patterns. We found that AI algorithms are increasingly being used to identify Gleason patterns in prostate cancer, with recent studies showing promising advancements that surpass traditional diagnostic methods. These findings highlight AI's potential to be integrated into clinical practice, enhancing pathologists' workflows and improving patient outcomes. The inter-observer variability in Gleason grading has seen an improvement in efficiency with the implementation of AI. Pathologists using AI have reported successful outcomes, demonstrating its effectiveness as a supplementary tool. While some refinements are still needed before AI can be fully implemented in clinical practice, its positive impact is anticipated soon.

Ameliorated Fick’s law algorithm based multi-threshold medical image segmentation

Medical image segmentation is a critical and demanding step in medical image processing, which provides a solid foundation for subsequent medical image data extraction and analysis. Multi-threshold image segmentation, one of the most commonly used and specialized image segmentation techniques, limits its application to medical images because it requires demanding computational performance and is difficult to produce satisfactory segmentation results. To overcome the above problems, an ameliorated Fick's law algorithm (MsFLA) for multi-threshold image segmentation is developed in this paper. First, an optimized sine–cosine strategy is introduced to extend the molecular diffusion process to alleviate the problem of easily falling into local optima, thus improving the convergence accuracy of the Fick's law algorithm (FLA). Secondly, the introduction of local minimal value avoidance enriches the individual molecular information and enhances the local search ability, thus improving computational accuracy. In addition, the optimal neighborhood learning strategy is added to ensure a more careful and reasonable reliance on the optimal solution, thus reducing the chance of convergence of a local solution. The efficient optimization capability of MsFLA is comprehensively validated by comparing MsFLA with the original FLA and other algorithms in 23 classical benchmark functions. Finally, MsFLA is applied to image segmentation of grayscale images of COVID-19 and brain and color images of Lung and Colon cancer histopathology by using Cross entropy to validate its segmentation capability. The experimental results show that the MsFLA obtains the best segmentation results in three medical image cases compared to other comparison algorithms, which indicates that MsFLA can effectively solve the multi-threshold medical image segmentation problem.Graphical abstract

SP3.11 - Beyond the microscope – Unravelling the Significance of Colorectal Cancer Histopathology Request Forms

Abstract Accurate histopathology reporting of colorectal cancer resected specimens (CCRS) is crucial for optimal patient care. The adequacy of clinical information provided by surgeons influences the quality of documentation. This audit aims to identify if histopathology request forms for CCRS in our trust meets national standards, as published by The Royal College of Pathologists. A retrospective study encompassing 40 consecutive histopathology request forms following colorectal cancer resections was conducted over 4 months (June to September 2023), where we assessed local adherence to national guidelines. The guidelines address a range of information from the surgical details to the patient’s medical background. Among the 40 request forms, 87.5% (35) were completed by consultants and 12.5% (5) by registrars. The information predominantly included were details on tumour site (98%) and nature of the resection (98%). Details on surgical procedure through a diagram and method of approach, were provided in 45% (18) and 48% (19) of the forms. Conversely, details on the pre-operative stage of tumour and types of therapy given was discerned in less than 25% of the forms. Only 3% (1) included information regarding presence of iatrogenic specimen defect and if specimen included other tissues or organ. Alarmingly, none addressed patient hereditary conditions or presence inflammatory bowel disease. The current histopathology request forms for CCRS lack information to facilitate comprehensive histopathological analysis. Addressing this necessitates raising awareness on national guidelines available. Additionally, a new structured form has been implemented in our trust, with the aim of encouraging surgeons to provide more comprehensive information.

Breast cancer histopathology image classification using an ensemble of optimized pretrained models with a trainable ensemble strategy classifier

Breast cancer histopathology image classification using an ensemble of optimized pretrained models with a trainable ensemble strategy classifier

Type-specific human papillomavirus prevalence in women referred for colposcopy in Tehran.

Although several studies have been achieved on the frequency of the HPV types among women with cervical cancer in Iran, HPV-positive samples were in some cases directed to specific-primer genotyping of HPV 16 and 18. Therefore, the other HPV types are underestimated. Several studies have also reported a greater prevalence of HPV 16 in cervical cancer in Iran than in the world. To clarify these subjects, the distribution of HPV types in women referred for colposcopy in Tehran was investigated. In this cross-sectional study, a total of 148 cervical samples from women with normal, atypical squamous cells of undetermined significance, cervical intraepithelial neoplasia I-III, and invasive cervical cancer histopathology were included. HPV was detected by PCR assay and all HPV-positive specimens were subjected to direct nucleotide sequencing. Our results demonstrated that the total prevalence of HPV was 92.5%. The five most common HPV types were HPV 16 (49.3%), 18 (14.8%), 6 (7.4%), 31 (4.1%), and 11 (2.7%). About the histopathological stage, HPV 16 and 18 were dominant in all studied groups. In cervical cancer, HPV 16 and 18 were detected in 60% and 20% of cases, respectively. HPV 16 and 18 were the most common in cervical cancer in Iran.