Automated image analysis for high-content screening and analysis

Modelling cell shape in 3D structured environments: A quantitative comparison with experiments

Cell shape plays a fundamental role in many biological processes, including adhesion, migration, division and development, but it is not clear which shape model best predicts three-dimensional cell shape in structured environments. Here, we compare different modelling approaches with experimental data. The shapes of single mesenchymal cells cultured in custom-made 3D scaffolds were compared by a Fourier method with surfaces that minimize area under the given adhesion and volume constraints. For the minimized surface model, we found marked differences to the experimentally observed cell shapes, which necessitated the use of more advanced shape models. We used different variants of the cellular Potts model, which effectively includes both surface and bulk contributions. The simulations revealed that the Hamiltonian with linear area energy outperformed the elastic area constraint in accurately modelling the 3D shapes of cells in structured environments. Explicit modelling the nucleus did not improve the accuracy of the simulated cell shapes. Overall, our work identifies effective methods for accurately modelling cellular shapes in complex environments.

FocA: A deep learning tool for reliable, near-real-time imaging focus analysis in automated cell assay pipelines

The increasing use of automation in cellular assays and cell culture presents significant opportunities to enhance the scale and throughput of imaging assays, but to do so, reliable data quality and consistency are critical. Realizing the full potential of automation will thus require the design of robust analysis pipelines that span the entire workflow in question. Here we present FocA, a deep learning tool that, in near real-time, identifies in-focus and out-of-focus images generated on a fully automated cell biology research platform, the NYSCF Global Stem Cell Array®. The tool is trained on small patches of downsampled images to maximize computational efficiency without compromising accuracy, and optimized to make sure no sub-quality images are stored and used in downstream analyses. The tool automatically generates balanced and maximally diverse training sets to avoid bias. The resulting model correctly identifies 100% of out-of-focus and 98% of in-focus images in under 4 s per 96-well plate, and achieves this result even in heavily downsampled data (∼30 times smaller than native resolution). Integrating the tool into automated workflows minimizes the need for human verification as well as the collection and usage of low-quality data. FocA thus offers a solution to ensure reliable image data hygiene and improve the efficiency of automated imaging workflows using minimal computational resources.

The Application of High-Throughput Approaches in Identifying Novel Therapeutic Targets and Agents to Treat Diabetes

During the past decades, unprecedented progress in technologies has revolutionized traditional research methodologies. Among these, advances in high-throughput drug screening approaches have permitted the rapid identification of potential therapeutic agents from drug libraries that contain thousands or millions of molecules. Moreover, high-throughput-based therapeutic target discovery strategies can comprehensively interrogate relationships between biomolecules (e.g., gene, RNA, and protein) and diseases and significantly increase the authors' knowledge of disease mechanisms. Diabetes is a chronic disease primarily characterized by the incapacity of the body to maintain normoglycemia. The prevalence of diabetes in modern society has become a severe public health issue that threatens the well-being of millions of patients. Although a number of pharmacological treatments are available, there is no permanent cure for diabetes, and discovering novel therapeutic targets and agents continues to be an urgent need. The present review discusses the technical details of high-throughput screening approaches in drug discovery, followed by introducing the applications of such approaches to diabetes research. This review aims to provide an example of the applicability of high-throughput technologies in facilitating different aspects of disease research.

Characterization, Quantification, and Visualization of Neutrophil Extracellular Traps

Following the discovery of neutrophil extracellular traps (NETs) in 2004 by Brinkmann and colleagues, there has been extensive research into the role of NETs in a number of inflammatory diseases, including periodontitis. This chapter describes the current methods for the isolation of peripheral blood neutrophils as well as of oral neutrophils for subsequent NET experiments, including approaches to quantify and visualize NET production, the ability of NETs to entrap and kill bacteria, and the removal of NETs by nuclease-containing plasma.

Characterization of drug effects on cell cultures from phase-contrast microscopy images

In this work, we classify chemotherapeutic agents (topoisomerase inhibitors) based on their effect on U-2 OS cells. We use phase-contrast microscopy images, which are faster and easier to obtain than fluorescence images and support live cell imaging. We use a convolutional neural network (CNN) trained end-to-end directly on the input images without requiring for manual segmentations or any other auxiliary data. Our method can distinguish between tested cytotoxic drugs with an accuracy of 98%, provided that their mechanism of action differs, outperforming previous work. The results are even better when substance-specific concentrations are used. We show the benefit of sharing the extracted features over all classes (drugs). Finally, a 2D visualization of these features reveals clusters, which correspond well to known class labels, suggesting the possible use of our methodology for drug discovery application in analyzing new, unseen drugs.

In vitro biochemical assessment of mixture effects of two endocrine disruptors on INS-1 cells

4-tert-Octylphenol (4-tOP) is a component of non-ionic surfactants alkylphenol polyethoxylates while triclosan (TCS) is an antibacterial present in personal care products. Both compounds can co-exist in environmental matrices such as soil and water. The mixture effects of these micropollutants in vitro remains unknown. INS-1 cells were exposed to 20 µM or 30 µM 4-tOP and 8 µM or 12.5 µM TCS as well as equimolar mixture of the chemicals (Mix) in total concentration of 12.5 µM or 25 µM for 48 h. Mitochondrial related parameters were investigated using high content analytical techniques. The cytotoxicity of the chemicals (IC₅₀) varied according to TCS > Mix > 4-tOP. Increased glucose uptake and loss of mitochondrial membrane potential were recorded in TCS and Mix treated cells. Fold values of glucose-galactose assay varied according to dinitrophenol > TCS > 4-tOP > Mix in decreasing order of mitochondrial toxicity. The loss of the intracellular Ca²⁺ influx by all the test substances and Mix was not substantial whereas glibenclamide and diazoxide increased the intracellular Ca²⁺ influx when compared with the Blank. The recorded increase in Ca²⁺ influx by diazoxide which contrasted with its primary role of inhibiting insulin secretion need be re-investigated. It is concluded that the toxic effects of TCS and Mix but not 4-tOP on INS-1 cells was mitochondria-mediated.

HeartCV: a tool for transferrable, automated measurement of heart rate and heart rate variability in transparent animals

Heart function is a key component of whole-organismal physiology. Bioimaging is commonly, but not exclusively, used for quantifying heart function in transparent individuals, including early developmental stages of aquatic animals, many of which are transparent. However, a central limitation of many imaging-related methods is the lack of transferability between species, life-history stages and experimental approaches. Furthermore, locating the heart in mobile individuals remains challenging. Here, we present HeartCV: an open-source Python package for automated measurement of heart rate and heart rate variability that integrates automated localization and is transferrable across a wide range of species. We demonstrate the efficacy of HeartCV by comparing its outputs with measurements made manually for a number of very different species with contrasting heart morphologies. Lastly, we demonstrate the applicability of the software to different experimental approaches and to different dataset types, such as those corresponding to longitudinal studies.

Applications of Neural Networks in Biomedical Data Analysis

Neural networks for deep-learning applications, also called artificial neural networks, are important tools in science and industry. While their widespread use was limited because of inadequate hardware in the past, their popularity increased dramatically starting in the early 2000s when it became possible to train increasingly large and complex networks. Today, deep learning is widely used in biomedicine from image analysis to diagnostics. This also includes special topics, such as forensics. In this review, we discuss the latest networks and how they work, with a focus on the analysis of biomedical data, particularly biomarkers in bioimage data. We provide a summary on numerous technical aspects, such as activation functions and frameworks. We also present a data analysis of publications about neural networks to provide a quantitative insight into the use of network types and the number of journals per year to determine the usage in different scientific fields.

An Imaging and Computational Algorithm for Efficient Identification and Quantification of Neutrophil Extracellular Traps

Neutrophil extracellular traps (NETs) are associated with multiple disease pathologies including sepsis, asthma, rheumatoid arthritis, cancer, systemic lupus erythematosus, acute respiratory distress syndrome, and COVID-19. NETs, being a disintegrated death form, suffered inconsistency in their identification, nomenclature, and quantifications that hindered therapeutic approaches using NETs as a target. Multiple strategies including microscopy, ELISA, immunoblotting, flow cytometry, and image-stream-based methods have exhibited drawbacks such as being subjective, non-specific, error-prone, and not being high throughput, and thus demand the development of innovative and efficient approaches for their analyses. Here, we established an imaging and computational algorithm using high content screening (HCS)-cellomics platform that aid in easy, rapid, and specific detection as well as analyses of NETs. This method employed membrane-permeable and impermeable DNA dyes in situ to identify NET-forming cells. Automated algorithm-driven single-cell analysis of change in nuclear morphology, increase in nuclear area, and change in intensities provided precise detection of NET-forming cells and eliminated user bias with other cell death modalities. Further combination with Annexin V staining in situ detected specific death pathway, e.g., apoptosis, and thus, discriminated between NETs, apoptosis, and necrosis. Our approach does not utilize fixation and permeabilization steps that disturb NETs, and thus, allows the time-dependent monitoring of NETs. Together, this specific imaging-based high throughput method for NETs analyses may provide a good platform for the discovery of potential inhibitors of NET formation and/or agents to modulate neutrophil death, e.g., NETosis-apoptosis switch, as an alternative strategy to enhance the resolution of inflammation.

A quantitative analysis of cell bridging kinetics on a scaffold using computer vision algorithms

Tissue engineering involves the seeding of cells into a structural scaffolding to regenerate the architecture of damaged or diseased tissue. To effectively design a scaffold, an understanding of how cells collectively sense and react to the geometry of their local environment is needed. Advances in the development of melt electro-writing have allowed micron and submicron polymeric fibres to be accurately printed into porous, complex and three-dimensional structures. By using melt electrowriting, we created a geometrically relevant in vitro scaffold model to study cellular spatial-temporal kinetics. These scaffolds were paired with custom computer vision algorithms to investigate cell nuclei, cell membrane actin and scaffold fibres over different pore sizes (200-600 µm) and time points (28 days). We find that cells proliferated much faster in the smaller (200 µm) pores which halved the time until confluence versus larger (500 and 600 µm) pores. Our analysis of stained actin fibres revealed that cells were highly aligned to the fibres and the leading edge of the pore filling front, and we found that cells behind the leading edge were not aligned in any particular direction. This study provides a systematic understanding of cellular spatial temporal kinetics within a 3D in vitro model to inform the design of more effective synthetic tissue engineering scaffolds for tissue regeneration. STATEMENT OF SIGNIFICANCE: Advances in the development of melt electro-writing have allowed micron and submicron polymeric fibres to be accurately printed into porous, complex and three-dimensional structures. By using melt electrowriting, we created a geometrically relevant in vitro model to study cellular spatial-temporal kinetics to provide a systematic understanding of cellular spatial temporal kinetics within a 3D in vitro model. The insights presented in this work help to inform the design of more effective synthetic tissue engineering scaffolds by reducing cell culture time; which is valuable information for the implant or lab-grown-meat industries.

Machine Learning for Analysis of Microscopy Images: A Practical Guide

The explosive growth of machine learning has provided scientists with insights into data in ways unattainable using prior research techniques. It has allowed the detection of biological features that were previously unrecognized and overlooked. However, because machine-learning methodology originates from informatics, many cell biology labs have experienced difficulties in implementing this approach. In this article, we target the rapidly expanding audience of cell and molecular biologists interested in exploiting machine learning for analysis of their research. We discuss the advantages of employing machine learning with microscopy approaches and describe the machine-learning pipeline. We also give practical guidelines for building models of cell behavior using machine learning. We conclude with an overview of the tools required for model creation, and share advice on their use. © 2020 by John Wiley & Sons, Inc.

Development and Validation of High-Content Analysis for Screening HDAC6-Selective Inhibitors

Throughout recent decades, histone deacetylase (HDAC) inhibitors have shown encouraging potential in cancer treatment, and several pan-HDAC inhibitors have been approved for treating malignant cancers. Numerous adverse effects of pan-HDAC inhibitors have been reported, however, during preclinical and clinical evaluations. To avoid undesirable responses, an increasing number of investigations are focusing on the development of isotype-selective HDAC inhibitors. In this study, we present an effective and quantitative cellular assay using high-content analysis (HCA) to determine compounds' inhibition of the activity of HDAC6 and Class I HDAC isoforms, by detecting the acetylation of their corresponding substrates (i.e., α-tubulin and histone H3). Several conditions that are critical for HCA assays, such as cell seeding number, fixation and permeabilization reagent, and antibody dilution, have been fully validated in this study. We used selective HDAC6 inhibitors and inhibitors targeting different HDAC isoforms to optimize and validate the capability of the HCA assay. The results indicated that the HCA assay is a robust assay for quantifying compounds' selectivity of HDAC6 and Class I HDAC isoforms in cells. Moreover, we screened a panel of compounds for HDAC6 selectivity using this HCA assay, which provided valuable information for the structure-activity relationship (SAR). In summary, our results suggest that the HCA assay is a powerful tool for screening selective HDAC6 inhibitors.

High-Content Imaging to Phenotype Human Primary and iPSC-Derived Cells

Increasingly powerful microscopy, liquid handling, and computational techniques have enabled cell imaging in high throughput. Microscopy images are quantified using high-content analysis platforms linking object features to cell behavior. This can be attempted on physiologically relevant cell models, including stem cells and primary cells, in complex environments, and conceivably in the presence of perturbations. Recently, substantial focus has been devoted to cell profiling for cell therapy, assays for drug discovery or biomarker identification for clinical decision-making protocols, bringing this wealth of information into translational applications. In this chapter, we focus on two protocols enabling to (1) benchmark human cells, in particular human endothelial cells as a case study and (2) extract cells from blood for follow-up experiments including image-based drug testing. We also present concepts of high-content imaging and discuss the benefits and challenges, with the aim of enabling readers to tailor existing pipelines and bring such approaches closer to translational research and the clinic.

Classifying changes in LN-18 glial cell morphology: a supervised machine learning approach to analyzing cell microscopy data via FIJI and WEKA

In cell-based research, the process of visually monitoring cells generates large image datasets that need to be evaluated for quantifiable information in order to track the effectiveness of treatments in vitro. With the traditional, end-point assay-based approach being error-prone, and existing computational approaches being complex, we tested existing machine learning frameworks to find methods that are relatively simple, yet powerful enough to accomplish the goal of analyzing cell microscopy data. This paper details the machine learning pipeline for pixel-based classification and object-based classification. Furthermore, it compares the performances of three classifiers. The classifiers evaluated were the fast-random forest (RF), the sequential minimal optimization (SMO), and the Bayesian network (BN). Images were first preprocessed using smoothing and contrast methods found in FIJI. For pixel-based classification, the preprocessed images were fed into the Trainable Waikato Segmentation (TWS). For object-based classification, training and classification were conducted within the Waikato Environment for Knowledge Analysis (WEKA) interface. All classifiers' performance was evaluated using the WEKA experimental explorer. In terms of performance, the BN had the lowest classification accuracy for both the pixel-based and object-based model. The object-based SMO classifier had the best performance with the lowest mean absolute error of 0.05. The TWS and WEKA interface allows users to easily create and train classifiers for image analysis. However, for analyzing large image datasets, they are not ideal. Grapical abstract.

KCML: a machine-learning framework for inference of multi-scale gene functions from genetic perturbation screens

Characterising context-dependent gene functions is crucial for understanding the genetic bases of health and disease. To date, inference of gene functions from large-scale genetic perturbation screens is based on ad hoc analysis pipelines involving unsupervised clustering and functional enrichment. We present Knowledge- and Context-driven Machine Learning (KCML), a framework that systematically predicts multiple context-specific functions for a given gene based on the similarity of its perturbation phenotype to those with known function. As a proof of concept, we test KCML on three datasets describing phenotypes at the molecular, cellular and population levels and show that it outperforms traditional analysis pipelines. In particular, KCML identified an abnormal multicellular organisation phenotype associated with the depletion of olfactory receptors, and TGFβ and WNT signalling genes in colorectal cancer cells. We validate these predictions in colorectal cancer patients and show that olfactory receptors expression is predictive of worse patient outcomes. These results highlight KCML as a systematic framework for discovering novel scale-crossing and context-dependent gene functions. KCML is highly generalisable and applicable to various large-scale genetic perturbation screens.

Applications and Challenges of Machine Learning to Enable Realistic Cellular Simulations

In this perspective, we examine three key aspects of an end-to-end pipeline for realistic cellular simulations: reconstruction and segmentation of cellular structures; generation of cellular structures; and mesh generation, simulation, and data analysis. We highlight some of the relevant prior work in these distinct but overlapping areas, with a particular emphasis on current use of machine learning technologies, as well as on future opportunities.

Quality evaluation based on color grading: quality discrimination of the Chinese medicine Corni Fructus by an E-eye

‘Quality evaluation based on color grading’ is one of the features used in Chinese medicine discrimination. In order to assess the feasibility of electronic eye (E-eye) in implementing ‘quality evaluation based on color grading’, the present study applied an IRIS VA400 E-eye to test 58 batches of Corni Fructus samples. Their optical data were acquired and combined with their corresponding classes. A total of four quality discrimination models were produced according to discrimination analysis (DA), least squares support vector machine (LS-SVM), partial least squares-discrimination analysis (PLS-DA), and principal component analysis-discrimination analysis (PCA-DA). The accuracy rate of the aforementioned 4 cross evaluation models were 86.21%, 89.66%, 81.03% and 91.38%, respectively. Therefore, the PCA-DA method was used to build the final discrimination model for classifying Corni Fructus or discriminating its quality.

Interacting adipose-derived stem cells and microvascular endothelial cells provide a beneficial milieu for soft tissue healing

There is growing evidence suggesting that healing of chronic soft tissue wounds profits from the presence of adipose-derived stem cells (ADSC). Among the large spectrum of mechanisms by which ADSC might act, especially the interaction with the microvascular endothelial cell, a main player during angiogenesis, is of special interest. In the present 2D model on the basis of endothelial cell ADSC co-cultures, we focused on the identification of characteristics of both cell types in response to a typical condition in acute and chronic wounds: hypoxia. Parameters like proliferation capacity, migration, myofibroblastoid differentiation of ADSC and the quantification of important paracrine factors related to angiogenesis and inflammation were used to correlate our experimental model with the in vivo situation of soft tissue healing. ADSC were not negatively affected by hypoxia in terms of proliferation, referring to their excellent hypoxia tolerance. Myofibroblastoid differentiation among ADSC was enhanced by hypoxia in mono- but not in co-culture. Furthermore, co-cultures were able to migrate under hypoxia. These effects might be caused to some extent by the distinct milieu created by interacting ADSC and endothelial cells, which was characterized by modulated levels of interleukin-6, interleukin-8, monocyte chemoattractant protein-1 and vascular endothelial growth factor. The identification of these cell characteristics in the present 2D in vitro model provide new insights into the process of human soft tissue healing, and underpin a beneficial role of ADSC by regulating inflammation and angiogenesis.

Translational AI and Deep Learning in Diagnostic Pathology

There has been an exponential growth in the application of AI in health and in pathology. This is resulting in the innovation of deep learning technologies that are specifically aimed at cellular imaging and practical applications that could transform diagnostic pathology. This paper reviews the different approaches to deep learning in pathology, the public grand challenges that have driven this innovation and a range of emerging applications in pathology. The translation of AI into clinical practice will require applications to be embedded seamlessly within digital pathology workflows, driving an integrated approach to diagnostics and providing pathologists with new tools that accelerate workflow and improve diagnostic consistency and reduce errors. The clearance of digital pathology for primary diagnosis in the US by some manufacturers provides the platform on which to deliver practical AI. AI and computational pathology will continue to mature as researchers, clinicians, industry, regulatory organizations and patient advocacy groups work together to innovate and deliver new technologies to health care providers: technologies which are better, faster, cheaper, more precise, and safe.

A fully-automated, robust, and versatile algorithm for long-term budding yeast segmentation and tracking

Live cell time-lapse microscopy, a widely-used technique to study gene expression and protein dynamics in single cells, relies on segmentation and tracking of individual cells for data generation. The potential of the data that can be extracted from this technique is limited by the inability to accurately segment a large number of cells from such microscopy images and track them over long periods of time. Existing segmentation and tracking algorithms either require additional dyes or markers specific to segmentation or they are highly specific to one imaging condition and cell morphology and/or necessitate manual correction. Here we introduce a fully automated, fast and robust segmentation and tracking algorithm for budding yeast that overcomes these limitations. Full automatization is achieved through a novel automated seeding method, which first generates coarse seeds, then automatically fine-tunes cell boundaries using these seeds and automatically corrects segmentation mistakes. Our algorithm can accurately segment and track individual yeast cells without any specific dye or biomarker. Moreover, we show how existing channels devoted to a biological process of interest can be used to improve the segmentation. The algorithm is versatile in that it accurately segments not only cycling cells with smooth elliptical shapes, but also cells with arbitrary morphologies (e.g. sporulating and pheromone treated cells). In addition, the algorithm is independent of the specific imaging method (bright-field/phase) and objective used (40X/63X/100X). We validate our algorithm's performance on 9 cases each entailing a different imaging condition, objective magnification and/or cell morphology. Taken together, our algorithm presents a powerful segmentation and tracking tool that can be adapted to numerous budding yeast single-cell studies.

Automated quality control in image segmentation: application to the UK Biobank cardiovascular magnetic resonance imaging study

BACKGROUND

The trend towards large-scale studies including population imaging poses new challenges in terms of quality control (QC). This is a particular issue when automatic processing tools such as image segmentation methods are employed to derive quantitative measures or biomarkers for further analyses. Manual inspection and visual QC of each segmentation result is not feasible at large scale. However, it is important to be able to automatically detect when a segmentation method fails in order to avoid inclusion of wrong measurements into subsequent analyses which could otherwise lead to incorrect conclusions.

METHODS

To overcome this challenge, we explore an approach for predicting segmentation quality based on Reverse Classification Accuracy, which enables us to discriminate between successful and failed segmentations on a per-cases basis. We validate this approach on a new, large-scale manually-annotated set of 4800 cardiovascular magnetic resonance (CMR) scans. We then apply our method to a large cohort of 7250 CMR on which we have performed manual QC.

RESULTS

We report results used for predicting segmentation quality metrics including Dice Similarity Coefficient (DSC) and surface-distance measures. As initial validation, we present data for 400 scans demonstrating 99% accuracy for classifying low and high quality segmentations using the predicted DSC scores. As further validation we show high correlation between real and predicted scores and 95% classification accuracy on 4800 scans for which manual segmentations were available. We mimic real-world application of the method on 7250 CMR where we show good agreement between predicted quality metrics and manual visual QC scores.

CONCLUSIONS

We show that Reverse classification accuracy has the potential for accurate and fully automatic segmentation QC on a per-case basis in the context of large-scale population imaging as in the UK Biobank Imaging Study.

RefCell: multi-dimensional analysis of image-based high-throughput screens based on 'typical cells'

BACKGROUND

Image-based high-throughput screening (HTS) reveals a high level of heterogeneity in single cells and multiple cellular states may be observed within a single population. Currently available high-dimensional analysis methods are successful in characterizing cellular heterogeneity, but suffer from the "curse of dimensionality" and non-standardized outputs.

RESULTS

Here we introduce RefCell, a multi-dimensional analysis pipeline for image-based HTS that reproducibly captures cells with typical combinations of features in reference states and uses these "typical cells" as a reference for classification and weighting of metrics. RefCell quantitatively assesses heterogeneous deviations from typical behavior for each analyzed perturbation or sample.

CONCLUSIONS

We apply RefCell to the analysis of data from a high-throughput imaging screen of a library of 320 ubiquitin-targeted siRNAs selected to gain insights into the mechanisms of premature aging (progeria). RefCell yields results comparable to a more complex clustering-based single-cell analysis method; both methods reveal more potential hits than a conventional analysis based on averages.

A review of image analysis and machine learning techniques for automated cervical cancer screening from pap-smear images

BACKGROUND AND OBJECTIVE

Early diagnosis and classification of a cancer type can help facilitate the subsequent clinical management of the patient. Cervical cancer ranks as the fourth most prevalent cancer affecting women worldwide and its early detection provides the opportunity to help save life. To that end, automated diagnosis and classification of cervical cancer from pap-smear images has become a necessity as it enables accurate, reliable and timely analysis of the condition's progress. This paper presents an overview of the state of the art as articulated in prominent recent publications focusing on automated detection of cervical cancer from pap-smear images.

METHODS

The survey reviews publications on applications of image analysis and machine learning in automated diagnosis and classification of cervical cancer from pap-smear images spanning 15 years. The survey reviews 30 journal papers obtained electronically through four scientific databases (Google Scholar, Scopus, IEEE and Science Direct) searched using three sets of keywords: (1) segmentation, classification, cervical cancer; (2) medical imaging, machine learning, pap-smear; (3) automated system, classification, pap-smear.

RESULTS

Most of the existing algorithms facilitate an accuracy of nearly 93.78% on an open pap-smear data set, segmented using CHAMP digital image software. K-nearest-neighbors and support vector machines algorithms have been reported to be excellent classifiers for cervical images with accuracies of over 99.27% and 98.5% respectively when applied to a 2-class classification problem (normal or abnormal).

CONCLUSION

The reviewed papers indicate that there are still weaknesses in the available techniques that result in low accuracy of classification in some classes of cells. Moreover, most of the existing algorithms work either on single or on multiple cervical smear images. This accuracy can be increased by varying various parameters such as the features to be extracted, improvement in noise removal, using hybrid segmentation and classification techniques such of multi-level classifiers. Combining K-nearest-neighbors algorithm with other algorithm(s) such as support vector machines, pixel level classifications and including statistical shape models can also improve performance. Further, most of the developed classifiers are tested on accurately segmented images using commercially available software such as CHAMP software. There is thus a deficit of evidence that these algorithms will work in clinical settings found in developing countries (where 85% of cervical cancer incidences occur) that lack sufficient trained cytologists and the funds to buy the commercial segmentation software.

Establishment and Validation of an In Vitro Screening Method for Traditional Chinese Medicine-Induced Nephrotoxicity

Renal injury is among the adverse drug reactions (ADRs) caused by herbal medicine products (HMPs). Traditional Chinese medicines (TCMs) have been practiced for over 2000 years in China and East Asia, and herbs are currently used worldwide for the treatment and prevention of chronic and acute disease. Operetta high content analysis (HCA, PerkinElmer, Waltham, MA, USA), which is an in vitro, sensitive, reproducible, multiparametric screening method, was used to evaluate the cytotoxicity of HMPs in cultures of HEK293 human embryo kidney cells. Cytotoxic results were validated by an animal-based subacute toxicity assay. The renal safety of 18 active pharmaceutical agents from 13 TCM herbs with known nephrotoxic potential was evaluated in HEK293 human embryonic kidney cells. A panel of five parameters, cell viability, nuclear area, nuclear roundness, mitochondrial mass, and mitochondrial membrane potential, was utilized to evaluate drug-induced renal mitochondrial and nuclear injury. HCA can be a useful tool for preclinical screening and postclinical evaluation of HMPs. The nephrotoxicity of diosbulbin B and other HMPs was evident at a concentration as low as 0.01 μM.

Development and Application of High-Content Biological Screening for Modulators of NET Production

Neutrophil extracellular traps (NETs) are DNA-based antimicrobial web-like structures whose release is predominantly mediated by reactive oxygen species (ROS); their purpose is to combat infections. However, unbalanced NET production and clearance is involved in tissue injury, circulation of auto-antibodies and development of several chronic diseases. Currently, there is lack of agreement regarding the high-throughput methods available for NET investigation. This study, therefore, aimed to develop and optimize a high-content analysis (HCA) approach, which can be applied for the assay of NET production and for the screening of compounds involved in the modulation of NET release. A suitable paraformaldehyde fixation protocol was established to enable HCA of neutrophils and NETs. Bespoke and in-built bioinformatics algorithms were validated by comparison with standard low-throughput approaches for application in HCA of NETs. Subsequently, the optimized protocol was applied to high-content screening (HCS) of a pharmaceutically derived compound library to identify modulators of NETosis. Of 56 compounds assessed, 8 were identified from HCS for further characterization of their effects on NET formation as being either inducers, inhibitors or biphasic modulators. The effects of these compounds on naïve neutrophils were evaluated by using specific assays for the induction of ROS and NET production, while their modulatory activity was validated in phorbol 12-myristate 13-acetate-stimulated neutrophils. Results indicated the involvement of glutathione reductase, Src family kinases, molecular-target-of-Rapamycin, and mitogen-activated-protein-kinase pathways in NET release. The compounds and pathways identified may provide targets for novel therapeutic approaches for treating NET-associated pathologies.

Cell dynamic morphology analysis by deep convolutional features

Computational analysis of cell dynamic morphology in time-lapse image is a challenging task in biomedical research. Inspired by the recent success of deep learning, we investigate the possibility to apply a deep neural network to cell dynamic morphology analysis in this paper. Specifically, a contour spectrum is composed as the input of neural network to characterize cell spatiotemporal deformation, then a pre-trained convolutional neural network model is performed for automatic feature extraction. Finally, the extracted deep convolutional features are analyzed by SVM. Experimental results demonstrate that the proposed strategy outperforms existing methods on the live-cell database, and the features extracted by the last layer and classified by linear kernel SVM allows for the state-of-the-art performance.

Automated classification of cell morphology by coherence-controlled holographic microscopy

In the last few years, classification of cells by machine learning has become frequently used in biology. However, most of the approaches are based on morphometric (MO) features, which are not quantitative in terms of cell mass. This may result in poor classification accuracy. Here, we study the potential contribution of coherence-controlled holographic microscopy enabling quantitative phase imaging for the classification of cell morphologies. We compare our approach with the commonly used method based on MO features. We tested both classification approaches in an experiment with nutritionally deprived cancer tissue cells, while employing several supervised machine learning algorithms. Most of the classifiers provided higher performance when quantitative phase features were employed. Based on the results, it can be concluded that the quantitative phase features played an important role in improving the performance of the classification. The methodology could be valuable help in refining the monitoring of live cells in an automated fashion. We believe that coherence-controlled holographic microscopy, as a tool for quantitative phase imaging, offers all preconditions for the accurate automated analysis of live cell behavior while enabling noninvasive label-free imaging with sufficient contrast and high-spatiotemporal phase sensitivity.

Target identification by image analysis

Covering: 1997 to the end of 2015Each biologically active compound induces phenotypic changes in target cells that are characteristic for its mode of action. These phenotypic alterations can be directly observed under the microscope or made visible by labelling structural elements or selected proteins of the cells with dyes. A comparison of the cellular phenotype induced by a compound of interest with the phenotypes of reference compounds with known cellular targets allows predicting its mode of action. While this approach has been successfully applied to the characterization of natural products based on a visual inspection of images, recent studies used automated microscopy and analysis software to increase speed and to reduce subjective interpretation. In this review, we give a general outline of the workflow for manual and automated image analysis, and we highlight natural products whose bacterial and eucaryotic targets could be identified through such approaches.

Machine learning applications in cell image analysis

Machine learning (ML) refers to a set of automatic pattern recognition methods that have been successfully applied across various problem domains, including biomedical image analysis. This review focuses on ML applications for image analysis in light microscopy experiments with typical tasks of segmenting and tracking individual cells, and modelling of reconstructed lineage trees. After describing a typical image analysis pipeline and highlighting challenges of automatic analysis (for example, variability in cell morphology, tracking in presence of clutters) this review gives a brief historical outlook of ML, followed by basic concepts and definitions required for understanding examples. This article then presents several example applications at various image processing stages, including the use of supervised learning methods for improving cell segmentation, and the application of active learning for tracking. The review concludes with remarks on parameter setting and future directions.

Dictionary-enhanced imaging cytometry

State-of-the-art high-throughput microscopes are now capable of recording image data at a phenomenal rate, imaging entire microscope slides in minutes. In this paper we investigate how a large image set can be used to perform automated cell classification and denoising. To this end, we acquire an image library consisting of over one quarter-million white blood cell (WBC) nuclei together with CD15/CD16 protein expression for each cell. We show that the WBC nucleus images alone can be used to replicate CD expression-based gating, even in the presence of significant imaging noise. We also demonstrate that accurate estimates of white blood cell images can be recovered from extremely noisy images by comparing with a reference dictionary. This has implications for dose-limited imaging when samples belong to a highly restricted class such as a well-studied cell type. Furthermore, large image libraries may endow microscopes with capabilities beyond their hardware specifications in terms of sensitivity and resolution. We call for researchers to crowd source large image libraries of common cell lines to explore this possibility.

Image Quality Ranking Method for Microscopy

Automated analysis of microscope images is necessitated by the increased need for high-resolution follow up of events in time. Manually finding the right images to be analyzed, or eliminated from data analysis are common day-to-day problems in microscopy research today, and the constantly growing size of image datasets does not help the matter. We propose a simple method and a software tool for sorting images within a dataset, according to their relative quality. We demonstrate the applicability of our method in finding good quality images in a STED microscope sample preparation optimization image dataset. The results are validated by comparisons to subjective opinion scores, as well as five state-of-the-art blind image quality assessment methods. We also show how our method can be applied to eliminate useless out-of-focus images in a High-Content-Screening experiment. We further evaluate the ability of our image quality ranking method to detect out-of-focus images, by extensive simulations, and by comparing its performance against previously published, well-established microscopy autofocus metrics.

High-Content Microscopy Analysis of Subcellular Structures: Assay Development and Application to Focal Adhesion Quantification

High-content analysis (HCA) converts raw light microscopy images to quantitative data through the automated extraction, multiparametric analysis, and classification of the relevant information content. Combined with automated high-throughput image acquisition, HCA applied to the screening of chemicals or RNAi-reagents is termed high-content screening (HCS). Its power in quantifying cell phenotypes makes HCA applicable also to routine microscopy. However, developing effective HCA and bioinformatic analysis pipelines for acquisition of biologically meaningful data in HCS is challenging. Here, the step-by-step development of an HCA assay protocol and an HCS bioinformatics analysis pipeline are described. The protocol's power is demonstrated by application to focal adhesion (FA) detection, quantitative analysis of multiple FA features, and functional annotation of signaling pathways regulating FA size, using primary data of a published RNAi screen. The assay and the underlying strategy are aimed at researchers performing microscopy-based quantitative analysis of subcellular features, on a small scale or in large HCS experiments. © 2016 by John Wiley & Sons, Inc.

Application of high-content image analysis for quantitatively estimating lipid accumulation in oleaginous yeasts with potential for use in biodiesel production

Biodiesel from oleaginous microorganisms is a viable substitute for a fossil fuel. Current methods for microorganism lipid productivity evaluation do not analyze lipid dynamics in single cells. Here, we described a high-content image analysis (HCA) as a promising strategy for screening oleaginous microorganisms for biodiesel production, while generating single-cell lipid dynamics data in large cell density. Rhodotorula slooffiae yeast were grown in standard (CTL) or lipid trigger medium (LTM), and lipid droplet (LD) accumulation was analyzed in deconvolved confocal microscopy images of cells stained with the lipophilic fluorescent Nile red (NR) dye using automated cell and LD segmentation. The 'vesicle segmentation' method yielded valid morphometric results for limited lipid accumulation in smaller LDs (CTL samples) and for high lipid accumulation in larger LDs (LTM samples), and detected LD localization changes. Thus, HCA can be used to analyze the lipid accumulation patterns likely to be encountered in screens for biodiesel production.

Visualizing quantitative microscopy data: History and challenges

Data visualization is a fundamental aspect of science. In the context of microscopy-based studies, visualization typically involves presentation of the images themselves. However, data visualization is challenging when microscopy experiments entail imaging of millions of cells, and complex cellular phenotypes are quantified in a high-content manner. Most well-established visualization tools are inappropriate for displaying high-content data, which has driven the development of new visualization methodology. In this review, we discuss how data has been visualized in both classical and high-content microscopy studies; as well as the advantages, and disadvantages, of different visualization methods.

The use of high-throughput screening in identifying chemotherapeutic agents for gastric cancer

Gastric cancer claims many lives around the world, particularly in Asia. Although diagnosis and treatment has improved, long-term survival of patients is still poor and there is an urgent need to develop more effective treatments for this disease. This review outlines some of the more innovative high-throughput screening-based approaches and strategies that may be used to identify compounds that have new or novel mechanisms of action and could be developed further as possible gastric cancer treatments in the future.

Identification of a functional nuclear translocation sequence in hPPIP5K2

BACKGROUND

Cells contain several inositol pyrophosphates (PP-InsPs; also known as diphosphoinositol polyphosphates), which play pivotal roles in cellular and organismic homeostasis. It has been proposed that determining mechanisms of compartmentation of the synthesis of a particular PP-InsP is key to understanding how each of them may exert a specific function. Human PPIP5K2 (hPPIP5K2), one of the key enzymes that synthesizes PP-InsPs, contains a putative consensus sequence for a nuclear localization signal (NLS). However, such in silico analysis has limited predictive power, and may be complicated by phosphorylation events that can dynamically modulate NLS function. We investigated if this candidate NLS is functional and regulated, using the techniques of cell biology, mutagenesis and mass spectrometry.

RESULTS

Multiple sequence alignments revealed that the metazoan PPIP5K2 family contains a candidate NLS within a strikingly well-conserved 63 amino-acid domain. By analyzing the distribution of hPPIP5K2-GFP in HEK293T cells with the techniques of confocal microscopy and imaging flow cytometry, we found that a distinct pool of hPPIP5K2 is present in the nucleus. Imaging flow cytometry yielded particular insight into the characteristics of the nuclear hPPIP5K2 sub-pool, through a high-throughput, statistically-robust analysis of many hundreds of cells. Mutagenic disruption of the candidate NLS in hPPIP5K2 reduced its degree of nuclear localization. Proximal to the NLS is a Ser residue (S1006) that mass spectrometry data indicate is phosphorylated inside cells. The degree of nuclear localization of hPPIP5K2 was increased when S1006 was rendered non-phosphorylatable by its mutation to Ala. Conversely, a S1006D phosphomimetic mutant of hPPIP5K2 exhibited a lower degree of nuclear localization.

CONCLUSIONS

The current study describes for the first time the functional significance of an NLS in the conserved PPIP5K2 family. We have further demonstrated that there is phosphorylation of a Ser residue that is proximal to the NLS of hPPIP5K2. These conclusions draw attention to nuclear compartmentation of PPIP5K2 as being a physiologically relevant and covalently-regulated event. Our study also increases general insight into the consensus sequences of other NLSs, the functions of which might be similarly regulated.

High-content screening technology for studying drug-induced hepatotoxicity in cell models

High-content screening is the application of automated microscopy and image analysis to both cell biology and drug discovery. Over the last decade, this technique has emerged as a useful technology that allows the simultaneous measurement of different parameters at a single-cell level. Hepatotoxicity is a compelling reason for drug nonapprovals and withdrawals. It is recognized that the safety of a compound cannot be based on a single in vitro assay, and existing methods are not predictive of drug-induced toxicity. However, different HCS assays have been recently demonstrated as being powerful for identifying different mechanisms implicated in drug-induced toxicity with high sensitivity and specificity. These assays integrate the data obtained from different cell function indicators and can be easily incorporated into basic screening processes for the safety evaluation and selection of drug candidates; thus, they contribute greatly to lessen the likelihood of drug failure. Exploring the use of cellular imaging technology in drug-induced liver injury by reviewing the different tests proposed provides evidence that this technology has a strong impact on drug discovery.

[High content screening in chemical biology: overview and main challenges]

The last two decades have seen the development of high content screening (HCS) methodology and its adaptation for the evaluation of small molecules as drug candidates or their use as chemical tools for research purpose. HCS was initially set-up for the understanding of the mechanism of action of compounds by testing them on cell based-assays for pharmacological and toxicological studies. Since the last decade, the use of HCS has been extended to academic research laboratories and this technology has become the starting point for numerous projects aiming at the identification of molecular targets and cellular pathways for a given disease on which novel type of drugs could act. This screening approach relies on image capture of fluorescently labeled cells therefore generating a large amount of data that must be handled by appropriate automated image analysis methods and storage instrumentation. These latter in addition to the integration and data sharing are current challenges that HCS must still tackle.

Image-based systems biology of infection

The successful treatment of infectious diseases requires interdisciplinary studies of all aspects of infection processes. The overarching combination of experimental research and theoretical analysis in a systems biology approach can unravel mechanisms of complex interactions between pathogens and the human immune system. Taking into account spatial information is especially important in the context of infection, since the migratory behavior and spatial interactions of cells are often decisive for the outcome of the immune response. Spatial information is provided by image and video data that are acquired in microscopy experiments and that are at the heart of an image-based systems biology approach. This review demonstrates how image-based systems biology improves our understanding of infection processes. We discuss the three main steps of this approach--imaging, quantitative characterization, and modeling--and consider the application of these steps in the context of studying infection processes. After summarizing the most relevant microscopy and image analysis approaches, we discuss ways to quantify infection processes, and address a number of modeling techniques that exploit image-derived data to simulate host-pathogen interactions in silico.

Visualizing cellular imaging data using PhenoPlot

Visualization is essential for data interpretation, hypothesis formulation and communication of results. However, there is a paucity of visualization methods for image-derived data sets generated by high-content analysis in which complex cellular phenotypes are described as high-dimensional vectors of features. Here we present a visualization tool, PhenoPlot, which represents quantitative high-content imaging data as easily interpretable glyphs, and we illustrate how PhenoPlot can be used to improve the exploration and interpretation of complex breast cancer cell phenotypes.

Image-based cell-resolved screening assays in flow

A parallel microfluidic cytometer (PMC) is based on a one-dimensional (1D) scanning detector, a parallel array of flow channels, and new multiparameter analysis algorithms that operate on low-pixel-count 1D images. In this article, we explore a series of image-based live- and fixed-cell screening assays, including two NF-kB nuclear translocations and T-cell capping. We then develop a new multiparametric linear weighted classifier that achieves a Z' factor sufficient for scaled pharmaceutical discovery with Jurkat cells in suspension. We conclude that the PMC should have the throughput and statistical power to permit a new capability for image-based high-sample-number pharmaceutical screening with suspension samples.