Histopathological Images Research Articles

Improving breast cancer classification using histopathology images through deep learning

Improving breast cancer classification using histopathology images through deep learning

A deep-learning framework to predict cancer treatment response from histopathology images through imputed transcriptomics.

Advances in artificial intelligence have paved the way for leveraging hematoxylin and eosin-stained tumor slides for precision oncology. We present ENLIGHT-DeepPT, an indirect two-step approach consisting of (1) DeepPT, a deep-learning framework that predicts genome-wide tumor mRNA expression from slides, and (2) ENLIGHT, which predicts response to targeted and immune therapies from the inferred expression values. We show that DeepPT successfully predicts transcriptomics in all 16 The Cancer Genome Atlas cohorts tested and generalizes well to two independent datasets. ENLIGHT-DeepPT successfully predicts true responders in five independent patient cohorts involving four different treatments spanning six cancer types, with an overall odds ratio of 2.28 and a 39.5% increased response rate among predicted responders versus the baseline rate. Notably, its prediction accuracy, obtained without any training on the treatment data, is comparable to that achieved by directly predicting the response from the images, which requires specific training on the treatment evaluation cohorts.

Utilizing a Pathomics Biomarker to Predict the Effectiveness of Bevacizumab in Ovarian Cancer Treatment.

The purpose of this investigation is to develop and initially assess a quantitative image analysis scheme that utilizes histopathological images to predict the treatment effectiveness of bevacizumab therapy in ovarian cancer patients. As a widely accessible diagnostic tool, histopathological slides contain copious information regarding underlying tumor progression that is associated with tumor prognosis. However, this information cannot be readily identified by conventional visual examination. This study utilizes novel pathomics technology to quantify this meaningful information for treatment effectiveness prediction. Accordingly, a total of 9828 features were extracted from segmented tumor tissue, cell nuclei, and cell cytoplasm, which were categorized into geometric, intensity, texture, and subcellular structure features. Next, the best performing features were selected as the input for SVM (support vector machine)-based prediction models. These models were evaluated on an open dataset containing a total of 78 patients and 288 whole slides images. The results indicated that the sufficiently optimized, best-performing model yielded an area under the receiver operating characteristic (ROC) curve of 0.8312. When examining the best model's confusion matrix, 37 and 25 cases were correctly predicted as responders and non-responders, respectively, achieving an overall accuracy of 0.7848. This investigation initially validated the feasibility of utilizing pathomics techniques to predict tumor responses to chemotherapy at an early stage.

Vision transformer based convolutional neural network for breast cancer histopathological images classification

Vision transformer based convolutional neural network for breast cancer histopathological images classification

Internalisation of immunoglobulin light chains by cardiomyocytes in AL amyloidosis: what can biopsies tell us?

Background Cardiac involvement in systemic light chain amyloidosis (AL) leads to chronic heart failure and is a major prognosis factor. Severe cellular defects are provoked in cardiac cells by tissue-deposited amyloid fibrils of misfolded free immunoglobulin light chains (LCs) and their prefibrillar oligomeric precursors. Objective Understanding the molecular mechanisms behind cardiac cell cytotoxicity is necessary to progress in therapy and to improve patient management. One key question is how extracellularly deposited molecules exert their toxic action inside cardiac cells. Here we searched for direct evidence of amyloid LC uptake by cardiomyocytes in patient biopsies. Methods We immunolocalized LCs in cardiac biopsies from four AL cardiac amyloidosis patients and analysed histopathological images by high resolution confocal microscopy and 3D image reconstruction. Results We show, for the first time directly in patient tissue, the presence of LCs inside cardiomyocytes, and report their proximity to nuclei and to caveolin-3-rich areas. Our observations point to macropinocytosis as a probable mechanism of LC uptake. Conclusions Internalisation of LCs occurs in patient cardiomyocytes. This event could have important consequences for the pathogenesis of the cardiac disease by enabling interactions between amyloid molecules and cellular organelles inducing specific signalling pathways, and might bring new insight regarding treatment.

Challenges of Breast Cancer Detection Based on Histopathology Images

Challenges of Breast Cancer Detection Based on Histopathology Images

Attention-Based Deep Learning Approach for Breast Cancer Histopathological Image Multi-Classification.

Breast cancer diagnosis from histopathology images is often time consuming and prone to human error, impacting treatment and prognosis. Deep learning diagnostic methods offer the potential for improved accuracy and efficiency in breast cancer detection and classification. However, they struggle with limited data and subtle variations within and between cancer types. Attention mechanisms provide feature refinement capabilities that have shown promise in overcoming such challenges. To this end, this paper proposes the Efficient Channel Spatial Attention Network (ECSAnet), an architecture built on EfficientNetV2 and augmented with a convolutional block attention module (CBAM) and additional fully connected layers. ECSAnet was fine-tuned using the BreakHis dataset, employing Reinhard stain normalization and image augmentation techniques to minimize overfitting and enhance generalizability. In testing, ECSAnet outperformed AlexNet, DenseNet121, EfficientNetV2-S, InceptionNetV3, ResNet50, and VGG16 in most settings, achieving accuracies of 94.2% at 40×, 92.96% at 100×, 88.41% at 200×, and 89.42% at 400× magnifications. The results highlight the effectiveness of CBAM in improving classification accuracy and the importance of stain normalization for generalizability.

Stain SAN: simultaneous augmentation and normalization for histopathology images.

We address the need for effective stain domain adaptation methods in histopathology to enhance the performance of downstream computational tasks, particularly classification. Existing methods exhibit varying strengths and weaknesses, prompting the exploration of a different approach. The focus is on improving stain color consistency, expanding the stain domain scope, and minimizing the domain gap between image batches. We introduce a new domain adaptation method, Stain simultaneous augmentation and normalization (SAN), designed to adjust the distribution of stain colors to align with a target distribution. Stain SAN combines the merits of established methods, such as stain normalization, stain augmentation, and stain mix-up, while mitigating their inherent limitations. Stain SAN adapts stain domains by resampling stain color matrices from a well-structured target distribution. Experimental evaluations of cross-dataset clinical estrogen receptor status classification demonstrate the efficacy of Stain SAN and its superior performance compared with existing stain adaptation methods. In one case, the area under the curve (AUC) increased by 11.4%. Overall, our results clearly show the improvements made over the history of the development of these methods culminating with substantial enhancement provided by Stain SAN. Furthermore, we show that Stain SAN achieves results comparable with the state-of-the-art generative adversarial network-based approach without requiring separate training for stain adaptation or access to the target domain during training. Stain SAN's performance is on par with HistAuGAN, proving its effectiveness and computational efficiency. Stain SAN emerges as a promising solution, addressing the potential shortcomings of contemporary stain adaptation methods. Its effectiveness is underscored by notable improvements in the context of clinical estrogen receptor status classification, where it achieves the best AUC performance. The findings endorse Stain SAN as a robust approach for stain domain adaptation in histopathology images, with implications for advancing computational tasks in the field.

Synthetic Genitourinary Image Synthesis via Generative Adversarial Networks: Enhancing Artificial Intelligence Diagnostic Precision.

In the realm of computational pathology, the scarcity and restricted diversity of genitourinary (GU) tissue datasets pose significant challenges for training robust diagnostic models. This study explores the potential of Generative Adversarial Networks (GANs) to mitigate these limitations by generating high-quality synthetic images of rare or underrepresented GU tissues. We hypothesized that augmenting the training data of computational pathology models with these GAN-generated images, validated through pathologist evaluation and quantitative similarity measures, would significantly enhance model performance in tasks such as tissue classification, segmentation, and disease detection. To test this hypothesis, we employed a GAN model to produce synthetic images of eight different GU tissues. The quality of these images was rigorously assessed using a Relative Inception Score (RIS) of 1.27 ± 0.15 and a Fréchet Inception Distance (FID) that stabilized at 120, metrics that reflect the visual and statistical fidelity of the generated images to real histopathological images. Additionally, the synthetic images received an 80% approval rating from board-certified pathologists, further validating their realism and diagnostic utility. We used an alternative Spatial Heterogeneous Recurrence Quantification Analysis (SHRQA) to assess the quality of prostate tissue. This allowed us to make a comparison between original and synthetic data in the context of features, which were further validated by the pathologist's evaluation. Future work will focus on implementing a deep learning model to evaluate the performance of the augmented datasets in tasks such as tissue classification, segmentation, and disease detection. This will provide a more comprehensive understanding of the utility of GAN-generated synthetic images in enhancing computational pathology workflows. This study not only confirms the feasibility of using GANs for data augmentation in medical image analysis but also highlights the critical role of synthetic data in addressing the challenges of dataset scarcity and imbalance. Future work will focus on refining the generative models to produce even more diverse and complex tissue representations, potentially transforming the landscape of medical diagnostics with AI-driven solutions.

Development of high-quality artificial intelligence for computer-aided diagnosis in determining subtypes of colorectal cancer.

There are no previous studies in which computer-aided diagnosis (CAD) diagnosed colorectal cancer (CRC) subtypes correctly. In this study, we developed an original CAD for the diagnosis of CRC subtypes. Pretraining for the CAD based on ResNet was performed using ImageNet and five open histopathological pretraining image datasets (HiPreD) containing 3 million images. In addition, sparse attention was introduced to improve the CAD compared to other attention networks. One thousand and seventy-two histopathological images from 29 early CRC cases at Kyoto Prefectural University of Medicine from 2019 to 2022 were collected (857 images for training and validation, 215 images for test). All images were annotated by a qualified histopathologist for segmentation of normal mucosa, adenoma, pure well-differentiated adenocarcinoma (PWDA), and moderately/poorly differentiated adenocarcinoma (MPDA). Diagnostic ability including dice sufficient coefficient (DSC) and diagnostic accuracy were evaluated. Our original CAD, named Colon-seg, with the pretraining of both HiPreD and ImageNET showed a better DSC (88.4%) compared to CAD without both pretraining (76.8%). Regarding the attentional mechanism, Colon-seg with sparse attention showed a better DSC (88.4%) compared to other attentional mechanisms (dual: 79.7%, ECA: 80.7%, shuffle: 84.7%, SK: 86.9%). In addition, the DSC of Colon-seg (88.4%) was better than other types of CADs (TransUNet: 84.7%, MultiResUnet: 86.1%, Unet++: 86.7%). The diagnostic accuracy of Colon-seg for each histopathological type was 94.3% for adenoma, 91.8% for PWDA, and 92.8% for MPDA. A deep learning-based CAD for CRC subtype differentiation was developed with pretraining and fine-tuning of abundant histopathological images.

MIST: Multi-instance selective transformer for histopathological subtype prediction

Accurate histopathological subtype prediction is clinically significant for cancer diagnosis and tumor microenvironment analysis. However, achieving accurate histopathological subtype prediction is a challenging task due to (1) instance-level discrimination of histopathological images, (2) low inter-class and large intra-class variances among histopathological images in their shape and chromatin texture, and (3) heterogeneous feature distribution over different images. In this paper, we formulate subtype prediction as fine-grained representation learning and propose a novel multi-instance selective transformer (MIST) framework, effectively achieving accurate histopathological subtype prediction. The proposed MIST designs an effective selective self-attention mechanism with multi-instance learning (MIL) and vision transformer (ViT) to adaptive identify informative instances for fine-grained representation. Innovatively, the MIST entrusts each instance with different contributions to the bag representation based on its interactions with instances and bags. Specifically, a SiT module with selective multi-head self-attention (S-MSA) is well-designed to identify the representative instances by modeling the instance-to-instance interactions. On the contrary, a MIFD module with the information bottleneck is proposed to learn the discriminative fine-grained representation for histopathological images by modeling instance-to-bag interactions with the selected instances. Substantial experiments on five clinical benchmarks demonstrate that the MIST achieves accurate histopathological subtype prediction and obtains state-of-the-art performance with an accuracy of 0.936. The MIST shows great potential to handle fine-grained medical image analysis, such as histopathological subtype prediction in clinical applications.

Dual-stream multi-dependency graph neural network enables precise cancer survival analysis

Histopathology image-based survival prediction aims to provide a precise assessment of cancer prognosis and can inform personalized treatment decision-making in order to improve patient outcomes. However, existing methods cannot automatically model the complex correlations between numerous morphologically diverse patches in each whole slide image (WSI), thereby preventing them from achieving a more profound understanding and inference of the patient status. To address this, here we propose a novel deep learning framework, termed dual-stream multi-dependency graph neural network (DM-GNN), to enable precise cancer patient survival analysis. Specifically, DM-GNN is structured with the feature updating and global analysis branches to better model each WSI as two graphs based on morphological affinity and global co-activating dependencies. As these two dependencies depict each WSI from distinct but complementary perspectives, the two designed branches of DM-GNN can jointly achieve the multi-view modeling of complex correlations between the patches. Moreover, DM-GNN is also capable of boosting the utilization of dependency information during graph construction by introducing the affinity-guided attention recalibration module as the readout function. This novel module offers increased robustness against feature perturbation, thereby ensuring more reliable and stable predictions. Extensive benchmarking experiments on five TCGA datasets demonstrate that DM-GNN outperforms other state-of-the-art methods and offers interpretable prediction insights based on the morphological depiction of high-attention patches. Overall, DM-GNN represents a powerful and auxiliary tool for personalized cancer prognosis from histopathology images and has great potential to assist clinicians in making personalized treatment decisions and improving patient outcomes.

AI-based histopathology image analysis reveals a distinct subset of endometrial cancers

Endometrial cancer (EC) has four molecular subtypes with strong prognostic value and therapeutic implications. The most common subtype (NSMP; No Specific Molecular Profile) is assigned after exclusion of the defining features of the other three molecular subtypes and includes patients with heterogeneous clinical outcomes. In this study, we employ artificial intelligence (AI)-powered histopathology image analysis to differentiate between p53abn and NSMP EC subtypes and consequently identify a sub-group of NSMP EC patients that has markedly inferior progression-free and disease-specific survival (termed ‘p53abn-like NSMP’), in a discovery cohort of 368 patients and two independent validation cohorts of 290 and 614 from other centers. Shallow whole genome sequencing reveals a higher burden of copy number abnormalities in the ‘p53abn-like NSMP’ group compared to NSMP, suggesting that this group is biologically distinct compared to other NSMP ECs. Our work demonstrates the power of AI to detect prognostically different and otherwise unrecognizable subsets of EC where conventional and standard molecular or pathologic criteria fall short, refining image-based tumor classification. This study’s findings are applicable exclusively to females.

Deep self-reconstruction driven joint nonnegative matrix factorization model for identifying multiple genomic imaging associations in complex diseases

ObjectiveComprehensive analysis of histopathology images and transcriptomics data enables the identification of candidate biomarkers and multimodal association patterns. Most existing multimodal data association studies are derived from extensions of the joint nonnegative matrix factorization model for identifying complex data associations, which can make full use of clinical prior information. However, the raw data were usually taken as the input without considering the underlying complex multi-subspace structure, influencing the subsequent integration analysis results. MethodsThis study proposed a deep-self reconstructed joint nonnegative matrix factorization (DSRJNMF) model to use self-expressive properties to reconstruct the raw data to characterize the similarity structure associated with clinical labels. Then, the sparsity, orthogonality, and regularization constraints constructed from prior information are added to the DSRJNMF model to determine the sparse set of biologically relevant features across modalities. ResultsThe algorithm has been applied to identify the imaging genetic association of triple negative breast cancer (TNBC). Multilevel experimental results demonstrate that the proposed algorithm better estimates potential associations between pathological image features and miRNA-gene and identifies consistent multimodal imaging genetic biomarkers to guide the interpretation of TNBC. ConclusionThe propose method provides a novel idea of data association analysis oriented to complex diseases.

Implementation of Improved U-Net and Optimized XGBoost-SVM Classifier for Early Detection of Masses and Microcalcifications in Breast

Purpose: In contemporary time, breast cancer has been extensively found among women at a global rate of 24.2% as of 2018. This exposes the significant and imperative need for detecting masses and micro calcifications in the breast to avoid death rates. The existing histopathological images have gained a golden standard considering they would afford reliable results. However, these images have been entangled with various complexities including insufficient image contrast, noise, and misdiagnosis. These pitfalls might negatively impact the detection rate for which automatic recognition has become vital. With the momentous evolvement of Machine Learning (ML) and Deep Learning (DL), various researchers have endeavoured to consider ML and DL for accomplishing this prediction. However, they need to improve in accuracy rate due to ineffective feature extraction, and most studies averted to consider segmentation. Hence, this study regards accomplishing a high classification and segmentation process. Materials and Methods: The study proposes Modified Weight Updated Convolutional Block-U-Net (MWu-Conv-U-Net) to handle the image dimensions optimally. In this case, U-Net based model is regarded by improvising it with the inclusion of an additional convolutional layer in individual encoder-decoder. Further, the study proposes an Optimized Weight Updated eXtreme Gradient Boosting-Support Vector Machine (OWu-XGBoost-SVM) for determining the optimal gradient, which would eventually enhance the prediction rate. Results: The overall performance is assessed through performance metrics to confirm its effectiveness in classifying and segmenting the Breast Cancer Histopathological (BACH) image dataset. Comparison is undertaken with conventional systems in accordance with metrics (recall, F1-score, precision, and accuracy). The results revealed the efficacy of the proposed system with 99% accuracy, 99% F1-score, 99% precision, and 99% recall. Conclusion: High accuracy procured through the analysis reveals its efficacy and hence it could be applicable for real-time execution for assisting medical experts in early breast cancer prognosis.

Stochastic Class‐Attention Net to Detect the Breast Carcinoma Subtypes With Test Time Augmentation

ABSTRACTDespite advances in medical sciences, breast cancer remains a deadly disease globally, primarily affecting women. Fortunately, studies claim that breast cancer is treatable if diagnosed early. Late diagnoses have poor prognoses and can affect the patient's quality of life. Therefore, a significant research body is dedicated to establishing and identifying the disease at an initial stage. Deep learning (DL) techniques are garnering attention for aiding medical professionals in detecting this disease using histopathology (HP) image modality. The heterogeneous nature of this disease subtypes results in the imbalances of benign and malignant subtypes. From a DL point of view, this becomes an imbalanced problem deserving special care. Unfortunately, current DL‐based techniques do not fully address this issue and suffer from poor metrics and robustness. In this work, we present a DL‐based breast cancer automatic detection system (BCADS) using a novel architecture stochastic class‐attention net (SCAN). This technique performed better when combined with label smoothing and test time augmentation. This work outperforms the previously reported results for binary and multiclass on the BreaKHis dataset. Also, we validated our method on separate BACH and BCNB datasets to prove its effectiveness and clinical relevancy. We hope that the designed BCADS will help the treating doctor and pathologist in a meaningful way and thus help to reduce the impact of this deadly disease.

A survey on deep learning and machine learning techniques over histopathology image based Osteosarcoma Detection

A survey on deep learning and machine learning techniques over histopathology image based Osteosarcoma Detection

Enhanced Power Law Transformation for Histopathology Images of Breast Cancer

Different image enhancement techniques are applied to improve the visual quality of an image on a display device. Contrast stretching, intensity level slicing with and without background, histogram equalization, logarithmic transformation and power law transformation are some image enhancement techniques. Most of the research work focuses on adaptive gamma correcting factors for better visualization of extremely low contrast images, giving less importance to the constant for enhanced visualization. This research proposes an efficient and less complex enhanced power law transformation (EPLT) approach to improve the contrast of dimmed and extremely bright images. The approach is a quick way to compute the value of C, i.e. constant for enhanced visualization. For better picture quality, it is very important to determine C automatically and the gamma correcting factor. This technique offers a novel and unique perspective on image contrast manipulation. The proposed enhancement technique is experimented on histopathology images of breast cancer, bright images and extremely dark images. The average peak signal-to-noise ratio (PSNR) for clinical data and BreakHis dataset is high for the proposed method are 16.52487 and 17.69335 respectively. The average RMSE for clinical data and BreakHis dataset is low for the proposed method are 40.88251 and 44.2546 respectively. It is observed that the proposed method yields the most satisfactory contrast enhancements based on performance comparison with other state-of-art enhancement algorithms and works efficiently on all types of images.

Optimasi Convolutional Neural Networks untuk Deteksi Kanker Payudara menggunakan Arsitektur DenseNet

Breast cancer is a disease commonly suffered by women worldwide, ranking as the second-largest disease burden. In response to the urgent need for improved detection accuracy, Convolutional Neural Networks (CNNs) promise significant advancements. The objective of this research is to optimize the use of CNNs with the DenseNet architecture for breast cancer detection. The study employs quantitative methods, leveraging Deep Learning through CNNs. Mammography data is sourced from Kaggle, specifically the “Breast Histopathology Images” dataset. This dataset comprises 90,000 digital mammography images, which are preprocessed and divided proportionally for training, validation, and model testing. Research variables encompass CNN model parameters, training techniques, and the integration of imaging modalities to enhance breast cancer detection performance. The research focuses on processed mammography data, with accuracy and image quality as key evaluation metrics for breast cancer sample identification. Our findings demonstrate that the DenseNet architecture within CNNs achieves an impressive 92% accuracy in breast cancer detection. This remarkable performance signifies success in enhancing image quality and class prediction, aligning with the DenseNet architecture’s flow diagram. Ultimately, these results contribute significantly to effective breast cancer diagnosis by optimizing CNNs with the DenseNet architecture to improve image quality during breast cancer sampling.

Predictive breast cancer diagnosis using ensemble fuzzy model

Breast cancer continues to be a major global health challenge, necessitating reliable diagnostic methods for early detection and improved patient outcomes. This study introduces a novel ensemble fuzzy model for predictive breast cancer diagnosis, integrating multiple deep-learning classifiers with fuzzy logic to enhance decision-making. Traditional diagnostic approaches often struggle with the complexity and heterogeneity of breast cancer data, which this new model addresses through an innovative ensemble technique. The ensemble model combines the strengths of Inception-V4, Inception-ResNet, and Inception V3/V4 + BN, with fuzzy logic for adaptive priority assignment based on confidence scores. The method employs a re-parameterized Gompertz function to assign fuzzy ranks to constituent classifiers, allowing flexible fusion strategies. The proposed model is evaluated on two benchmark breast cancer datasets: the Digital Database for Screening Mammography (DDSM) and the Breast Cancer Histopathological Image Classification (BACH) dataset. It achieves high performance across key metrics, including accuracy, precision, recall, and F1 score, consistently outperforming individual classifiers. On the DDSM dataset, the ensemble fuzzy model attains an accuracy of 0.97, a recall of 0.93, a precision of 0.95, and an F1 score of 0.96. Similarly, on the BACH dataset, the proposed method records an accuracy of 97.05%, a recall of 99.31%, a precision of 95.44%, and an F1 score of 97.37%, demonstrating its robust capability to identify positive instances and maintain a balanced performance. These results highlight the potential of the ensemble fuzzy model to improve breast cancer diagnosis, offering a reliable solution to the inherent challenges in this field.