This review covers computer-assisted analysis of images in the field of medical imaging. Recent advances in machine learning, especially with regard to deep learning, are helping to identify, classify, and quantify patterns in medical images. At the core of these advances is the ability to exploit hierarchical feature representations learned solely from data, instead of features designed by hand according to domain-specific knowledge. Deep learning is rapidly becoming the state of the art, leading to enhanced performance in various medical applications. We introduce the fundamentals of deep learning methods and review their successes in image registration, detection of anatomical and cellular structures, tissue segmentation, computer-aided disease diagnosis and prognosis, and so on. We conclude by discussing research issues and suggesting future directions for further improvement.

The ubiquity and high bioavailability of microplastics have an unknown risk on the marine environment. Biomonitoring should be used to investigate biotic impacts of microplastic exposure. While many studies have used mussels as indicators for marine microplastic pollution, a robust and clear justification for their selection as indicator species is still lacking. Here, we review published literature from field investigations and laboratory experiments on microplastics in mussels and critically discuss the suitability and challenges of mussels as bioindicator for microplastic pollution. Mussels are suitable bioindicator for microplastic pollution because of their wide distribution, vital ecological niches, susceptibility to microplastic uptake and close connection with marine predators and human health. Field investigations highlight a wide occurrence of microplastics in mussels from all over the world, yet their abundance varies enormously. Problematically, these studies are not comparable due to the lack of a standardized approach, as well as temporal and spatial variability. Interestingly, microplastic abundance in field-collected mussels is closely related to human activity, and there is evidence for a positive and quantitative correlation between microplastics in mussels and surrounding waters. Laboratory studies collectively demonstrate that mussels may be good model organisms in revealing microplastic uptake, accumulation and toxicity. Consequently, we propose the use of mussels as target species to monitor microplastics and call for a uniform, efficient and economical approach that is suitable for a future large-scale monitoring program.Copyright © 2018 Elsevier Ltd. All rights reserved.

Mussel beds along the wave-exposed coast of the eastern North Pacific Ocean serve as an important habitat, harboring a high diversity of species. A comparison of California mussel bed community diversity in 2002 to historical data (1960s to 1970s) revealed large declines (mean loss 58.9%), including some declines >141 species (approximately 80% loss). Concurrent work revealed inconsistent changes in mussel populations (biomass and bed thickness) along the California coast, suggesting that diversity declines may be related to large-scale processes rather than local habitat destruction. Potential factors causing declines in mussel community diversity are discussed, with regional climate change associated with the Pacific Decadal Oscillation and climate change induced alterations of ecological interactions and biological processes suggested as likely causes. Although extensive literature has predicted the potential effects of climate change on global diversity, this study is one of the few examples of declines attributed to climate change.

The Gulf of La Spezia (northern Tyrrhenian Sea, Italy) is a commercially important area both as a shipping port and for mussel farming. Recently, there has been increased concern over environmental disturbances caused by anthropogenic activities such as ship traffic and dredging and the effects they have on the health of farmed mussels. This paper reports the results of microbiological and histological analyses, as well as of measurement of several biomarkers which were performed to assess the health status of mussels (Mytilus galloprovincialis) from four rearing sites in the Gulf of La Spezia. Mussels were collected between October 2015 and September 2016 and histological analyses (including gonadal maturation stage), as well as the presence of pathogenic bacteria (Vibrio splendidus clade, V. aestuarianus and V. harveyi), viruses (Herpes virus and ostreid Herpes virus 1) and protozoa (Marteilia spp., in the summer season only) were carried out on a monthly basis. Conversely, biomarker responses in haemocyte/haemolymph (total haemocyte count, haemocyte diameter and volume, lysozyme and lactate dehydrogenase activities in cell-free haemolymph, and micronuclei frequency) and in gills and digestive gland (cortisol-like steroids and lipid peroxidation levels), were evaluated bimonthly. Microbiological data indicated that mussels contain a reservoir of potentially pathogenic bacteria, viruses and protozoa that in certain environmental conditions may cause a weakening of the immune system of animals leading to mortality episodes. The percentage of parasites detected in the mussels was generally low (9.6% for Steinhausia mytilovum, that is 17 samples out of 177 examined females; 3.4% for Proctoeces maculatus; 0.9% for Mytilicola intestinalis and 2% for ciliated protozoa), while symbiont loads were higher (31% for Eugymnanthea inquilina and Urastoma cyprinae). Interestingly, a previously undescribed haplosporidian was detected in a single mussel sample (0.2%) and was confirmed by in situ hybridization. Cells morphologically similar to Perkinsus sp. trophozoites were observed in 0.7% of the mussels analysed; however, infection with Perkinsus spp. could neither be confirmed by ISH nor by PCR. Different pathological aspects, such as host defence responses and regressive/progressive changes were detected in the gills, digestive glands, gonads and mantle. Only one single case of disseminated neoplasia (0.2%) was observed. As for the biomarker evaluation, the MANOVA analysis revealed the statistically significant effect that the variable "sampling site" had on the biological parameter measured, thus suggesting that the multibiomarker approach was able to differentiate the rearing sites.Copyright © 2018 Elsevier Inc. All rights reserved.

Bivalve histopathology has become an important tool in aquatic toxicology, having been implemented in many biomonitoring programmes worldwide. However, there are various gaps in the knowledge of many sentinel organisms and the interference of confounding factors. This work aimed (i) to develop a detailed semi-quantitative histopathological index of the digestive gland and gonad of the Mytilus galloprovincialis mussel collected from five sites contaminated with distinct patterns of organic and inorganic toxicants along the Basque coast (SE Bay of Biscay) and (ii) to investigate whether seasonal variability and parasitosis act as confounding factors. A total of twenty-three histopathological alterations were analysed in the digestive gland and gonad following a weighed condition index approach. The alterations were integrated into a single value for a better understanding of the mussels' health status. The digestive gland was consistently more damaged than the gonad. Mussels from the most impacted sites endured the most significant deleterious effects showing inflammation-related alterations together with digestive tubule atrophy and necrosis. Neoplastic diseases were scarce, with only a few cases of fibromas (benign neoplasia). In contrast, in moderately or little impacted sites, contamination levels did not cause significant tissue damage. However, parasites contributed to overestimating the values of histopathological indices (i.e. more severe tissue damage) in mussels from little impacted sites, whilst the opposite occurred in mussels from highly polluted sites. Accordingly, inter-site differences were more pronounced in autumn when natural physiological responses of advanced maturation stages did not interfere in the histological response. In conclusion, although seasonal variability and parasitosis mask the response of histopathological indices, this biomonitoring approach may provide good sensitivity for assessing the health status of mussels if fluctuations of these confounding factors are considered. Copyright © 2015 Elsevier B.V. All rights reserved.

Comparative pathology as a scientific discipline studies animal diseases in relation to their aetiology, pathogenesis and prognosis. Among the main aspects of this discipline, regressive changes, host defense responses with pathological implications and progressive changes, represent the majority of the possible responses of cells and tissues to pathogens and exposure to chemicals. One of the most persistent issues in the field of invertebrate pathology is the variability in terminology and definition, which has led to confusion in scientific communication. The aim of this paper is to provide an overview of the pathological basis of bivalve disease (defensive, regressive and progressive phenomena) and contribute to the standardised terminology for bivalve molluscan disease in the context of comparative pathology. Copyright © 2015 Elsevier Inc. All rights reserved.

During the diagnostic workup of lung adenocarcinomas (LAC), pathologists evaluate distinct histological tumor growth patterns. The percentage of each pattern on multiple slides bears prognostic significance. To assist with the quantification of growth patterns, we constructed a pipeline equipped with a convolutional neural network (CNN) and soft-voting as the decision function to recognize solid, micropapillary, acinar, and cribriform growth patterns, and non-tumor areas. Slides of primary LAC were obtained from Cedars-Sinai Medical Center (CSMC), the Military Institute of Medicine in Warsaw and the TCGA portal. Several CNN models trained with 19,924 image tiles extracted from 78 slides (MIMW and CSMC) were evaluated on 128 test slides from the three sites by F1-score and accuracy using manual tumor annotations by pathologist. The best CNN yielded F1-scores of 0.91 (solid), 0.76 (micropapillary), 0.74 (acinar), 0.6 (cribriform), and 0.96 (non-tumor) respectively. The overall accuracy of distinguishing the five tissue classes was 89.24%. Slide-based accuracy in the CSMC set (88.5%) was significantly better (p < 2.3E-4) than the accuracy in the MIMW (84.2%) and TCGA (84%) sets due to superior slide quality. Our model can work side-by-side with a pathologist to accurately quantify the percentages of growth patterns in tumors with mixed LAC patterns.

Classifying breast cancer histopathological images automatically is an important task in computer assisted pathology analysis. However, extracting informative and non-redundant features for histopathological image classification is challenging due to the appearance variability caused by the heterogeneity of the disease, the tissue preparation, and staining processes. In this paper, we propose a new feature extractor, called deep manifold preserving autoencoder, to learn discriminative features from unlabeled data. Then, we integrate the proposed feature extractor with a softmax classifier to classify breast cancer histopathology images. Specifically, it learns hierarchal features from unlabeled image patches by minimizing the distance between its input and output, and simultaneously preserving the geometric structure of the whole input data set. After the unsupervised training, we connect the encoder layers of the trained deep manifold preserving autoencoder with a softmax classifier to construct a cascade model and fine-tune this deep neural network with labeled training data. The proposed method learns discriminative features by preserving the structure of the input datasets from the manifold learning view and minimizing reconstruction error from the deep learning view from a large amount of unlabeled data. Extensive experiments on the public breast cancer dataset (BreaKHis) demonstrate the effectiveness of the proposed method.

Deep learning algorithms, in particular convolutional networks, have rapidly become a methodology of choice for analyzing medical images. This paper reviews the major deep learning concepts pertinent to medical image analysis and summarizes over 300 contributions to the field, most of which appeared in the last year. We survey the use of deep learning for image classification, object detection, segmentation, registration, and other tasks. Concise overviews are provided of studies per application area: neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal, musculoskeletal. We end with a summary of the current state-of-the-art, a critical discussion of open challenges and directions for future research.Copyright © 2017 Elsevier B.V. All rights reserved.

The analysis of glandular morphology within colon histopathology images is an important step in determining the grade of colon cancer. Despite the importance of this task, manual segmentation is laborious, time-consuming and can suffer from subjectivity among pathologists. The rise of computational pathology has led to the development of automated methods for gland segmentation that aim to overcome the challenges of manual segmentation. However, this task is non-trivial due to the large variability in glandular appearance and the difficulty in differentiating between certain glandular and non-glandular histological structures. Furthermore, a measure of uncertainty is essential for diagnostic decision making. To address these challenges, we propose a fully convolutional neural network that counters the loss of information caused by max-pooling by re-introducing the original image at multiple points within the network. We also use atrous spatial pyramid pooling with varying dilation rates for preserving the resolution and multi-level aggregation. To incorporate uncertainty, we introduce random transformations during test time for an enhanced segmentation result that simultaneously generates an uncertainty map, highlighting areas of ambiguity. We show that this map can be used to define a metric for disregarding predictions with high uncertainty. The proposed network achieves state-of-the-art performance on the GlaS challenge dataset and on a second independent colorectal adenocarcinoma dataset. In addition, we perform gland instance segmentation on whole-slide images from two further datasets to highlight the generalisability of our method. As an extension, we introduce MILD-Net for simultaneous gland and lumen segmentation, to increase the diagnostic power of the network.Copyright © 2018 Elsevier B.V. All rights reserved.

The development of deep neural networks is facilitating more advanced digital analysis of histopathologic images. We trained a convolutional neural network for multiclass segmentation of digitized kidney tissue sections stained with periodic acid-Schiff (PAS).We trained the network using multiclass annotations from 40 whole-slide images of stained kidney transplant biopsies and applied it to four independent data sets. We assessed multiclass segmentation performance by calculating Dice coefficients for ten tissue classes on ten transplant biopsies from the Radboud University Medical Center in Nijmegen, The Netherlands, and on ten transplant biopsies from an external center for validation. We also fully segmented 15 nephrectomy samples and calculated the network's glomerular detection rates and compared network-based measures with visually scored histologic components (Banff classification) in 82 kidney transplant biopsies.The weighted mean Dice coefficients of all classes were 0.80 and 0.84 in ten kidney transplant biopsies from the Radboud center and the external center, respectively. The best segmented class was "glomeruli" in both data sets (Dice coefficients, 0.95 and 0.94, respectively), followed by "tubuli combined" and "interstitium." The network detected 92.7% of all glomeruli in nephrectomy samples, with 10.4% false positives. In whole transplant biopsies, the mean intraclass correlation coefficient for glomerular counting performed by pathologists versus the network was 0.94. We found significant correlations between visually scored histologic components and network-based measures.This study presents the first convolutional neural network for multiclass segmentation of PAS-stained nephrectomy samples and transplant biopsies. Our network may have utility for quantitative studies involving kidney histopathology across centers and provide opportunities for deep learning applications in routine diagnostics.Copyright © 2019 by the American Society of Nephrology.

Visual inspection of histopathology images of stained biopsy tissue by expert pathologists is the standard method for grading of prostate cancer (PCa). However, this process is time-consuming and subject to high inter-observer variability. Machine learning-based methods have the potential to improve efficient throughput of large volumes of slides while decreasing variability, but they are not easy to develop because they require substantial amounts of labeled training data. In this paper, we propose a deep learning-based classification technique and data augmentation methods for accurate grading of PCa in histopathology images in the presence of limited data. Our method combines the predictions of three separate convolutional neural networks (CNNs) that work with different patch sizes. This enables our method to take advantage of the greater amount of contextual information in larger patches as well as greater quantity of smaller patches in the labeled training data. The predictions produced by the three CNNs are combined using a logistic regression model, which is trained separately after the CNN training. To effectively train our models, we propose new data augmentation methods and empirically study their effects on the classification accuracy. The proposed method achieves an accuracy of [Formula: see text] in classifying cancerous patches versus benign patches and an accuracy of [Formula: see text] in classifying low-grade (i.e., Gleason grade 3) from high-grade (i.e., Gleason grades 4 and 5) patches. The agreement level of our automatic grading method with expert pathologists is within the range of agreement between pathologists. Our experiments indicate that data augmentation is necessary for achieving expert-level performance with deep learning-based methods. A combination of image-space augmentation and feature-space augmentation leads to the best results. Our study shows that well-designed and properly trained deep learning models can achieve PCa Gleason grading accuracy that is comparable to an expert pathologist.

Colorectal adenocarcinoma originating in intestinal glandular structures is the most common form of colon cancer. In clinical practice, the morphology of intestinal glands, including architectural appearance and glandular formation, is used by pathologists to inform prognosis and plan the treatment of individual patients. However, achieving good inter-observer as well as intra-observer reproducibility of cancer grading is still a major challenge in modern pathology. An automated approach which quantifies the morphology of glands is a solution to the problem. This paper provides an overview to the Gland Segmentation in Colon Histology Images Challenge Contest (GlaS) held at MICCAI'2015. Details of the challenge, including organization, dataset and evaluation criteria, are presented, along with the method descriptions and evaluation results from the top performing methods.Copyright © 2016 Elsevier B.V. All rights reserved.

In this review, a wide array of bioaccumulation markers and biomarkers, used to demonstrate exposure to and effects of environmental contaminants, has been discussed in relation to their feasibility in environmental risk assessment (ERA). Fish bioaccumulation markers may be applied in order to elucidate the aquatic behavior of environmental contaminants, as bioconcentrators to identify certain substances with low water levels and to assess exposure of aquatic organisms. Since it is virtually impossible to predict the fate of xenobiotic substances with simple partitioning models, the complexity of bioaccumulation should be considered, including toxicokinetics, metabolism, biota-sediment accumulation factors (BSAFs), organ-specific bioaccumulation and bound residues. Since it remains hard to accurately predict bioaccumulation in fish, even with highly sophisticated models, analyses of tissue levels are required. The most promising fish bioaccumulation markers are body burdens of persistent organic pollutants, like PCBs and DDTs. Since PCDD and PCDF levels in fish tissues are very low as compared with the sediment levels, their value as bioaccumulation markers remains questionable. Easily biodegradable compounds, such as PAHs and chlorinated phenols, do not tend to accumulate in fish tissues in quantities that reflect the exposure. Semipermeable membrane devices (SPMDs) have been successfully used to mimic bioaccumulation of hydrophobic organic substances in aquatic organisms. In order to assess exposure to or effects of environmental pollutants on aquatic ecosystems, the following suite of fish biomarkers may be examined: biotransformation enzymes (phase I and II), oxidative stress parameters, biotransformation products, stress proteins, metallothioneins (MTs), MXR proteins, hematological parameters, immunological parameters, reproductive and endocrine parameters, genotoxic parameters, neuromuscular parameters, physiological, histological and morphological parameters. All fish biomarkers are evaluated for their potential use in ERA programs, based upon six criteria that have been proposed in the present paper. This evaluation demonstrates that phase I enzymes (e.g. hepatic EROD and CYP1A), biotransformation products (e.g. biliary PAH metabolites), reproductive parameters (e.g. plasma VTG) and genotoxic parameters (e.g. hepatic DNA adducts) are currently the most valuable fish biomarkers for ERA. The use of biomonitoring methods in the control strategies for chemical pollution has several advantages over chemical monitoring. Many of the biological measurements form the only way of integrating effects on a large number of individual and interactive processes in aquatic organisms. Moreover, biological and biochemical effects may link the bioavailability of the compounds of interest with their concentration at target organs and intrinsic toxicity. The limitations of biomonitoring, such as confounding factors that are not related to pollution, should be carefully considered when interpreting biomarker data. Based upon this overview there is little doubt that measurements of bioaccumulation and biomarker responses in fish from contaminated sites offer great promises for providing information that can contribute to environmental monitoring programs designed for various aspects of ERA.