The Gold Standard Paradox in Digital Image Analysis: Manual Versus Automated Scoring as Ground Truth

Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

This paper discusses some overlooked challenges faced when working with machine learning models for histopathology and presents a novel opportunity to support "Learning Health Systems" with them. Initially, the authors elaborate on these challenges after separating them according to their mitigation strategies: those that need innovative approaches, time, or future technological capabilities and those that require a conceptual reappraisal from a critical perspective. Then, a novel opportunity to support "Learning Health Systems" by integrating hidden information extracted by ML models from digitalized histopathology slides with other healthcare big data is presented.

Proteomic profiling of FFPE specimens: Discovery of HNRNPA2/B1 and STT3B as biomarkers for determining formalin fixation durations

Recent advancements in proteomics technologies using formalin-fixed paraffin-embedded (FFPE) samples have significantly advanced biomarker discovery. Yet, the effects of varying sample preparation protocols on proteomic analyses remain poorly understood. We analyzed mouse liver FFPE samples that varied in fixatives, fixation duration, and storage temperature using LC/MS. We found that variations in fixation duration significantly affected the abundance of specific proteins, showing that HNRNPA2/B1 demonstrated a significant decrease in abundance in samples fixed for long periods, whereas STT3B exhibited a significant increase in abundance in samples fixed for long durations. These findings were supported by immunohistochemical analysis across liver, spleen, and lung tissues, demonstrating a significant decrease in the nuclear staining of HNRNPA2/B1 in long-duration acid formalin(AF)-fixed FFPE samples, and an increase in cytoplasmic staining of STT3B in long-duration neutral buffered formalin-fixed liver and lung tissues and granular staining in all long-duration AF-fixed FFPE tissue types. Similar trends were observed in the long-duration fixed HeLa cells. These results demonstrate that fixation duration critically affects the proteomic integrity of FFPE samples, emphasizing the urgent need for standardized fixation protocols to ensure consistent and reliable proteomic data. SIGNIFICANCE: The quality of FFPE samples is primarily influenced by the fixation and storage conditions. However, previous studies have mainly focused on their impact on nucleic acids and the extent to which different fixation conditions affect changes in proteins has not been evaluated. In addition, to our knowledge, proteomic research focusing on differences in formalin fixation conditions has not yet been conducted. Here, we analyzed FFPE samples with different formalin fixation and storage conditions using LC/MS and evaluated the impact of different fixation conditions on protein variations. Our study unequivocally established formalin fixation duration as a critical determinant of protein variation in FFPE specimens and successfully identified HNRNPA2/B1 and STT3B as potential biomarkers for predicting formalin fixation duration for the first time. The study findings open new avenues for quality assessment in biomedical research and diagnostics.

Computational pathology: an evolving concept

The initial enthusiasm about computational pathology (CP) and artificial intelligence (AI) was that they will replace pathologists entirely on the way to fully automated diagnostics. It is becoming clear that currently this is not the immediate model to pursue. On top of the legal and regulatory complexities surrounding its implementation, the majority of tested machine learning (ML)-based predictive algorithms do not display the exquisite performance needed to render them unequivocal, standalone decision makers for matters with direct implications to human health. We are thus moving into a different model of "computer-assisted diagnostics", where AI is there to provide support, rather than replacing, the pathologist. Herein we focus on the practical aspects of CP, from a pathologist perspective. There is a wide range of potential applications where CP can enhance precision of pathology diagnosis, tailor prognostic and predictive information, as well as save time. There are, however, a number of potential limitations for CP that currently hinder their wider adoption in the clinical setting. We address the key necessary steps towards clinical implementation of computational pathology, discuss the significant obstacles that hinders its adoption in the clinical context and summarize some proposed solutions. We conclude that the advancement of CP in the clinic is a promising resource-intensive endeavour that requires broad and inclusive collaborations between academia, industry, and regulatory bodies.

Automated quantification of avian influenza virus antigen in different organs

As immunohistochemistry is valuable for determining tissue and cell tropism of avian influenza viruses (AIV), but time-consuming, an artificial intelligence-based workflow was developed to automate the AIV antigen quantification. Organ samples from experimental AIV infections including brain, heart, lung and spleen on one slide, and liver and kidney on another slide were stained for influenza A-matrixprotein and analyzed with QuPath: Random trees algorithms were trained to identify the organs on each slide, followed by threshold-based quantification of the immunoreactive area. The algorithms were trained and tested on two different slide sets, then retrained on both and validated on a third set. Except for the kidney, the best algorithms for organ selection correctly identified the largest proportion of the organ area. For most organs, the immunoreactive area assessed following organ selection was significantly and positively correlated to a manually assessed semiquantitative score. In the validation set, intravenously infected chickens showed a generally higher percentage of immunoreactive area than chickens infected oculonasally. Variability between the slide sets and a similar tissue texture of some organs limited the ability of the algorithms to select certain organs. Generally, suitable correlations of the immunoreactivity data results were achieved, facilitating high-throughput analysis of AIV tissue tropism.

Digital counting of tissue cells for molecular analysis: the QuANTUM pipeline

The estimation of tumor cellular fraction (TCF) is a crucial step in predictive molecular pathology, representing an entry adequacy criterion also in the next-generation sequencing (NGS) era. However, heterogeneity of quantification practices and inter-pathologist variability hamper the robustness of its evaluation, stressing the need for more reliable results. Here, 121 routine histological samples from non-small cell lung cancer (NSCLC) cases with complete NGS profiling were used to evaluate TCF interobserver variability among three different pathologists (pTCF), developing a computational tool (cTCF) and assessing its reliability vs ground truth (GT) tumor cellularity and potential impact on the final molecular results. Inter-pathologist reproducibility was fair to good, with overall Wk ranging between 0.46 and 0.83 (avg. 0.59). The obtained cTCF was comparable to the GT (p = 0.129, 0.502, and 0.130 for surgical, biopsies, and cell block, respectively) and demonstrated good reliability if elaborated by different pathologists (Wk = 0.9). Overall cTCF was lower as compared to pTCF (30 ± 10 vs 52 ± 19, p < 0.001), with more cases < 20% (17, 14%, p = 0.690), but none containing < 100 cells for the algorithm. Similarities were noted between tumor area estimation and pTCF (36 ± 29, p < 0.001), partly explaining variability in the human assessment of tumor cellularity. Finally, the cTCF allowed a reduction of the copy number variations (CNVs) called (27 vs 29, - 6.9%) with an increase of effective CNVs detection (13 vs 7, + 85.7%), some with potential clinical impact previously undetected with pTCF. An automated computational pipeline (Qupath Analysis of Nuclei from Tumor to Uniform Molecular tests, QuANTUM) has been created and is freely available as a QuPath extension. The computational method used in this study has the potential to improve efficacy and reliability of TCF estimation in NSCLC, with demonstrated impact on the final molecular results.

SFE-Net: Spatial-Frequency Enhancement Network for robust nuclei segmentation in histopathology images

Morphological features of individual nuclei serve as a dependable foundation for pathologists in making accurate diagnoses. Existing methods that rely on spatial information for feature extraction have achieved commendable results in nuclei segmentation tasks. However, these approaches are not sufficient to extract edge information of nuclei with small sizes and blurred outlines. Moreover, the lack of attention to the interior of the nuclei leads to significant internal inconsistencies. To address these challenges, we introduce a novel Spatial-Frequency Enhancement Network (SFE-Net) to incorporate spatial-frequency features and promote intra-nuclei consistency for robust nuclei segmentation. Specifically, SFE-Net incorporates a distinctive Spatial-Frequency Feature Extraction module and a Spatial-Guided Feature Enhancement module, which are designed to preserve spatial-frequency information and enhance feature representation respectively, to achieve comprehensive extraction of edge information. Furthermore, we introduce the Label-Guided Distillation method, which utilizes semantic features to guide the segmentation network in strengthening boundary constraints and learning the intra-nuclei consistency of individual nuclei, to improve the robustness of nuclei segmentation. Extensive experiments on three publicly available histopathology image datasets (MoNuSeg, TNBC and CryoNuSeg) demonstrate the superiority of our proposed method, which achieves 79.23%, 81.96% and 73.26% Aggregated Jaccard Index, respectively. The proposed model is available at https://github.com/jinshachen/SFE-Net.

Digital Pathology Applications for PD-L1 Scoring in Head and Neck Squamous Cell Carcinoma: A Challenging Series

The assessment of programmed death-ligand 1 (PD-L1) combined positive scoring (CPS) in head and neck squamous cell carcinoma (HNSCC) is challenged by pre-analytical and inter-observer variabilities. An educational program to compare the diagnostic performances between local pathologists and a board of pathologists on 11 challenging cases from different Italian pathology centers stained with PD-L1 immunohistochemistry on a digital pathology platform is reported. A laboratory-developed test (LDT) using both 22C3 (Dako) and SP263 (Ventana) clones on Dako or Ventana platforms was compared with the companion diagnostic (CDx) Dako 22C3 pharm Dx assay. A computational approach was performed to assess possible correlations between stain features and pathologists' visual assessments. Technical discordances were noted in five cases (LDT vs. CDx, 45%), due to an abnormal nuclear/cytoplasmic diaminobenzidine (DAB) stain in LDT (n = 2, 18%) and due to variation in terms of intensity, dirty background, and DAB droplets (n = 3, 27%). Interpretative discordances were noted in six cases (LDT vs. CDx, 54%). CPS remained unchanged, increased, or decreased from LDT to CDx in three (27%) cases, two (18%) cases, and one (9%) case, respectively, around relevant cutoffs (1 and 20, k = 0.63). Differences noted in DAB intensity/distribution using computational pathology partly explained the LDT vs. CDx differences in two cases (18%). Digital pathology may help in PD-L1 scoring, serving as a second opinion consultation platform in challenging cases. Computational and artificial intelligence tools will improve clinical decision-making and patient outcomes.

A Robust Method for the Unsupervised Scoring of Immunohistochemical Staining

Immunohistochemistry is a powerful technique that is widely used in biomedical research and clinics; it allows one to determine the expression levels of some proteins of interest in tissue samples using color intensity due to the expression of biomarkers with specific antibodies. As such, immunohistochemical images are complex and their features are difficult to quantify. Recently, we proposed a novel method, including a first separation stage based on non-negative matrix factorization (NMF), that achieved good results. However, this method was highly dependent on the parameters that control sparseness and non-negativity, as well as on algorithm initialization. Furthermore, the previously proposed method required a reference image as a starting point for the NMF algorithm. In the present work, we propose a new, simpler and more robust method for the automated, unsupervised scoring of immunohistochemical images based on bright field. Our work is focused on images from tumor tissues marked with blue (nuclei) and brown (protein of interest) stains. The new proposed method represents a simpler approach that, on the one hand, avoids the use of NMF in the separation stage and, on the other hand, circumvents the need for a control image. This new approach determines the subspace spanned by the two colors of interest using principal component analysis (PCA) with dimension reduction. This subspace is a two-dimensional space, allowing for color vector determination by considering the point density peaks. A new scoring stage is also developed in our method that, again, avoids reference images, making the procedure more robust and less dependent on parameters. Semi-quantitative image scoring experiments using five categories exhibit promising and consistent results when compared to manual scoring carried out by experts.

Pseudo-spectral angle mapping for automated pixel-level analysis of highly multiplexed tissue image data

The rapid development of highly multiplexed microscopy systems has enabled the study of cells embedded within their native tissue, which is providing exciting insights into the spatial features of human disease [1]. However, computational methods for analyzing these high-content images are still emerging, and there is a need for more robust and generalizable tools for evaluating the cellular constituents and underlying stroma captured by high-plex imaging [2]. To address this need, we have adapted spectral angle mapping - an algorithm used widely in hyperspectral image analysis - to compress the channel dimension of high-plex immunofluorescence images. As many high-plex immunofluorescence imaging experiments probe unique sets of protein markers, existing cell and pixel classification models do not typically generalize well. Pseudospectral angle mapping (pSAM) uses reference pseudospectra - or pixel vectors - to assign each pixel in an image a similarity score to several cell class reference vectors, which are defined by each unique staining panel. Here, we demonstrate that the class maps provided by pSAM can directly provide insight into the prevalence of each class defined by reference pseudospectra. In a dataset of high-plex images of colon biopsies from patients with gut autoimmune conditions, sixteen pSAM class representation maps were combined with instance segmentation of cells to provide cell class predictions. Finally, pSAM detected a diverse set of structure and immune cells when applied to a novel dataset of kidney biopsies imaged with a 43-marker panel. In summary, pSAM provides a powerful and readily generalizable method for evaluating high-plex immunofluorescence image data.

Digital pathology in cardiac transplant diagnostics: from biopsies to algorithms

In the field of heart transplantation, the ability to accurately and promptly diagnose cardiac allograft rejection is crucial. This comprehensive review explores the transformative role of digital pathology and computational pathology, especially through machine learning, in this critical domain. These methodologies harness large datasets to extract subtle patterns and valuable information that extend beyond human perceptual capabilities, potentially enhancing diagnostic outcomes. Current research indicates that these computer-based systems could offer accuracy and performance matching, or even exceeding, that of expert pathologists, thereby introducing more objectivity and reducing observer variability. Despite promising results, several challenges such as limited sample sizes, diverse data sources, and the absence of standardized protocols pose significant barriers to the widespread adoption of these techniques. The future of digital pathology in heart transplantation diagnostics depends on utilizing larger, more diverse patient cohorts, standardizing data collection, processing, and evaluation protocols, and fostering collaborative research efforts. The integration of various data types, including clinical, demographic, and imaging information, could further refine diagnostic precision. As researchers address these challenges and promote collaborative efforts, digital pathology has the potential to become an integral part of clinical practice, ultimately improving patient care in heart transplantation.

Steroid receptor coactivator-2 drives epithelial reprogramming that enables murine embryo implantation

Although we have shown that steroid receptor coactivator-2 (SRC-2), a member of the p160/SRC family of transcriptional coregulators, is essential for decidualization of both human and murine endometrial stromal cells, SRC-2's role in the earlier stages of the implantation process have not been adequately addressed. Using a conditional SRC-2 knockout mouse (SRC-2d/d ) in timed natural pregnancy studies, we show that endometrial SRC-2 is required for embryo attachment and adherence to the luminal epithelium. Implantation failure is associated with the persistent expression of Mucin 1 and E-cadherin on the apical surface and basolateral adherens junctions of the SRC-2d/d luminal epithelium, respectively. These findings indicate that the SRC-2d/d luminal epithelium fails to exhibit a plasma membrane transformation (PMT) state known to be required for the development of uterine receptivity. Transcriptomics demonstrated that the expression of genes involved in steroid hormone control of uterine receptivity were significantly disrupted in the SRC-2d/d endometrium as well as genes that control epithelial tight junctional biology and the emergence of the epithelial mesenchymal transition state, with the latter sharing similar biological properties with PMT. Collectively, these findings uncover a new role for endometrial SRC-2 in the induction of the luminal epithelial PMT state, which is a prerequisite for the development of uterine receptivity and early pregnancy establishment.

Using Sound Location to Monitor Farrowing in Sows

Precision Livestock Farming systems can help pig farmers prevent health and welfare issues around farrowing. Five sows were monitored in two field studies. A Sorama L642V sound camera, visualising sound sources as coloured spots using a 64-microphone array, and a Bascom XD10-4 security camera with a built-in microphone were used in a farrowing unit. Firstly, sound spots were compared with audible sounds, using the Observer XT (Noldus Information Technology), analysing video data at normal speed. This gave many false positives, including visible sound spots without audible sounds. In total, 23 of 50 piglet births were visible, but none were audible. The sow's behaviour changed when farrowing started. One piglet was silently crushed. Secondly, data were analysed at a 10-fold slower speed when comparing sound spots with audible sounds and sow behaviour. This improved results, but accuracy and specificity were still low. When combining audible sound with visible sow behaviour and comparing sound spots with combined sound and behaviour, the accuracy was 91.2%, the error was 8.8%, the sensitivity was 99.6%, and the specificity was 69.7%. We conclude that sound cameras are promising tools, detecting sound more accurately than the human ear. There is potential to use sound cameras to detect the onset of farrowing, but more research is needed to detect piglet births or crushing.

Histological classification of canine and feline lymphoma using a modular approach based on deep learning and advanced image processing

Histopathological examination of tissue samples is essential for identifying tumor malignancy and the diagnosis of different types of tumor. In the case of lymphoma classification, nuclear size of the neoplastic lymphocytes is one of the key features to differentiate the different subtypes. Based on the combination of artificial intelligence and advanced image processing, we provide a workflow for the classification of lymphoma with regards to their nuclear size (small, intermediate, and large). As the baseline for our workflow testing, we use a Unet++ model trained on histological images of canine lymphoma with individually labeled nuclei. As an alternative to the Unet++, we also used a publicly available pre-trained and unmodified instance segmentation model called Stardist to demonstrate that our modular classification workflow can be combined with different types of segmentation models if they can provide proper nuclei segmentation. Subsequent to nuclear segmentation, we optimize algorithmic parameters for accurate classification of nuclear size using a newly derived reference size and final image classification based on a pathologists-derived ground truth. Our image classification module achieves a classification accuracy of up to 92% on canine lymphoma data. Compared to the accuracy ranging from 66.67 to 84% achieved using measurements provided by three individual pathologists, our algorithm provides a higher accuracy level and reproducible results. Our workflow also demonstrates a high transferability to feline lymphoma, as shown by its accuracy of up to 84.21%, even though our workflow was not optimized for feline lymphoma images. By determining the nuclear size distribution in tumor areas, our workflow can assist pathologists in subtyping lymphoma based on the nuclei size and potentially improve reproducibility. Our proposed approach is modular and comprehensible, thus allowing adaptation for specific tasks and increasing the users' trust in computer-assisted image classification.

CaMeL-Net: Centroid-aware metric learning for efficient multi-class cancer classification in pathology images

BACKGROUND AND OBJECTIVE

Cancer grading in pathology image analysis is a major task due to its importance in patient care, treatment, and management. The recent developments in artificial neural networks for computational pathology have demonstrated great potential to improve the accuracy and quality of cancer diagnosis. These improvements are generally ascribable to the advance in the architecture of the networks, often leading to increase in the computation and resources. In this work, we propose an efficient convolutional neural network that is designed to conduct multi-class cancer classification in an accurate and robust manner via metric learning.

METHODS

We propose a centroid-aware metric learning network for an improved cancer grading in pathology images. The proposed network utilizes centroids of different classes within the feature embedding space to optimize the relative distances between pathology images, which manifest the innate similarities/dissimilarities between them. For improved optimization, we introduce a new loss function and a training strategy that are tailored to the proposed network and metric learning.

RESULTS

We evaluated the proposed approach on multiple datasets of colorectal and gastric cancers. For the colorectal cancer, two different datasets were employed that were collected from different acquisition settings. the proposed method achieved an accuracy, F1-score, quadratic weighted kappa of 88.7%, 0.849, and 0.946 for the first dataset and 83.3%, 0.764, and 0.907 for the second dataset, respectively. For the gastric cancer, the proposed method obtained an accuracy of 85.9%, F1-score of 0.793, and quadratic weighted kappa of 0.939. We also found that the proposed method outperforms other competing models and is computationally efficient.

CONCLUSIONS

The experimental results demonstrate that the prediction results by the proposed network are both accurate and reliable. The proposed network not only outperformed other related methods in cancer classification but also achieved superior computational efficiency during training and inference. The future study will entail further development of the proposed method and the application of the method to other problems and domains.

Clinical Proteomics for Solid Organ Tissues

The evaluation of biopsied solid organ tissue has long relied on visual examination using a microscope. Immunohistochemistry is critical in this process, labeling and detecting cell lineage markers and therapeutic targets. However, while the practice of immunohistochemistry has reshaped diagnostic pathology and facilitated improvements in cancer treatment, it has also been subject to pervasive challenges with respect to standardization and reproducibility. Efforts are ongoing to improve immunohistochemistry, but for some applications, the benefit of such initiatives could be impeded by its reliance on monospecific antibody-protein reagents and limited multiplexing capacity. This perspective surveys the relevant challenges facing traditional immunohistochemistry and describes how mass spectrometry, particularly liquid chromatography-tandem mass spectrometry, could help alleviate problems. In particular, targeted mass spectrometry assays could facilitate measurements of individual proteins or analyte panels, using internal standards for more robust quantification and improved interlaboratory reproducibility. Meanwhile, untargeted mass spectrometry, showcased to date clinically in the form of amyloid typing, is inherently multiplexed, facilitating the detection and crude quantification of 100s to 1000s of proteins in a single analysis. Further, data-independent acquisition has yet to be applied in clinical practice, but offers particular strengths that could appeal to clinical users. Finally, we discuss the guidance that is needed to facilitate broader utilization in clinical environments and achieve standardization.

Automated diagnosis of 7 canine skin tumors using machine learning on H&E-stained whole slide images

Microscopic evaluation of hematoxylin and eosin-stained slides is still the diagnostic gold standard for a variety of diseases, including neoplasms. Nevertheless, intra- and interrater variability are well documented among pathologists. So far, computer assistance via automated image analysis has shown potential to support pathologists in improving accuracy and reproducibility of quantitative tasks. In this proof of principle study, we describe a machine-learning-based algorithm for the automated diagnosis of 7 of the most common canine skin tumors: trichoblastoma, squamous cell carcinoma, peripheral nerve sheath tumor, melanoma, histiocytoma, mast cell tumor, and plasmacytoma. We selected, digitized, and annotated 350 hematoxylin and eosin-stained slides (50 per tumor type) to create a database divided into training, n = 245 whole-slide images (WSIs), validation (n = 35 WSIs), and test sets (n = 70 WSIs). Full annotations included the 7 tumor classes and 6 normal skin structures. The data set was used to train a convolutional neural network (CNN) for the automatic segmentation of tumor and nontumor classes. Subsequently, the detected tumor regions were classified patch-wise into 1 of the 7 tumor classes. A majority of patches-approach led to a tumor classification accuracy of the network on the slide-level of 95% (133/140 WSIs), with a patch-level precision of 85%. The same 140 WSIs were provided to 6 experienced pathologists for diagnosis, who achieved a similar slide-level accuracy of 98% (137/140 correct majority votes). Our results highlight the feasibility of artificial intelligence-based methods as a support tool in diagnostic oncologic pathology with future applications in other species and tumor types.

Prognostic effect of CD73 in pancreatic ductal adenocarcinoma for disease-free survival after radical surgery

PURPOSE

Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease with a high potency of metastasis or recurrence after radical resection. Effective predictors for metastasis and recurrence postoperatively were dominant for the development of systemic adjuvant treatment regimens. The ATP hydrolase correlated gene CD73 was described as a promoter in tumor growth and immune escape of PDAC. However, there lacked research focused on the role of CD73 in PDAC metastasis. This study aimed to investigate the expression of CD73 in PDAC patients with different outcomes as well as the prognostic effect of CD73 for disease-free survival (DFS).

METHODS

The expression level of CD73 in cancerous samples from 301 PDAC patients was evaluated by immunohistochemistry (IHC) and translated into a histochemistry score (H-score) by the HALO analysis system. Then, the CD73 H-score was involved in multivariate Cox regression along with other clinicopathological characteristics to find independent prognostic factors for DFS. Finally, a nomogram was constructed based on those independent prognostic factors for DFS prediction.

RESULTS

Higher CD73 expression was found in PDAC patients with tumor metastasis postoperatively. Meanwhile, higher CD73 expressions were also investigated in PDAC patients diagnosed with advanced N stage and T stage. Furthermore, CD73 H-score along with tumor margin status, CA19-9, 8th N stage, and adjuvant chemotherapy was indicated as independent prognostic factors for DFS in PDAC patients. The nomogram based on these factors predicted DFS in a good manner.

CONCLUSION

CD73 was associated with PDAC metastasis and served as an effective prognostic factor for DFS in PDAC patients after radical surgery.

Results of the European Society of Toxicologic Pathology Survey on the Use of Artificial Intelligence in Toxicologic Pathology

The European Society of Toxicologic Pathology (ESTP) initiated a survey through its Pathology 2.0 workstream in partnership with sister professional societies in Europe and North America to generate a snapshot of artificial intelligence (AI) usage in the field of toxicologic pathology. In addition to demographic information, some general questions explored AI relative to (1) the current status of adoption across organizations; (2) technical and methodological aspects; (3) perceived business value and finally; and (4) roadblocks and perspectives. AI has become increasingly established in toxicologic pathology with most pathologists being supportive of its development despite some areas of uncertainty. A salient feature consisted of the variability of AI awareness and adoption among the responders, as the spectrum extended from pathologists having developed familiarity and technical skills in AI, to colleagues who had no interest in AI as a tool in toxicologic pathology. Despite a general enthusiasm for these techniques, the overall understanding and trust in AI algorithms as well as their added value in toxicologic pathology were generally low, suggesting room for the need for increased awareness and education. This survey will serve as a basis to evaluate the evolution of AI penetration and acceptance in this domain.

ClusterSeg: A crowd cluster pinpointed nucleus segmentation framework with cross-modality datasets

The detection and segmentation of individual cells or nuclei is often involved in image analysis across a variety of biology and biomedical applications as an indispensable prerequisite. However, the ubiquitous presence of crowd clusters with morphological variations often hinders successful instance segmentation. In this paper, nuclei cluster focused annotation strategies and frameworks are proposed to overcome this challenging practical problem. Specifically, we design a nucleus segmentation framework, namely ClusterSeg, to tackle nuclei clusters, which consists of a convolutional-transformer hybrid encoder and a 2.5-path decoder for precise predictions of nuclei instance mask, contours, and clustered-edges. Additionally, an annotation-efficient clustered-edge pointed strategy pinpoints the salient and error-prone boundaries, where a partially-supervised PS-ClusterSeg is presented using ClusterSeg as the segmentation backbone. The framework is evaluated with four privately curated image sets and two public sets with characteristic severely clustered nuclei across a variety range of image modalities, e.g., microscope, cytopathology, and histopathology images. The proposed ClusterSeg and PS-ClusterSeg are modality-independent and generalizable, and superior to current state-of-the-art approaches in multiple metrics empirically. Our collected data, the elaborate annotations to both public and private set, as well the source code, are released publicly at https://github.com/lu-yizhou/ClusterSeg.

European Society of Toxicologic Pathology (Pathology 2.0 Molecular Pathology Special Interest Group): Review of In Situ Hybridization Techniques for Drug Research and Development

In situ hybridization (ISH) is used for the localization of specific nucleic acid sequences in cells or tissues by complementary binding of a nucleotide probe to a specific target nucleic acid sequence. In the last years, the specificity and sensitivity of ISH assays were improved by innovative techniques like synthetic nucleic acids and tandem oligonucleotide probes combined with signal amplification methods like branched DNA, hybridization chain reaction and tyramide signal amplification. These improvements increased the application spectrum for ISH on formalin-fixed paraffin-embedded tissues. ISH is a powerful tool to investigate DNA, mRNA transcripts, regulatory noncoding RNA, and therapeutic oligonucleotides. ISH can be used to obtain spatial information of a cell type, subcellular localization, or expression levels of targets. Since immunohistochemistry and ISH share similar workflows, their combination can address simultaneous transcriptomics and proteomics questions. The goal of this review paper is to revisit the current state of the scientific approaches in ISH and its application in drug research and development.

Validation of automated positive cell and region detection of immunohistochemically stained laryngeal tumor tissue using digital image analysis

Objectives

This study aimed to validate a digital image analysis (DIA) workflow for automatic positive cell detection and positive region delineation for immunohistochemical hypoxia markers with a nuclear (hypoxia-inducible factor 1α [HIF-1α]) and a cytoplasmic (pimonidazole [PIMO]) staining pattern.

Materials and methods

101 tissue fragments from 44 laryngeal tumor biopsies were immunohistochemically stained for HIF-1α and PIMO. QuPath was used to determine the percentage of positive cells and to delineate positive regions automatically. For HIF-1α, only cells with strong staining were considered positive. Three dedicated head and neck pathologists scored the percentage of positive cells using three categories (0: <1%; 1: 1%-33%; 2: >33%;). The pathologists also delineated the positive regions on 14 corresponding PIMO and HIF-1α-stained fragments. The consensus between observers was used as the reference standard and was compared to the automatic delineation.

Results

Agreement between categorical positivity scores was 76.2% and 65.4% for PIMO and HIF-1α, respectively. In all cases of disagreement in HIF-1α fragments, the DIA underestimated the percentage of positive cells. As for the region detection, the DIA correctly detected most positive regions on PIMO fragments (false positive area=3.1%, false negative area=0.7%). In HIF-1α, the DIA missed some positive regions (false positive area=1.3%, false negative area=9.7%).

Conclusions

Positive cell and region detection on biopsy material is feasible, but further optimization is needed before unsupervised use. Validation at varying DAB staining intensities is hampered by lack of reliability of the gold standard (i.e., visual human interpretation). Nevertheless, the DIA method has the potential to be used as a tool to assist pathologists in the analysis of IHC staining.

The efficacy of a machine learning algorithm for assessing tumour components as a prognostic marker of surgically resected stage IA lung adenocarcinoma

BACKGROUND

The importance of the stromal components in tumour progression has been discussed widely, but their prognostic role in small size tumours with lepidic components is not fully understood. Applying digital tissue image analysis to whole-slide imaging may enhance the accuracy and reproducibility of pathological assessment. This study aimed to evaluate the prognostic value of tumour components of lung adenocarcinoma by measuring the dimensions of the tumour consisting elements separately, using a machine learning algorithm.

METHODS

Between September 2002 and December 2016, 317 patients with surgically resected, pathological stage IA adenocarcinoma with lepidic components were analysed. We assessed the whole tumour area, including the lepidic components, and measured the epithelium, collagen, elastin areas and alveolar air space. We analysed the prognostic impact of each tumour component.

RESULTS

The dimensions of the epithelium and collagen areas were independent significant risk factors for recurrence-free survival (hazard ratio, 8.38; 95% confidence interval, 1.14-61.88; P = 0.037, and hazard ratio, 2.58; 95% confidence interval, 1.14-5.83; P = 0.022, respectively). According to the subgroup analysis when combining the epithelium and collagen areas as risk factors, patients with tumours consisting of both large epithelium and collagen areas showed significantly poor prognoses (P = 0.002).

CONCLUSIONS

We assessed tumour components using a machine learning algorithm to stratify the post-operative prognosis of surgically resected stage IA adenocarcinomas. This method might guide the selection of patients with a high risk of recurrence.

Heterogeneity of Glycan Biomarker Clusters as an Indicator of Recurrence in Pancreatic Cancer

Outcomes following tumor resection vary dramatically among patients with pancreatic cancer. A challenge in defining predictive biomarkers is to discern within the complex tumor tissue the specific subpopulations and relationships that drive recurrence. Multiplexed immunofluorescence is valuable for such studies when supplied with markers of relevant subpopulations and analysis methods to sort out the intra-tumor relationships that are informative of tumor behavior. We hypothesized that the glycan biomarkers CA19-9 and STRA, which detect separate subpopulations of cancer cells, define intra-tumoral features associated with recurrence. We probed this question using automated signal thresholding and spatial cluster analysis applied to the immunofluorescence images of the STRA and CA19-9 glycan biomarkers in whole-block tumor sections. The tumors (N = 22) displayed extreme diversity between them in the amounts of the glycans and in the levels of spatial clustering, but neither the amounts nor the clusters of the individual and combined glycans associated with recurrence. The combined glycans, however, marked divergent types of spatial clusters, alternatively only STRA, only CA19-9, or both. The co-occurrence of more than one cluster type within a tumor associated significantly with disease recurrence, in contrast to the independent occurrence of each type of cluster. In addition, intra-tumoral regions with heterogeneity in biomarker clusters spatially aligned with pathology-confirmed cancer cells, whereas regions with homogeneous biomarker clusters aligned with various non-cancer cells. Thus, the STRA and CA19-9 glycans are markers of distinct and co-occurring subpopulations of cancer cells that in combination are associated with recurrence. Furthermore, automated signal thresholding and spatial clustering provides a tool for quantifying intra-tumoral subpopulations that are informative of outcome.

Cytologic scoring of equine exercise-induced pulmonary hemorrhage: Performance of human experts and a deep learning-based algorithm

Exercise-induced pulmonary hemorrhage (EIPH) is a relevant respiratory disease in sport horses, which can be diagnosed by examination of bronchoalveolar lavage fluid (BALF) cells using the total hemosiderin score (THS). The aim of this study was to evaluate the diagnostic accuracy and reproducibility of annotators and to validate a deep learning-based algorithm for the THS. Digitized cytological specimens stained for iron were prepared from 52 equine BALF samples. Ten annotators produced a THS for each slide according to published methods. The reference methods for comparing annotator's and algorithmic performance included a ground truth dataset, the mean annotators' THSs, and chemical iron measurements. Results of the study showed that annotators had marked interobserver variability of the THS, which was mostly due to a systematic error between annotators in grading the intracytoplasmatic hemosiderin content of individual macrophages. Regarding overall measurement error between the annotators, 87.7% of the variance could be reduced by using standardized grades based on the ground truth. The algorithm was highly consistent with the ground truth in assigning hemosiderin grades. Compared with the ground truth THS, annotators had an accuracy of diagnosing EIPH (THS of < or ≥ 75) of 75.7%, whereas, the algorithm had an accuracy of 92.3% with no relevant differences in correlation with chemical iron measurements. The results show that deep learning-based algorithms are useful for improving reproducibility and routine applicability of the THS. For THS by experts, a diagnostic uncertainty interval of 40 to 110 is proposed. THSs within this interval have insufficient reproducibility regarding the EIPH diagnosis.

Heterogeneity of glycan biomarker clusters as an indicator of recurrence in pancreatic cancer

Introduction

Outcomes following tumor resection vary dramatically among patients with pancreatic ductal adenocarcinoma (PDAC). A challenge in defining predictive biomarkers is to discern within the complex tumor tissue the specific subpopulations and relationships that drive recurrence. Multiplexed immunofluorescence is valuable for such studies when supplied with markers of relevant subpopulations and analysis methods to sort out the intra-tumor relationships that are informative of tumor behavior. We hypothesized that the glycan biomarkers CA19-9 and STRA, which detect separate subpopulations of cancer cells, define intra-tumoral features associated with recurrence.

Methods

We probed this question using automated signal thresholding and spatial cluster analysis applied to the immunofluorescence images of the STRA and CA19-9 glycan biomarkers in whole-block sections of PDAC tumors collected from curative resections.

Results

The tumors (N = 22) displayed extreme diversity between them in the amounts of the glycans and in the levels of spatial clustering, but neither the amounts nor the clusters of the individual and combined glycans associated with recurrence. The combined glycans, however, marked divergent types of spatial clusters, alternatively only STRA, only CA19-9, or both. The co-occurrence of more than one cluster type within a tumor associated significantly with disease recurrence, in contrast to the independent occurrence of each type of cluster. In addition, intra-tumoral regions with heterogeneity in biomarker clusters spatially aligned with pathology-confirmed cancer cells, whereas regions with homogeneous biomarker clusters aligned with various non-cancer cells.

Conclusion

Thus, the STRA and CA19-9 glycans are markers of distinct and co-occurring subpopulations of cancer cells that in combination are associated with recurrence. Furthermore, automated signal thresholding and spatial clustering provides a tool for quantifying intra-tumoral subpopulations that are informative of outcome.

An update on computational pathology tools for genitourinary pathology practice: A review paper from the Genitourinary Pathology Society (GUPS)

Machine learning has been leveraged for image analysis applications throughout a multitude of subspecialties. This position paper provides a perspective on the evolutionary trajectory of practical deep learning tools for genitourinary pathology through evaluating the most recent iterations of such algorithmic devices. Deep learning tools for genitourinary pathology demonstrate potential to enhance prognostic and predictive capacity for tumor assessment including grading, staging, and subtype identification, yet limitations in data availability, regulation, and standardization have stymied their implementation.

MEED: A novel robust contrast enhancement procedure yielding highly-convergent thresholding of biofilm images

Morphological and statistical investigation of biofilm images may be even more critical than the image acquisition itself, in particular in the presence of morphologically complex distributions, due to the unavoidable impact of the measurement technique too. Hence, digital image pre-processing is mandatory for reliable feature extraction and enhancement preliminary to segmentation. Also, pattern recognition in automated deep learning (both supervised and unsupervised) models often requires a preliminary effective contrast-enhancement. However, no universal consensus exists on the optimal contrast enhancement approach. This paper presents and discusses a new general, robust, reproducible, accurate and easy to implement contrast enhancement procedure, briefly named MEED-procedure, able to work on images with different bacterial coverages and biofilm structures, coming from different imaging instrumentations (herein stereomicroscope and transmission microscope). It exploits a proper succession of basic morphological operations (erosion and dilation) and a horizontal line structuring element, to minimize the impact on size and shape of the even finer bacterial features. It systematically enhances the objects of interest, without histogram stretching and/or undesirable artifacts yielded by common automated methods. The quality of the MEED-procedure is ascertained by segmentation tests which demonstrate its robustness regarding the determination of threshold and convergence of the thresholding algorithm. Extensive validation tests over a rich image database, comparison with the literature and comprehensive discussion of the conceptual background support the superiority of the MEED-procedure over the existing methods and demonstrate it is not a routine application of morphological operators.

Application of Artificial Intelligence in Pathology: Trends and Challenges

Given the recent success of artificial intelligence (AI) in computer vision applications, many pathologists anticipate that AI will be able to assist them in a variety of digital pathology tasks. Simultaneously, tremendous advancements in deep learning have enabled a synergy with artificial intelligence (AI), allowing for image-based diagnosis on the background of digital pathology. There are efforts for developing AI-based tools to save pathologists time and eliminate errors. Here, we describe the elements in the development of computational pathology (CPATH), its applicability to AI development, and the challenges it faces, such as algorithm validation and interpretability, computing systems, reimbursement, ethics, and regulations. Furthermore, we present an overview of novel AI-based approaches that could be integrated into pathology laboratory workflows.

Artificial intelligence assisted compositional analyses of human abdominal aortic aneurysms ex vivo

Quantification of histological information from excised human abdominal aortic aneurysm (AAA) specimens may provide essential information on the degree of infiltration of inflammatory cells in different regions of the AAA. Such information will support mechanistic insight in AAA pathology and can be linked to clinical measures for further development of AAA treatment regimens. We hypothesize that artificial intelligence can support high throughput analyses of histological sections of excised human AAA. We present an analysis framework based on supervised machine learning. We used TensorFlow and QuPath to determine the overall architecture of the AAA: thrombus, arterial wall, and adventitial loose connective tissue. Within the wall and adventitial zones, the content of collagen, elastin, and specific inflammatory cells was quantified. A deep neural network (DNN) was trained on manually annotated, Weigert stained, tissue sections (14 patients) and validated on images from two other patients. Finally, we applied the method on 95 new patient samples. The DNN was able to segment the sections according to the overall wall architecture with Jaccard coefficients after 65 epocs of 92% for the training and 88% for the validation data set, respectively. Precision and recall both reached 92%. The zone areas were highly variable between patients, as were the outputs on total cell count and elastin/collagen fiber content. The number of specific cells or stained area per zone was deterministically determined. However, combining the masks based on the Weigert stainings, with images of immunostained serial sections requires addition of landmark recognition to the analysis path. The combination of digital pathology, the DNN we developed, and landmark registration will provide a strong tool for future analyses of the histology of excised human AAA. In combination with biomechanical testing and microstructurally motivated mathematical models of AAA remodeling, the method has the potential to be a strong tool to provide mechanistic insight in the disease. In combination with each patients’ demographic and clinical profile, the method can be an interesting tool to in supportof a better treatment regime for the patients.

IT/QA and Regulatory Aspects of Digital Pathology: Results of the 8th ESTP International Workshop

Digital toxicologic histopathology has been broadly adopted in preclinical compound development for informal consultation and peer review. There is now increased interest in implementing the technology for good laboratory practice-regulated study evaluations. However, the implementation is not straightforward because systems and work processes require qualification and validation, with consideration also given to security. As a result of the high-throughput, high-volume nature of safety evaluations, computer performance, ergonomics, efficiency, and integration with laboratory information management systems are further key considerations. The European Society of Toxicologic Pathology organized an international expert workshop with participation by toxicologic pathologists, quality assurance/regulatory experts, and information technology experts to discuss qualification and validation of digital histopathology systems in a good laboratory practice environment, and to share the resulting conclusions broadly in the toxicologic pathology community.

Digital quantitative tissue image analysis of hypoxia in resected pancreatic ductal adenocarcinomas

Background

Tumor hypoxia is theorized to contribute to the aggressive biology of pancreatic ductal adenocarcinoma (PDAC). We previously reported that hypoxia correlated with rapid tumor growth and metastasis in patient-derived xenografts. Anticipating a prognostic relevance of hypoxia in patient tumors, we developed protocols for automated semi-quantitative image analysis to provide an objective, observer-independent measure of hypoxia. We further validated this method which can reproducibly estimate pimonidazole-detectable hypoxia in a high-through put manner.

Methods

We studied the performance of three automated image analysis platforms in scoring pimonidazole-detectable hypoxia in resected PDAC (n = 10) in a cohort of patients enrolled in PIMO-PANC. Multiple stained tumor sections were analyzed on three independent image-analysis platforms, Aperio Genie (AG), Definiens Tissue Studio (TS), and Definiens Developer (DD), which comprised of a customized rule set.

Results

The output from Aperio Genie (AG) had good concordance with manual scoring, but the workflow was resource-intensive and not suited for high-throughput analysis. TS analysis had high levels of variability related to misclassification of cells class, while the customized rule set of DD had a high level of reliability with an intraclass coefficient of more than 85%.

Discussion

This work demonstrates the feasibility of developing a robust, high-performance pipeline for an automated, quantitative scoring of pimonidazole-detectable hypoxia in patient tumors.

Evaluation of a Probability-Based Predictive Tool on Pathologist Agreement Using Urinary Bladder as a Pilot Tissue

Simple Summary

There is a common joke in pathology—put three pathologists in a room and you will obtain three different answers. This saying comes from the fact that pathology can be subjective; pathologists’ diagnoses can be influenced by many different biases, and pathologists are also influenced by the presence or absence of animal information and medical history. Compared to pathology, statistics is a much more objective field. This study aimed to develop a probability-based tool using statistics obtained by analyzing 338 histopathology slides of canine and feline urinary bladders, then see if the tool affected agreement between the test pathologists. Four pathologists diagnosed 25 canine and feline bladder slides and they conducted this three times: without animal and clinical information, then with this information, and finally using the probability tool. Results showed large differences in the pathologists’ interpretation of bladder slides, with kappa agreement values (low value for digital slide images, high value for glass slides) of 7–37% without any animal or clinical information, 23–37% with animal signalment and history, and 31–42% when our probability tool was used. This study provides a starting point for the use of probability-based tools in standardizing pathologist agreement in veterinary pathology.

Abstract

Inter-pathologist variation is widely recognized across human and veterinary pathology and is often compounded by missing animal or clinical information on pathology submission forms. Variation in pathologist threshold levels of resident inflammatory cells in the tissue of interest can further decrease inter-pathologist agreement. This study applied a predictive modeling tool to bladder histology slides that were assessed by four pathologists: first without animal and clinical information, then with this information, and finally using the predictive tool. All three assessments were performed twice, using digital whole-slide images (WSI) and then glass slides. Results showed marked variation in pathologists’ interpretation of bladder slides, with kappa agreement values of 7–37% without any animal or clinical information, 23–37% with animal signalment and history, and 31–42% when our predictive tool was applied, for digital WSI and glass slides. The concurrence of test pathologists to the reference diagnosis was 60% overall. This study provides a starting point for the use of predictive modeling in standardizing pathologist agreement in veterinary pathology. It also highlights the importance of high-quality whole-slide imaging to limit the effect of digitization on inter-pathologist agreement and the benefit of continued standardization of tissue assessment in veterinary pathology.

Comparison of manual and automated digital image analysis systems for quantification of cellular protein expression

OBJECTIVE

Quantifying protein expression in immunohistochemically stained histological slides is an important tool for oncologic research. The use of computer-aided evaluation of IHC-stained slides significantly contributes to objectify measurements. Manual digital image analysis (mDIA) requires a user-dependent annotation of the region of interest (ROI). Others have built-in machine learning algorithms with automated digital image analysis (aDIA) and can detect the ROIs automatically. We aimed to investigate the agreement between the results obtained by aDIA and those derived from mDIA systems.

METHODS

We quantified chromogenic intensity (CI) and calculated the positive index (PI) in cohorts of tissue microarrays (TMA) using mDIA and aDIA. To consider the different distributions of staining within cellular sub-compartments and different tumor architecture our study encompassed nuclear and cytoplasmatic stainings in adenocarcinomas and squamous cell carcinomas.

RESULTS

Within all cohorts, we were able to show a high correlation between mDIA and aDIA for the CI (p<0.001) along with high agreement for the PI. Moreover, we were able to show that the cell detections of the programs were comparable as well and both proved to be reliable when compared to manual counting.

CONCLUSION

mDIA and aDIA show a high correlation in acquired IHC data. Both proved to be suitable to stratify patients for evaluation with clinical data. As both produce the same level of information, aDIA might be preferable as it is time-saving, can easily be reproduced, and enables regular and efficient output in large studies in a reasonable time period.

Techniques for digital histological morphometry of the pineal gland

INTRODUCTION

The pineal gland is a small photo-neuro-endocrine organ. This study used human post-mortem pineal glands to microscopically assess immunohistochemical marker intensity and percentage of positivity using known and novel digital techniques.

MATERIALS AND METHODS

An experimental non-inferiority study has been performed on 72 pineal glands harvested from post-mortem examinations. The glands have been stained with glial fibrillary acidic protein (GFAP), synaptophysin (SYN), neuron-specific enolase (NSE), and neurofilament (NF). Slides were digitally scanned. Morphometric data were obtained using optical analysis, CaseViewer, ImageJ, and MorphoRGB RESULTS: Strong and statistically significant correlations were found and plotted using Bland-Altman diagrams between the two image analysis software in the case of mean percentage and intensity of GFAP, NSE, NF, and SYN.

DISCUSSIONS

Software such as SlideViewer and ImageJ, with our novel software MorphoRGB were used to perform histological morphometry of the pineal gland. Digital morphometry of a small organ such as the pineal gland is easy to do by using whole slide imaging (WSI) and digital image analysis software, with potential use in clinical settings. MorphoRGB provides slightly more accurate data than ImageJ and is more user-friendly regarding measurements of parenchyma percentage stained by immunohistochemistry. The results show that MorphoRGB is not inferior in functionality.

CONCLUSIONS

The described morphometric techniques have potential value in current practice, experimental small animal models and human pineal glands, or other small endocrine organs that can be fully included in a whole slide image. The software we used has applications in quantifying immunohistochemical stains.

A Method for Unsupervised Semi-Quantification of Inmunohistochemical Staining with Beta Divergences

In many research laboratories, it is essential to determine the relative expression levels of some proteins of interest in tissue samples. The semi-quantitative scoring of a set of images consists of establishing a scale of scores ranging from zero or one to a maximum number set by the researcher and assigning a score to each image that should represent some predefined characteristic of the IHC staining, such as its intensity. However, manual scoring depends on the judgment of an observer and therefore exposes the assessment to a certain level of bias. In this work, we present a fully automatic and unsupervised method for comparative biomarker quantification in histopathological brightfield images. The method relies on a color separation method that discriminates between two chromogens expressed as brown and blue colors robustly, independent of color variation or biomarker expression level. For this purpose, we have adopted a two-stage stain separation approach in the optical density space. First, a preliminary separation is performed using a deconvolution method in which the color vectors of the stains are determined after an eigendecomposition of the data. Then, we adjust the separation using the non-negative matrix factorization method with beta divergences, initializing the algorithm with the matrices resulting from the previous step. After that, a feature vector of each image based on the intensity of the two chromogens is determined. Finally, the images are annotated using a systematically initialized k-means clustering algorithm with beta divergences. The method clearly defines the initial boundaries of the categories, although some flexibility is added. Experiments for the semi-quantitative scoring of images in five categories have been carried out by comparing the results with the scores of four expert researchers yielding accuracies that range between 76.60% and 94.58%. These results show that the proposed automatic scoring system, which is definable and reproducible, produces consistent results.

Rheumatoid Arthritis Synovial Inflammation Quantification Using Computer Vision

OBJECTIVE

We quantified inflammatory burden in rheumatoid arthritis (RA) synovial tissue by using computer vision to automate the process of counting individual nuclei in hematoxylin and eosin images.

METHODS

We adapted and applied computer vision algorithms to quantify nuclei density (count of nuclei per unit area of tissue) on synovial tissue from arthroplasty samples. A pathologist validated algorithm results by labeling nuclei in synovial images that were mislabeled or missed by the algorithm. Nuclei density was compared with other measures of RA inflammation such as semiquantitative histology scores, gene-expression data, and clinical measures of disease activity.

RESULTS

The algorithm detected a median of 112,657 (range 8,160-821,717) nuclei per synovial sample. Based on pathologist-validated results, the sensitivity and specificity of the algorithm was 97% and 100%, respectively. The mean nuclei density calculated by the algorithm was significantly higher (P < 0.05) in synovium with increased histology scores for lymphocytic inflammation, plasma cells, and lining hyperplasia. Analysis of RNA sequencing identified 915 significantly differentially expressed genes in correlation with nuclei density (false discovery rate is less than 0.05). Mean nuclei density was significantly higher (P < 0.05) in patients with elevated levels of C-reactive protein, erythrocyte sedimentation rate, rheumatoid factor, and cyclized citrullinated protein antibody.

CONCLUSION

Nuclei density is a robust measurement of inflammatory burden in RA and correlates with multiple orthogonal measurements of inflammation.

IL-MCAM: An interactive learning and multi-channel attention mechanism-based weakly supervised colorectal histopathology image classification approach

In recent years, colorectal cancer has become one of the most significant diseases that endanger human health. Deep learning methods are increasingly important for the classification of colorectal histopathology images. However, existing approaches focus more on end-to-end automatic classification using computers rather than human-computer interaction. In this paper, we propose an IL-MCAM framework. It is based on attention mechanisms and interactive learning. The proposed IL-MCAM framework includes two stages: automatic learning (AL) and interactivity learning (IL). In the AL stage, a multi-channel attention mechanism model containing three different attention mechanism channels and convolutional neural networks is used to extract multi-channel features for classification. In the IL stage, the proposed IL-MCAM framework continuously adds misclassified images to the training set in an interactive approach, which improves the classification ability of the MCAM model. We carried out a comparison experiment on our dataset and an extended experiment on the HE-NCT-CRC-100K dataset to verify the performance of the proposed IL-MCAM framework, achieving classification accuracies of 98.98% and 99.77%, respectively. In addition, we conducted an ablation experiment and an interchangeability experiment to verify the ability and interchangeability of the three channels. The experimental results show that the proposed IL-MCAM framework has excellent performance in the colorectal histopathological image classification tasks.

Proteomic-Based Machine Learning Analysis Reveals PYGB as a Novel Immunohistochemical Biomarker to Distinguish Inverted Urothelial Papilloma From Low-Grade Papillary Urothelial Carcinoma With Inverted Growth

Background

The molecular biology of inverted urothelial papilloma (IUP) as a precursor disease of urothelial carcinoma is poorly understood. Furthermore, the overlapping histology between IUP and papillary urothelial carcinoma (PUC) with inverted growth is a diagnostic pitfall leading to frequent misdiagnoses.

Methods

To identify the oncologic significance of IUP and discover a novel biomarker for its diagnosis, we employed mass spectrometry-based proteomic analysis of IUP, PUC, and normal urothelium (NU). Machine learning analysis shortlisted candidate proteins, while subsequent immunohistochemical validation was performed in an independent sample cohort.

Results

From the overall proteomic landscape, we found divergent ‘NU-like’ (low-risk) and ‘PUC-like’ (high-risk) signatures in IUP. The latter were characterized by altered metabolism, biosynthesis, and cell–cell interaction functions, indicating oncologic significance. Further machine learning-based analysis revealed SERPINH1, PKP2, and PYGB as potential diagnostic biomarkers discriminating IUP from PUC. The immunohistochemical validation confirmed PYGB as a specific biomarker to distinguish between IUP and PUC with inverted growth.

Conclusion

In conclusion, we suggest PYGB as a promising immunohistochemical marker for IUP diagnosis in routine practice.

A Study of Thyroid Fine Needle Aspiration of Follicular Adenoma in the "Atypia of Undetermined Significance" Bethesda Category Using Digital Image Analysis

Background

Originally designed for computerized image analysis, ThinPrep is underutilized in that role outside gynecological cytology. It can be used to address the inter/intra-observer variability in the evaluation of thyroid fine needle aspiration (TFNA) biopsy and help pathologists to gain additional insight into thyroid cytomorphology.

Methods

We designed and validated a feature engineering and supervised machine learning-based digital image analysis method using ImageJ and Python scikit-learn . The method was trained and validated from 400 low power (100x) and 400 high power (400x) images generated from 40 TFNA cases.

Result

The area under the curve (AUC) for receiver operating characteristics (ROC) is 0.75 (0.74–0.82) for model based from low-power images and 0.74 (0.69–0.79) for the model based from high-power images. Cytomorphologic features were synthesized using feature engineering and when performed in isolation, they achieved AUC of 0.71 (0.64–0.77) for chromatin, 0.70 (0.64–0.73) for cellularity, 0.65 (0.60–0.69) for cytoarchitecture, 0.57 (0.51–0.61) for nuclear size, and 0.63 (0.57–0.68) for nuclear shape.

Conclusion

Our study proves that ThinPrep is an excellent preparation method for digital image analysis of thyroid cytomorphology. It can be used to quantitatively harvest morphologic information for diagnostic purpose.

Deep Learning Accurately Quantifies Plasma Cell Percentages on CD138-Stained Bone Marrow Samples

The diagnosis of plasma cell neoplasms requires accurate, and ideally precise, percentages. This plasma cell percentage is often determined by visual estimation of CD138-stained bone marrow biopsies and clot sections. While not necessarily inaccurate, estimates are by definition imprecise. For this study, we hypothesized that deep learning can be used to improve precision. We trained a semantic segmentation-based convolutional neural network (CNN) using annotations of CD138+ and CD138- cells provided by one pathologist on small image patches of bone marrow and validated the CNN on an independent test set of image patches using annotations from two pathologists and a non-deep learning commercial software. On validation, we found that the intraclass correlation coefficients for plasma cell percentages between the CNN and pathologist #1, a non-deep learning commercial software and pathologist #1, and pathologists #1 and #2 were 0.975, 0.892, and 0.994, respectively. The overall results show that CNN labels were almost as accurate as pathologist labels at a cell-by-cell level. Once satisfied with performance, we scaled-up the CNN to evaluate whole slide images (WSIs), and deployed the system as a workflow friendly web application to measure plasma cell percentages using snapshots taken from microscope cameras.

Research-Relevant Conditions and Pathology of Laboratory Mice, Rats, Gerbils, Guinea Pigs, Hamsters, Naked Mole Rats, and Rabbits

Animals are valuable resources in biomedical research in investigations of biological processes, disease pathogenesis, therapeutic interventions, safety, toxicity, and carcinogenicity. Interpretation of data from animals requires knowledge not only of the processes or diseases (pathophysiology) under study but also recognition of spontaneous conditions and background lesions (pathology) that can influence or confound the study results. Species, strain/stock, sex, age, anatomy, physiology, spontaneous diseases (noninfectious and infectious), and neoplasia impact experimental results and interpretation as well as animal welfare. This review and the references selected aim to provide a pathology resource for researchers, pathologists, and veterinary personnel who strive to achieve research rigor and validity and must understand the spectrum of "normal" and expected conditions to accurately identify research-relevant experimental phenotypes as well as unusual illness, pathology, or other conditions that can compromise studies involving laboratory mice, rats, gerbils, guinea pigs, hamsters, naked mole rats, and rabbits.

Activation of angiotensin-converting enzyme 2 ameliorates metabolic syndrome-induced renal damage in rats by renal TLR4 and nuclear transcription factor κB downregulation

BACKGROUND

Metabolic syndrome (MetS) is an independent risk factor for chronic kidney disease (CKD) through many mechanisms, including activation of the renin-angiotensin system. The deleterious effects of angiotensin II (Ang II) can be counterbalanced by angiotensin-converting enzyme 2 (ACE2). Diminazene aceturate (DIZE), an anti-trypanosomal drug, can activate ACE2.

OBJECTIVE

This study aimed to investigate the possible reno-protective effects of DIZE in MetS rats with elucidation of related mechanisms.

MATERIALS AND METHODS

Thirty adult male Wistar albino rats were divided equally into control, MetS, and MetS + DIZE groups. Body weight, systolic blood pressure (SBP), and urinary albumin levels were measured. Serum levels of fasting blood glucose (FBG), insulin, uric acid, lipid profile, urea, and creatinine were measured. Homeostasis Model Assessment Index (HOMA-IR) was estimated. Subsequently, renal levels of ACE2, Ang II, malondialdehyde (MDA), reduced glutathione (GSH), and tumor necrosis factor-α (TNF-α) were measured with histopathological and immunohistochemical assessment of TLR4 and NF-κB in renal tissues.

RESULTS

MetS caused dyslipidemia with significant increases in body weight, SBP, FBG, serum insulin, HOMA-IR, uric acid, urea, creatinine, urinary albumin, and renal levels of Ang II, MDA, and TNF-α, whereas renal ACE2 and GSH were significantly decreased. Renal TLR4 and NF-κB immunoreactivity in MetS rats was upregulated. DIZE supplementation of MetS rats induced significant improvements in renal function parameters; this could be explained by the ability of DIZE to activate renal ACE2 and decrease renal Ang II levels with downregulation of renal TLR4 and NF-κB expression.

CONCLUSION

DIZE exerts a reno-protective effect in MetS, mainly by downregulating renal TLR4 and NF-κB levels.

Immunohistochemistry scoring of breast tumor tissue microarrays: A comparison study across three software applications

Digital pathology can efficiently assess immunohistochemistry (IHC) data on tissue microarrays (TMAs). Yet, it remains important to evaluate the comparability of the data acquired by different software applications and validate it against pathologist manual interpretation. In this study, we compared the IHC quantification of 5 clinical breast cancer biomarkers-estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), and cytokeratin 5/6 (CK5/6)-across 3 software applications (Definiens Tissue Studio, inForm, and QuPath) and benchmarked the results to pathologist manual scores. IHC expression for each marker was evaluated across 4 TMAs consisting of 935 breast tumor tissue cores from 367 women within the Nurses' Health Studies; each women contributing three 0.6-mm cores. The correlation and agreement between manual and software-derived results were primarily assessed using Spearman's ρ, percentage agreement, and area under the curve (AUC). At the TMA core-level, the correlations between manual and software-derived scores were the highest for HER2 (ρ ranging from 0.75 to 0.79), followed by ER (0.69-0.71), PR (0.67-0.72), CK5/6 (0.43-0.47), and EGFR (0.38-0.45). At the case-level, there were good correlations between manual and software-derived scores for all 5 markers (ρ ranging from 0.43 to 0.82), where QuPath had the highest correlations. Software-derived scores were highly comparable to each other (ρ ranging from 0.80 to 0.99). The average percentage agreements between manual and software-derived scores were excellent for ER (90.8%-94.5%) and PR (78.2%-85.2%), moderate for HER2 (65.4%-77.0%), highly variable for EGFR (48.2%-82.8%), and poor for CK5/6 (22.4%-45.0%). All AUCs across markers and software applications were ≥0.83. The 3 software applications were highly comparable to each other and to manual scores in quantifying these 5 markers. QuPath consistently produced the best performance, indicating this open-source software is an excellent alternative for future use.

Comparison of Semi-Quantitative Scoring and Artificial Intelligence Aided Digital Image Analysis of Chromogenic Immunohistochemistry

Semi-quantitative scoring is a method that is widely used to estimate the quantity of proteins on chromogen-labelled immunohistochemical (IHC) tissue sections. However, it suffers from several disadvantages, including its lack of objectivity and the fact that it is a time-consuming process. Our aim was to test a recently established artificial intelligence (AI)-aided digital image analysis platform, Pathronus, and to compare it to conventional scoring by five observers on chromogenic IHC-stained slides belonging to three experimental groups. Because Pathronus operates on grayscale 0-255 values, we transformed the data to a seven-point scale for use by pathologists and scientists. The accuracy of these methods was evaluated by comparing statistical significance among groups with quantitative fluorescent IHC reference data on subsequent tissue sections. The pairwise inter-rater reliability of the scoring and converted Pathronus data varied from poor to moderate with Cohen’s kappa, and overall agreement was poor within every experimental group using Fleiss’ kappa. Only the original and converted that were obtained from Pathronus original were able to reproduce the statistical significance among the groups that were determined by the reference method. In this study, we present an AI-aided software that can identify cells of interest, differentiate among organelles, protein specific chromogenic labelling, and nuclear counterstaining after an initial training period, providing a feasible and more accurate alternative to semi-quantitative scoring.

QuPath Digital Immunohistochemical Analysis of Placental Tissue

Background:

QuPath is an open-source digital image analyzer notable for its user-friendly design, cross-platform compatibility, and customizable functionality. Since it was first released in 2016, at least 624 publications have reported its use, and it has been applied in a wide spectrum of settings. However, there are currently limited reports of its use in placental tissue. Here, we present the use of QuPath to quantify staining of G-protein coupled receptor 18 (GPR18), the receptor for the pro-resolving lipid mediator Resolvin D2, in placental tissue.

Methods:

Whole slide images of vascular smooth muscle (VSM) and extravillous trophoblast (EVT) cells stained for GPR18 were annotated for areas of interest. Visual scoring was performed on these images by trained and in-training pathologists, while QuPath scoring was performed with the methodology described herein.

Results:

Bland–Altman analyses showed that, for the VSM category, the two methods were comparable across all staining levels. For EVT cells, the high-intensity staining level was comparable across methods, but the medium and low staining levels were not comparable.

Conclusions:

Digital image analysis programs offer great potential to revolutionize pathology practice and research by increasing accuracy and decreasing the time and cost of analysis. Careful study is needed to optimize this methodology further.

Deep Learning Approaches and Applications in Toxicologic Histopathology: Current Status and Future Perspectives

Whole slide imaging enables the use of a wide array of digital image analysis tools that are revolutionizing pathology. Recent advances in digital pathology and deep convolutional neural networks have created an enormous opportunity to improve workflow efficiency, provide more quantitative, objective, and consistent assessments of pathology datasets, and develop decision support systems. Such innovations are already making their way into clinical practice. However, the progress of machine learning - in particular, deep learning (DL) - has been rather slower in nonclinical toxicology studies. Histopathology data from toxicology studies are critical during the drug development process that is required by regulatory bodies to assess drug-related toxicity in laboratory animals and its impact on human safety in clinical trials. Due to the high volume of slides routinely evaluated, low-throughput, or narrowly performing DL methods that may work well in small-scale diagnostic studies or for the identification of a single abnormality are tedious and impractical for toxicologic pathology. Furthermore, regulatory requirements around good laboratory practice are a major hurdle for the adoption of DL in toxicologic pathology. This paper reviews the major DL concepts, emerging applications, and examples of DL in toxicologic pathology image analysis. We end with a discussion of specific challenges and directions for future research.

Artificial Intelligence in Toxicological Pathology: Quantitative Evaluation of Compound-Induced Follicular Cell Hypertrophy in Rat Thyroid Gland Using Deep Learning Models

Digital pathology has recently been more broadly deployed, fueling artificial intelligence (AI) application development and more systematic use of image analysis. Here, two different AI models were developed to evaluate follicular cell hypertrophy in hematoxylin and eosin-stained whole-slide-images of rat thyroid gland, using commercial AI-based-software. In the first, mean cytoplasmic area measuring approach (MCA approach), mean cytoplasmic area was calculated via several sequential deep learning (DL)-based algorithms including segmentation in microanatomical structures (separation of colloid and stroma from thyroid follicular epithelium), nuclear detection, and area measurements. With our additional second, hypertrophy area fraction predicting approach (HAF approach), we present for the first time DL-based direct detection of the histopathological change follicular cell hypertrophy in the thyroid gland with similar results. For multiple studies, increased output parameters (mean cytoplasmic area and hypertrophic area fraction) were shown in groups given different hypertrophy-inducing reference compounds in comparison to control groups. Quantitative results correlated with the gold standard of board-certified veterinary pathologists' diagnoses and gradings as well as thyroid hormone dependent gene expressions. Accuracy and repeatability of diagnoses and grading by pathologists are expected to be improved by additional evaluation of mean cytoplasmic area or direct detection of hypertrophy, combined with standard histopathological observations.

Reduced tissue and serum resistin expression as a clinical marker for esophageal squamous cell carcinoma

Esophageal cancer is one of the most common malignancies and leading cause of cancer-associated mortality worldwide. However, the molecular mechanisms underlying esophageal cancer progression and the development of clinical tools for effective diagnosis remain unclear. Resistin, which was originally identified as an adipose tissue-secretory factor, has been associated with obesity-related diseases, including certain types of cancer. Thus, the present study aimed to investigate the expression levels of resistin in tissue and serum specimens from patients with esophageal squamous cell carcinoma (ESCC) to determine the potential biological effects of resistin on ESCC cells. The results demonstrated that both tissue and serum resistin levels were significantly lower in patients with ESCC compared with healthy controls. In addition, resistin expression was positively associated with the body mass index of patients with ESCC. In vitro studies revealed that resistin inhibited the migratory ability of ESCC cells, while having no effect on ESCC cell proliferation. Taken together, these results suggest that resistin may have the potential to be developed into a clinical marker for ESCC. However, further studies are required to investigate resistin receptor expression and determine the potential involvement of resistin-associated biological pathways, which may provide insight for future development of targeted therapies for resistin-mediated ESCC.