Academic literature on the topic 'Multi-stain segmentation'

Existing nuclei segmentation methods have obtained limited results with multi-center and multi-organ whole-slide images (WSIs) due to the use of different stains, scanners, overlapping, clumped nuclei, and the ambiguous boundary between adjacent cell nuclei. In an attempt to address these problems, we propose an efficient stain-aware nuclei segmentation method based on deep learning for multi-center WSIs. Unlike all related works that exploit a single-stain template from the dataset to normalize WSIs, we propose an efficient algorithm to select a set of stain templates based on stain clustering. Individual deep learning models are trained based on each stain template, and then, an aggregation function based on the Choquet integral is employed to combine the segmentation masks of the individual models. With a challenging multi-center multi-organ WSIs dataset, the experimental results demonstrate that the proposed method outperforms the state-of-art nuclei segmentation methods with aggregated Jaccard index (AJI) and F1-scores of 73.23% and 89.32%, respectively, while achieving a lower number of parameters.

Existing nuclei segmentation methods face challenges with hematoxylin and eosin (H&E) whole slide imaging (WSI) due to the variations in staining methods and nuclei shapes and sizes. Most existing approaches require a stain normalization step that may cause losing source information and fail to handle the inter-scanner feature instability problem. To mitigate these issues, this article proposes an efficient staining-invariant nuclei segmentation method based on self-supervised contrastive learning and an effective weighted hybrid dilated convolution (WHDC) block. In particular, we propose a staining-invariant encoder (SIE) that includes convolution and transformers blocks. We also propose the WHDC block allowing the network to learn multi-scale nuclei-relevant features to handle the variation in the sizes and shapes of nuclei. The SIE network is trained on five unlabeled WSIs datasets using self-supervised contrastive learning and then used as a backbone for the downstream nuclei segmentation network. Our method outperforms existing approaches in challenging multiple WSI datasets without stain color normalization.

Abstract Objective To investigate the characteristics of dysplasia in myelodysplastic syndrome (MDS). Methods Collect 716 samples of adult patients with abnormal blood routine but unclear cause between July 04, 2003 and March 14, 2007. Based on the gold standard of WHO MDS classification, all cases were detected on cytomorphological observation, cytochemical stain, bone marrow pathological study, cytogenetics, flow cytometry, and ect. The bone marrow cytological study on some abnormal hematopoietic cells has a diagnostic value to determine clonal or non-clonal diseases and assess sensitivity and specificity. Results In the complicated various dysplasia of hematopoietic cells, the following characteristics can be the main basis of cytomorphological diagnosis: One of granular Auer bodies, micronucleus (MN), or nuclear budding; Erythroid nuclear budding; Megakaryocytes presented in peripheral blood; Myeloblast or prorubricyte exhibited in peripheral blood; ringed sideroblasts>1%. The subordinate basis of cytomorphological diagnosis was as follows: Granular pseudo Pelger-H≥et anomaly, hard nucleus segmentation, unsynchronous development of nuclei, ring-shaped nuclei, and aggregation of nuclear chromatin. Erythroid multi-nuclei, odd nucleus, mother-daughter nucleus, nuclear fragmentation, vacuole, and anisocytosis; micromegakaryocytes. Conclusion Cytomorphologic is the base for the diagnosis of MDS, however, it presents certain limit, especially cytomorphological change does not possess specificity for early MDS, hereby, it requires to combine other detection methods.

Abstract Background: Non-small cell lung cancer (NSCLC) tissue is a valuable resource for diagnosis, treatment planning, and drug development. Current 2D histopathological techniques introduce under-sampling error (i.e., a single 5 um section represents 0.5% of a 1 mm thick biopsy), interobserver variability, and fail to capture the biology contained within the entire tissue sample. We have developed a suite of technologies to stain, chemically clarify, image, visualize, and analyze entire intact NSCLC tissue samples. Methods: Human NSCLC tissue, stored frozen in OCT, was fixed in 4% paraformaldehyde, stained with nuclear (TOPRO-3) and general protein (Eosin) fluorescent dyes, and optically cleared using a modified iDISCO protocol with ethyl cinnamate as the refractive index matching solution. Whole, intact tissue samples, roughly 1-5mm3 in volume, were imaged at 2 microns/pixel resolution with an open-top light-sheet microscope (3Di, Alpenglow Biosciences). Smaller regions of interest (ROIs) with key pathologic features were reimaged at higher resolution, 0.17 microns/pixel, to reveal subnuclear features and for cell typing. Visualization was performed using Aivia software. Results: NSCLC tissue samples were successfully imaged in 3D. Low resolution images (2 microns/pixel) were obtained within 4-31 minutes, depending on the tissue volume. The 3D distribution of cancer cells, immune cells, vessels, and fibrosis varied substantially throughout the volume of the tissue. Recognizable histologic features, including nests of tumor cells surrounded by vasculature and immune cells, were readily visualized. Squamous and adenocarcinoma with its subtypes (solid, acinar, lepidic, and micropapillary) morphologies were recognizable in 2D optical sections of the 3D datasets. Imaging quality degraded in tissue deeper than 1 mm due to light scattering. Conclusion: We assessed intact NSCLC tissue samples measuring up to 5 mm3 using our custom light-sheet microscope and tissue clearing techniques. This novel method enables us to visualize key features of NSCLC such as the tumor interfaces, tertiary lymphoid structures, vessels and fibrosis in the entire tissue sample, preventing under sampling error, and potentially enabling new biologic insights. Next steps include segmentation and quantification of key tissue structures such as tumor volume, immune cell distribution, and fibrosis/immune cell exclusion. This proof-of-concept study provides motivation for further investigation into the significance of 3D tissue features in NSCLC tissue samples. Citation Format: Alexandra Alvarsson, Carl Storey, Brandy Olin Pope, Caleb Stoltzfus, Robert Vierkant, Jessica Tufariello, Aaron Bungum, Julia Naso, Cheuk Ki Chan, Eric Edell, Christopher Hartley, Janani Reisenauer, Nicholas Reder. 3D assessment of the lung cancer microenvironment using multi-resolution open-top light-sheet microscopy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6624.

Abstract Introduction The bone marrow (BM) microenvironment consists of various cell types such as mesenchymal stromal cells, endothelial cells, osteoblastic cells and multiple immune cell types including mature myeloid cells and lymphocytes. Recent studies have shown that leukemias can create and maintain a leukemia-supporting BM microenvironment, and vice versa, a dysfunctional BM microenvironment can contribute to leukemia development and progression. Moreover, in tumors the microenvironment is often immunosuppressive and restrains effective anti-tumoral immune responses by adaptive and innate immunity. A better understanding of the precise localization of microenvironmental and immune cell types in intact tissue, and how they physically interact with each other and with tumor cells, will improve our understanding of the mechanisms by which cancer reprograms its microenvironment and may form the basis for novel immunotherapies. Methods CO-Detection by antibody indEXing (CODEX) is a multiplex fluorescence microscopy platform based on DNA-conjugated antibodies that allows the analysis of 50+ markers in a single tissue section. After staining with an antibody cocktail, tissues are imaged in a multi-cycle reaction using a microfluidics system and a fluorescence microscope with a computer automated X/Y/Z stage. DNA-conjugated antibodies are rendered visible using complementary fluorescent DNA probes, followed by imaging, probe stripping, washing and re-rendering. This process is repeated until all the antibodies present in the initial cocktail have been rendered and imaged. Here, we used CODEX to analyze intact BM at the single-cell level (~200nm resolution) during leukemic progression. Chronic myeloid leukemia (CML)-like disease was induced in non-irradiated mice using BCR-ABL1-GFP retrovirus. Tissue sections of femoral bones harvested at different time points after leukemia onset were stained using a 50+ marker CODEX antibody panel to simultaneously identify hematopoietic and leukemic stem and progenitor cells, multiple BM microenvironmental cell types, myeloid and lymphoid cells as well as functional markers. Results We have built an integrated computational pipeline for the analysis of high-dimensional CODEX data that enables the identification and characterization of BM cell types as well as their spatial organization in situ. Raw images were concatenated and aligned using Hoechst nuclear stain as a reference, followed by deconvolution, segmentation, marker expression quantification and spatial compensation. Exported data were clustered in an unsupervised manner using VorteX algorithm, which identified 28 distinct cellular clusters based on marker expression values. All major BM compartments including stromal (vascular, pericytes, osteoblastic), lymphoid (T and B cell subsets), myeloid (megakaryocytes, macrophages, dendritic cells, granulocytes) and progenitor cell types, as well as leukemic cells, were represented. During leukemic progression, the BM microenvironment was dramatically rearranged. Besides the expected growth of the leukemic clone, we observed a massive increase in vascular and osteoblastic cell types, whereas immune cell clusters were significantly reduced. Interestingly, CD71, the transferrin receptor, was strongly up-regulated on tumor cells in advanced leukemia, indicating towards a role for iron metabolism in malignant progression. Furthermore, hierarchical clustering of tissue regions based on cellular composition using X/Y/Z positional information pointed towards the emergence of specific cell-cell interaction modules that developed during leukemic progression, including mutual attraction between B cells and central arterioles. Conclusions High-dimensional imaging of the BM microenvironment by CODEX allows studying the abundance and distribution of cellular elements that are often underestimated or missed by traditional flow cytometry, such as stromal cells, vasculature and megakaryocytes. Importantly, CODEX identifies single cells in their tissue context during leukemic progression and facilitates the discovery of novel cell-cell interactions and cell types as well as unexpected marker constellations. Disclosures Samusik: Akoya Biosciences: Consultancy, Equity Ownership, Honoraria, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties. Nolan:Akoya Biosciences: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties. Goltsev:Akoya Biosciences: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties.

AbstractChemical staining of biological specimens is commonly utilised to boost contrast in soft tissue structures, but unambiguous identification of staining location and distribution is difficult without confirmation of the elemental signature, especially for chemicals of similar density contrast. Hyperspectral X-ray computed tomography (XCT) enables the non-destructive identification, segmentation and mapping of elemental composition within a sample. With the availability of hundreds of narrow, high resolution (~ 1 keV) energy channels, the technique allows the simultaneous detection of multiple contrast agents across different tissue structures. Here we describe a hyperspectral imaging routine for distinguishing multiple chemical agents, regardless of contrast similarity. Using a set of elemental calibration phantoms, we perform a first instance of direct stain concentration measurement using spectral absorption edge markers. Applied to a set of double- and triple-stained biological specimens, the study analyses the extent of stain overlap and uptake regions for commonly used contrast markers. An improved understanding of stain concentration as a function of position, and the interaction between multiple stains, would help inform future studies on multi-staining procedures, as well as enable future exploration of heavy metal uptake across medical, agricultural and ecological fields.

AbstractNuclei segmentation is a prerequisite and an essential step in cancer detection and prognosis. Automatic nuclei segmentation from the histopathological images is challenging due to nuclear overlap, disease types, chromatic stain variability, and cytoplasmic morphology differences. Furthermore, it is demanding to develop a single accurate method for segmenting nuclei of different organs because of the diversity in nuclei size, shape, and appearance across the various organs. To address these challenges, we developed a robust Encoder‐Decoder network for nuclei segmentation from the multi‐organ histopathological images. In this approach, we utilize a pre‐trained EfficientNet‐B4 as an Encoder subnetwork and design a new Decoder subnetwork architecture. Additionally, we have applied morphological operation‐based post‐processing to improve the segmentation results. The performance of our approach has been evaluated on three public datasets, namely, Kumar, TNBC, and CPM‐17 datasets, which contain histopathological images of seven organs, one organ, and four organs, respectively. The proposed method achieved an aggregated Jacquard index of 0.636, 0.611, and 0.706 on Kumar, TNBC, and CPM‐17 datasets, respectively. Our proposed approach also shows superiority over the existing methods.

AbstractAdvances in computational algorithms and tools have made the prediction of cancer patient outcomes using computational pathology feasible. However, predicting clinical outcomes from pre-treatment histopathologic images remains a challenging task, limited by the poor understanding of tumor immune micro-environments. In this study, an automatic, accurate, comprehensive, interpretable, and reproducible whole slide image (WSI) feature extraction pipeline known as, IMage-based Pathological REgistration and Segmentation Statistics (IMPRESS), is described. We used both H&E and multiplex IHC (PD-L1, CD8+, and CD163+) images, investigated whether artificial intelligence (AI)-based algorithms using automatic feature extraction methods can predict neoadjuvant chemotherapy (NAC) outcomes in HER2-positive (HER2+) and triple-negative breast cancer (TNBC) patients. Features are derived from tumor immune micro-environment and clinical data and used to train machine learning models to accurately predict the response to NAC in breast cancer patients (HER2+ AUC = 0.8975; TNBC AUC = 0.7674). The results demonstrate that this method outperforms the results trained from features that were manually generated by pathologists. The developed image features and algorithms were further externally validated by independent cohorts, yielding encouraging results, especially for the HER2+ subtype.

Abstract Introduction Artificial intelligence (AI) integration in nephropathology has been growing rapidly in recent years, facing several challenges including the wide range of histological techniques used, the low occurrence of certain diseases, and the need for data sharing. This narrative review retraces the history of AI in nephropathology and provides insights into potential future developments. Methods Electronic searches in PubMed-MEDLINE and Embase were made to extract pertinent articles from the literature. Works about automated image analysis or the application of an AI algorithm on non-neoplastic kidney histological samples were included and analyzed to extract information such as publication year, AI task, and learning type. Prepublication servers and reviews were not included. Results Seventy-six (76) original research articles were selected. Most of the studies were conducted in the United States in the last 7 years. To date, research has been mainly conducted on relatively easy tasks, like single-stain glomerular segmentation. However, there is a trend towards developing more complex tasks such as glomerular multi-stain classification. Conclusion Deep learning has been used to identify patterns in complex histopathology data and looks promising for the comprehensive assessment of renal biopsy, through the use of multiple stains and virtual staining techniques. Hybrid and collaborative learning approaches have also been explored to utilize large amounts of unlabeled data. A diverse team of experts, including nephropathologists, computer scientists, and clinicians, is crucial for the development of AI systems for nephropathology. Collaborative efforts among multidisciplinary experts result in clinically relevant and effective AI tools. Graphical abstract

AbstractHere we apply hyperspectral bright field imaging to collect computed tomographic images with excellent energy resolution (~ 1 keV), applying it for the first time to map the distribution of stain in a fixed biological sample through its characteristic K-edge. Conventionally, because the photons detected at each pixel are distributed across as many as 200 energy channels, energy-selective images are characterised by low count-rates and poor signal-to-noise ratio. This means high X-ray exposures, long scan times and high doses are required to image unique spectral markers. Here, we achieve high quality energy-dispersive tomograms from low dose, noisy datasets using a dedicated iterative reconstruction algorithm. This exploits the spatial smoothness and inter-channel structural correlation in the spectral domain using two carefully chosen regularisation terms. For a multi-phase phantom, a reduction in scan time of 36 times is demonstrated. Spectral analysis methods including K-edge subtraction and absorption step-size fitting are evaluated for an ex vivo, single (iodine)-stained biological sample, where low chemical concentration and inhomogeneous distribution can affect soft tissue segmentation and visualisation. The reconstruction algorithms are available through the open-source Core Imaging Library. Taken together, these tools offer new capabilities for visualisation and elemental mapping, with promising applications for multiply-stained biological specimens.

Un défi majeur dans l'application de l'apprentissage profond à l'histopathologie réside dans la variation des colorations, à la fois inter et intra-coloration. Les modèles d'apprentissage profond entraînés sur une seule coloration (ou domaine) échouent souvent sur d'autres, même pour la même tâche (par exemple, la segmentation des glomérules rénaux). L'annotation de chaque coloration est coûteuse et chronophage, ce qui pousse les chercheurs à explorer des méthodes de transfert de coloration basées sur l'adaptation de domaine. Celles-ci visent à réaliser une segmentation multi-coloration en utilisant des annotations d'une seule coloration, mais sont limitées par l'introduction d'un décalage de domaine, réduisant ainsi les performances. La détection et la quantification de ce décalage sont essentielles. Cette thèse se concentre sur des méthodes non supervisées pour développer une métrique de détection du décalage et propose une approche de transfert de coloration pour le minimiser. Bien que ces algorithmes réduisent le besoin d'annotations, ils peuvent être limités pour certains tissus. Cette thèse propose donc une amélioration via l'auto-supervision
A key challenge in applying deep learning to histopathology is the variation in stainings, both inter and intra-stain. Deep learning models trained on one stain (or domain) often fail on others, even for the same task (e.g., kidney glomeruli segmentation). Labelling each stain is expensive and time-consuming, prompting researchers to explore domain adaptation based stain-transfer methods. These aim to perform multi-stain segmentation using labels from only one stain but are limited by the introduction of domain shift, reducing performance. Detecting and quantifying this domain shift is important. This thesis focuses on unsupervised methods to develop a metric for detecting domain shift and proposes a novel stain-transfer approach to minimise it. While multi-stain algorithms reduce the need for labels in target stains, they may struggle with tissue types lacking source-stain labels. To address this, the thesis focuses to improve multi-stain segmentation with less reliance on labelled data using self-supervision. While this thesis focused on kidney glomeruli segmentation, the proposed methods are designed to be applicable to other histopathology tasks and domains, including medical imaging and computer vision