[Short paths to diagnosis with artificial intelligence: systematic literature review on diagnostic decision support systems]

BACKGROUND

Rare diseases are often recognized late. Their diagnosis is particularly challenging due to the diversity, complexity and heterogeneity of clinical symptoms. Computer-aided diagnostic aids, often referred to as diagnostic decision support systems (DDSS), are promising tools for shortening the time to diagnosis. Despite initial positive evaluations, DDSS are not yet widely used, partly due to a lack of integration with existing clinical or practice information systems.

OBJECTIVE

This article provides an insight into currently existing diagnostic support systems that function without access to electronic patient records and only require information that is easily obtainable.

MATERIALS AND METHODS

A systematic literature search identified eight articles on DDSS that can assist in the diagnosis of rare diseases with no need for access to electronic patient records or other information systems in practices and hospitals. The main advantages and disadvantages of the identified rare disease diagnostic support systems were extracted and summarized.

RESULTS

Symptom checkers and DDSS based on portrait photos and pain drawings already exist. The degree of maturity of these applications varies.

CONCLUSION

DDSS currently still face a number of challenges, such as concerns about data protection and accuracy, and acceptance and awareness continue to be rather low. On the other hand, there is great potential for faster diagnosis, especially for rare diseases, which are easily overlooked due to their large number and the low awareness of them. The use of DDSS should therefore be carefully considered by doctors on a case-by-case basis.

Deep learning algorithm for fully automated measurement of sagittal balance in adult spinal deformity

AIM

Deep learning (DL) algorithms can be used for automated analysis of medical imaging. The aim of this study was to assess the accuracy of an innovative, fully automated DL algorithm for analysis of sagittal balance in adult spinal deformity (ASD).

MATERIAL AND METHODS

Sagittal balance (sacral slope, pelvic tilt, pelvic incidence, lumbar lordosis and sagittal vertical axis) was evaluated in 141 preoperative and postoperative radiographs of patients with ASD. The DL, landmark-based measurements, were compared with the ground truth values from validated manual measurements.

RESULTS

The DL algorithm showed an excellent consistency with the ground truth measurements. The intra-class correlation coefficient between the DL and ground truth measurements was 0.71-0.99 for preoperative and 0.72-0.96 for postoperative measurements. The DL detection rate was 91.5% and 84% for preoperative and postoperative images, respectively.

CONCLUSION

This is the first study evaluating a complete automated DL algorithm for analysis of sagittal balance with high accuracy for all evaluated parameters. The excellent accuracy in the challenging pathology of ASD with long construct instrumentation demonstrates the eligibility and possibility for implementation in clinical routine.

Automated high-content image-based characterization of microorganism behavioral diversity and distribution

Microorganisms have evolved complex systems to respond to environmental signals. Gradients of particular molecules and elemental ions alter the behavior of microbes and their distribution within their environment. Microdevices coupled with automated image-based methods are now employed to analyze the instantaneous distribution and motion behaviors of microbial species in controlled environments at small temporal scales, mimicking, to some extent, macro conditions. Such technologies have so far been adopted for investigations mainly on individual species. Similar versatile approaches must now be developed for the characterization of multiple and complex interactions between a microbial community and its environment. Here, we provide a comprehensive step-by-step method for the characterization of species-specific behavior in a synthetic mixed microbial suspension in response to an environmental driver. By coupling accessible microfluidic devices with automated image analysis approaches, we evaluated the behavioral response of three morphologically different telluric species (Phytophthora parasitica, Vorticella microstoma, Enterobacter aerogenes) to a potassium gradient driver. Using the TrackMate plug-in algorithm, we performed morphometric and then motion analyses to characterize the response of each microbial species to the driver. Such an approach enabled to confirm the different morphological features of the three species and simultaneously characterize their specific motion in reaction to the driver and their co-interaction dynamics. By increasing the complexity of suspensions, this approach could be integrated in a framework for phenotypic analysis in microbial ecology research, helping to characterize how key drivers influence microbiota assembly at microbiota host-environment interfaces.

Invention of automated numerical algorithm adopting binarization for the evaluation of scalp hair coverage: An image analysis providing a substitute for phototrichogram and global photography assessment for hair diseases

BACKGROUND

The efficacy of therapeutic modalities for hair disease can be evaluated globally by photo assessment and more precisely by phototrichogram (PTG). However, the latter procedure is laborious, time consuming, subject to inter-observer variation, and requires hair clipping.

OBJECTIVE

To establish an automated and patient/investigator friendly methodology enabling quantitative hair amount evaluation for daily clinical practice.

METHODS

A novel automated numerical algorithm (aNA) adopting digital image binarization (i.e., black and white color conversion) was invented to evaluate hair coverage and measure PTG parameters in scalp images. Step-by-step improvement of aNA was attempted through comparative analyses of the data obtained respectively by the novel approach and conventional PTG/global photography assessment (GPA).

RESULTS

For measuring scalp hair coverage, the initial version of aNA generally agreed with the cumulative hair diameter as assessed using PTG, showing a coefficient of 0.60. However, these outcomes were influenced by the angle of hair near the parting line. By integrating an angle compensation formula, the standard deviation of aNA data decreased from 5.7% to 1.2%. Consequently, the coefficient of determination for hair coverage calculated using the modified aNA and cumulative hair diameter assessed by PTG increased to 0.90. Furthermore, the change in hair coverage as determined by the modified aNA protocol correlated well with changes in the GPA score of images obtained using clinical trials.

CONCLUSION

The novel aNA method provides a valuable tool for enabling simple and accurate evaluation of hair growth and volume for clinical trials and for treatment of hair disease.

Automated stereological image analysis approach of the human placenta: Surface areas and vascularization

Detecting and quantifying surface densities of placental villi and their vasculature adds important information on the development of the placenta under different exposures and pathological conditions. Today, a larger number of samples and tissue areas can be examined using automated Artificial Intelligence-based approaches. Although each image series calls for a particular approach, sharing the methods will help in facilitating reproducibility and comparability. Here we show the protocol of a software-based quantification of vessels (number and area) in villous tissues of human placentas, based on scanned images of full-size placental sections.

Deep machine learning for cell segmentation and quantitative analysis of radial plant growth

Plants produce the major part of terrestrial biomass and are long-term deposits of atmospheric carbon. This capacity is to a large extent due to radial growth of woody species - a process driven by cambium stem cells located in distinct niches of shoot and root axes. In the model species Arabidopsis thaliana, thousands of cells are produced by the cambium in radial orientation generating a complex organ anatomy enabling long-distance transport, mechanical support and protection against biotic and abiotic stressors. These complex organ dynamics make a comprehensive and unbiased analysis of radial growth challenging and asks for tools for automated quantification. Here, we combined the recently developed PlantSeg and MorphographX image analysis tools, to characterize tissue morphogenesis of the Arabidopsis hypocotyl. After sequential training of segmentation models on ovules, shoot apical meristems and adult hypocotyls using deep machine learning, followed by the training of cell type classification models, our pipeline segments complex images of transverse hypocotyl sections with high accuracy and classifies central hypocotyl cell types. By applying our pipeline on both wild type and phloem intercalated with xylem (pxy) mutants, we also show that this strategy faithfully detects major anatomical aberrations. Collectively, we conclude that our established pipeline is a powerful phenotyping tool comprehensively extracting cellular parameters and providing access to tissue topology during radial plant growth.

Novel image analysis tool for rapid screening of cell morphology in preclinical animal models of disease

The field of cell biology has seen major advances in both cellular imaging modalities and the development of automated image analysis platforms that increase rigor, reproducibility, and throughput for large imaging data sets. However, there remains a need for tools that provide accurate morphometric analysis of single cells with complex, dynamic cytoarchitecture in a high-throughput and unbiased manner. We developed a fully automated image-analysis algorithm to rapidly detect and quantify changes in cellular morphology using microglia cells, an innate immune cell within the central nervous system, as representative of cells that exhibit dynamic and complex cytoarchitectural changes. We used two preclinical animal models that exhibit robust changes in microglia morphology: (1) a rat model of acute organophosphate intoxication, which was used to generate fluorescently labeled images for algorithm development; and (2) a rat model of traumatic brain injury, which was used to validate the algorithm using cells labeled using chromogenic detection methods. All ex vivo brain sections were immunolabeled for IBA-1 using fluorescence or diaminobenzidine (DAB) labeling, images were acquired using a high content imaging system and analyzed using a custom-built algorithm. The exploratory data set revealed eight statistically significant and quantitative morphometric parameters that distinguished between phenotypically distinct groups of microglia. Manual validation of single-cell morphology was strongly correlated with the automated analysis and was further supported by a comparison with traditional stereology methods. Existing image analysis pipelines rely on high-resolution images of individual cells, which limits sample size and is subject to selection bias. However, our fully automated method integrates quantification of morphology and fluorescent/chromogenic signals in images from multiple brain regions acquired using high-content imaging. In summary, our free, customizable image analysis tool provides a high-throughput, unbiased method for accurately detecting and quantifying morphological changes in cells with complex morphologies.

Mass Balance Imaging: A Phase Portrait Analysis for Characterizing Growth Kinetics of Biomolecular Condensates

Biomolecular condensation has emerged as a key organizing principle governing the formation of membraneless cellular assemblies. Revealing the mechanism of formation of biomolecular condensates requires the quantitative examination of their growth kinetics. Here, we introduce mass balance imaging (MBI) as a general method to study compositional growth dynamics based on fluorescent images of multicomponent clusters. MBI allows the visualization and measurement of composition-dependent growth rates of biomolecular condensates and other assemblies. We provide a computational pipeline and demonstrate the applicability of our method by investigating cortical assemblies containing N-WASP (WSP-1) and F-actin that appear during oocyte cortex activation in C. elegans. In general, the method can be broadly implemented to identify interactions that underlie growth kinetics of multicomponent assemblies in vivo and in vitro.

Detecting variation in starch granule size and morphology by high-throughput microscopy and flow cytometry

Starch forms semi-crystalline, water-insoluble granules, the size and morphology of which vary according to biological origin. These traits, together with polymer composition and structure, determine the physicochemical properties of starch. However, screening methods to identify differences in starch granule size and shape are lacking. Here, we present two approaches for high-throughput starch granule extraction and size determination using flow cytometry and automated, high-throughput light microscopy. We evaluated the practicality of both methods using starch from different species and tissues and demonstrated their effectiveness by screening for induced variation in starch extracted from over 10,000 barley lines, yielding four with heritable changes in the ratio of large A-granules to small B-granules. Analysis of Arabidopsis lines altered in starch biosynthesis further demonstrates the applicability of these approaches. Identifying variation in starch granule size and shape will enable identification of trait-controlling genes for developing crops with desired properties, and could help optimise starch processing.

A Simple Ca2+-Imaging Approach of Network-Activity Analyses for Human Neurons

Rapid advances in light microscopy and development of all-optical electrophysiological imaging tools have greatly leveraged the speed and the depth of neurobiology studies. Calcium imaging is a common method that is useful for measuring calcium signals in cells and has been used as a functional proxy for neuronal activity. Here I describe a simple, stimulation-free approach that measures neuronal network activity and single-neuron dynamics in human neurons. This protocol provides the experimental workflow that includes step-wise illustrations of sample preparations, data processing, and analyses that can be used for quick phenotypical assessment and serves as a quick functional readout for mutagenesis or screen effort for neurodegenerative studies.

Artificial intelligence for image analysis in total hip and total knee arthroplasty : a scoping review

AIMS

Total hip arthroplasty (THA) and total knee arthroplasty (TKA) are common orthopaedic procedures requiring postoperative radiographs to confirm implant positioning and identify complications. Artificial intelligence (AI)-based image analysis has the potential to automate this postoperative surveillance. The aim of this study was to prepare a scoping review to investigate how AI is being used in the analysis of radiographs following THA and TKA, and how accurate these tools are.

METHODS

The Embase, MEDLINE, and PubMed libraries were systematically searched to identify relevant articles. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for scoping reviews and Arksey and O'Malley framework were followed. Study quality was assessed using a modified Methodological Index for Non-Randomized Studies tool. AI performance was reported using either the area under the curve (AUC) or accuracy.

RESULTS

Of the 455 studies identified, only 12 were suitable for inclusion. Nine reported implant identification and three described predicting risk of implant failure. Of the 12, three studies compared AI performance with orthopaedic surgeons. AI-based implant identification achieved AUC 0.992 to 1, and most algorithms reported an accuracy > 90%, using 550 to 320,000 training radiographs. AI prediction of dislocation risk post-THA, determined after five-year follow-up, was satisfactory (AUC 76.67; 8,500 training radiographs). Diagnosis of hip implant loosening was good (accuracy 88.3%; 420 training radiographs) and measurement of postoperative acetabular angles was comparable to humans (mean absolute difference 1.35° to 1.39°). However, 11 of the 12 studies had several methodological limitations introducing a high risk of bias. None of the studies were externally validated.

CONCLUSION

These studies show that AI is promising. While it already has the ability to analyze images with significant precision, there is currently insufficient high-level evidence to support its widespread clinical use. Further research to design robust studies that follow standard reporting guidelines should be encouraged to develop AI models that could be easily translated into real-world conditions. Cite this article: Bone Joint J 2022;104-B(8):929-937.

Artificial intelligence-based diagnostics of molar-incisor-hypomineralization (MIH) on intraoral photographs

Objective

The aim of this study was to develop and validate a deep learning–based convolutional neural network (CNN) for the automated detection and categorization of teeth affected by molar-incisor-hypomineralization (MIH) on intraoral photographs.

Materials and methods

The data set consisted of 3241 intraoral images (767 teeth with no MIH/no intervention, 76 with no MIH/atypical restoration, 742 with no MIH/sealant, 815 with demarcated opacity/no intervention, 158 with demarcated opacity/atypical restoration, 181 with demarcated opacity/sealant, 290 with enamel breakdown/no intervention, 169 with enamel breakdown/atypical restoration, and 43 with enamel breakdown/sealant). These images were divided into a training (N = 2596) and a test sample (N = 649). All images were evaluated by an expert group, and each diagnosis served as a reference standard for cyclic training and evaluation of the CNN (ResNeXt-101–32 × 8d). Statistical analysis included the calculation of contingency tables, areas under the receiver operating characteristic curve (AUCs) and saliency maps.

Results

The developed CNN was able to categorize teeth with MIH correctly with an overall diagnostic accuracy of 95.2%. The overall SE and SP amounted to 78.6% and 97.3%, respectively, which indicate that the CNN performed better in healthy teeth compared to those with MIH. The AUC values ranging from 0.873 (enamel breakdown/sealant) to 0.994 (atypical restoration/no MIH).

Conclusion

It was possible to categorize the majority of clinical photographs automatically by using a trained deep learning–based CNN with an acceptably high diagnostic accuracy.

Clinical relevance

Artificial intelligence-based dental diagnostics may support dental diagnostics in the future regardless of the need to improve accuracy.

Performances of Machine Learning in Detecting Glaucoma Using Fundus and Retinal Optical Coherence Tomography Images: A Meta-Analysis

PURPOSE

To evaluate the performance of machine learning (ML) in detecting glaucoma using fundus and retinal optical coherence tomography (OCT) images.

DESIGN

Meta-analysis.

METHODS

PubMed and EMBASE were searched on August 11, 2021. A bivariate random-effects model was used to pool ML's diagnostic sensitivity, specificity, and area under the curve (AUC). Subgroup analyses were performed based on ML classifier categories and dataset types.

RESULTS

One hundred and five studies (3.3%) were retrieved. Seventy-three (69.5%), 30 (28.6%), and 2 (1.9%) studies tested ML using fundus, OCT, and both image types, respectively. Total testing data numbers were 197,174 for fundus and 16,039 for OCT. Overall, ML showed excellent performances for both fundus (pooled sensitivity = 0.92 [95% CI, 0.91-0.93]; specificity = 0.93 [95% CI, 0.91-0.94]; and AUC = 0.97 [95% CI, 0.95-0.98]) and OCT (pooled sensitivity = 0.90 [95% CI, 0.86-0.92]; specificity = 0.91 [95% CI, 0.89-0.92]; and AUC = 0.96 [95% CI, 0.93-0.97]). ML performed similarly using all data and external data for fundus and the external test result of OCT was less robust (AUC = 0.87). When comparing different classifier categories, although support vector machine showed the highest performance (pooled sensitivity, specificity, and AUC ranges, 0.92-0.96, 0.95-0.97, and 0.96-0.99, respectively), results by neural network and others were still good (pooled sensitivity, specificity, and AUC ranges, 0.88-0.93, 0.90-0.93, 0.95-0.97, respectively). When analyzed based on dataset types, ML demonstrated consistent performances on clinical datasets (fundus AUC = 0.98 [95% CI, 0.97-0.99] and OCT AUC = 0.95 [95% 0.93-0.97]).

CONCLUSIONS

Performance of ML in detecting glaucoma compares favorably to that of experts and is promising for clinical application. Future prospective studies are needed to better evaluate its real-world utility.

Blood Vessel Detection Algorithm for Tissue Engineering and Quantitative Histology

Immunohistochemistry for vascular network analysis plays a fundamental role in basic science, translational research and clinical practice. However, identifying vascularization in histological tissue images is time consuming and markedly depends on the operator’s experience. In this study, we present “blood vessel detection—BVD”, an automatic algorithm for quantitative analysis of blood vessels in immunohistochemical images. BVD is based on extraction and analysis of low-level image features and spatial filtering techniques, which do not require a training phase. BVD algorithm performance was comparatively evaluated on histological sections from three different in vivo experiments. Collectively, 173 independent images were analyzed, and the algorithm's results were compared to those obtained by human operators. The developed BVD algorithm proved to be a robust and versatile tool, being able to quantify number, area, and spatial distribution of blood vessels within all three considered histologic datasets. BVD is provided as an open-source application working on different operating systems. BVD is supported by a user-friendly graphical interface designed to facilitate large-scale analysis.

Fractal dimension of the aortic annulus: a novel predictor of paravalvular leak after transcatheter aortic valve implantation

To evaluate the prognostic relevance of aortic annulus (AA) and left ventricular outflow tract (LVOT) Fractal dimension (FD). FD is a mathematical concept that describes geometric complexity of a structure and has been shown to predict adverse outcomes in several contexts. Computed tomography (CT) scans from the SOLVE-TAVI trial, which, in a 2 × 2 factorial design, randomized 447 patients to TAVI with the balloon-expandable Edwards Sapien 3 or the self-expanding Medtronic Evolut R, and conscious sedation or general anesthesia, were analyzed semi-automatically with a custom-built software to determine border of AA and LVOT. FD was measured by box counting using grid calibers between 0.8 and 6.75 mm and was compared between patients with none/trivial and mild/moderate paravalvular regurgitation (PVR). Overall, 122 patients had CT scans sufficient for semi-automatic PVR in 30-day echocardiography. PVR was none in 65(53.3%) patients, trace in 9(7.4%), mild in 46(37.7%), moderate in 2(1.6%) and severe in 0 patients. FD determined in diastolic images was significantly higher in patients with mild/moderate PVR (1.0558 ± 0.0289 vs. 1.0401 ± 0.0284, p = 0.017). Annulus eccentricity was the only conventional measure of AA and LVOT geometry significantly correlated to FD (R = 0.337, p < 0.01). Area under the curve (AUC) of diastolic annular FD for prediction of mild/moderate PVR in ROC analysis was 0.661 (0.542-0.779, p = 0.014). FD shows promise in prediction of PVR after TAVI. Further evaluation using larger patient numbers and refined algorithms to better understand its predictive performance is warranted.Trial Registration: www.clinicaltrials.gov , identifier: NCT02737150, date of registration: 13.04.2016.

Image-Based Screening for Stress Granule Regulators

Stress granule (SG)-based RNA interference (RNAi) screening is a powerful method to discover factors that control protein synthesis and aggregation, as well as regulators of SG assembly and disassembly. Here, we describe how to set up and optimize a large-scale siRNA screen, and give a detailed outline for the automated quantification of SGs as a visual readout. Hit evaluation via calculated Z scores provides a list of candidates for further in-depth studies.

ExplantAnalyzer: An advanced automated neurite outgrowth analysis evaluated by means of organotypic auditory neuron explant cultures

BACKGROUND

Neuronal outgrowth assays using organotypic explant cultures are commonly utilized to study neuroregenerative and -protective effects of drugs such as neurotrophins. While this approach offers higher organized tissue compared to single cell cultures and less experimental effort than in-vivo studies, quantitative evaluation of the neuronal network is often time consuming. Thus, we developed ExplantAnlayzer, a time-saving high-throughput evaluation method, yielding numerous metrics to objectively describe neuronal outgrowth.

NEW METHOD

Spiral ganglion explants were cultured in 24-well plates, mechanically fixed in a collagen matrix and immunolabeled against beta-III-tubulin. The explants were imaged using a fluorescent tile-scan microscope and resulting images were stitched. The evaluation was developed as an open-source MATLAB routine and involves several image processing steps, including adaptive thresholding. The neurite network was eventually converted to a graph to track neurites from their terminals back to the explant body.

COMPARISON WITH EXISTING METHOD(S)

We compared ExplantAnlayzer quantitatively and qualitatively to common existing methods, such as Sholl analyses and manual fiber tracing, using representative explant images. ExplantAnlayzer is able to achieve similar and as detailed results as manual tracing while decreasing manual interaction and required time dramatically.

RESULTS

After an initial setup phase, the explant images could be batch-processed altogether. Bright bundles as well as faint fibers were reliably detected. Several metrics describing the outgrowth morphology, including total outgrowth, neurite numbers and length estimations, as well as their growth directions, were computed.

CONCLUSIONS

ExplantAnalyzer is a time-saving and objective method for an in-depth evaluation of organotypic explant outgrowth.

eHooke: A tool for automated image analysis of spherical bacteria based on cell cycle progression

Fluorescence microscopy is a critical tool for cell biology studies on bacterial cell division and morphogenesis. Because the analysis of fluorescence microscopy images evolved beyond initial qualitative studies, numerous images analysis tools were developed to extract quantitative parameters on cell morphology and organization. To understand cellular processes required for bacterial growth and division, it is particularly important to perform such analysis in the context of cell cycle progression. However, manual assignment of cell cycle stages is laborious and prone to user bias. Although cell elongation can be used as a proxy for cell cycle progression in rod-shaped or ovoid bacteria, that is not the case for cocci, such as Staphylococcus aureus. Here, we describe eHooke, an image analysis framework developed specifically for automated analysis of microscopy images of spherical bacterial cells. eHooke contains a trained artificial neural network to automatically classify the cell cycle phase of individual S. aureus cells. Users can then apply various functions to obtain biologically relevant information on morphological features of individual cells and cellular localization of proteins, in the context of the cell cycle.

A novel automated image analysis pipeline for quantifying morphological changes to the endoplasmic reticulum in cultured human cells

Background

In mammalian cells the endoplasmic reticulum (ER) comprises a highly complex reticular morphology that is spread throughout the cytoplasm. This organelle is of particular interest to biologists, as its dysfunction is associated with numerous diseases, which often manifest themselves as changes to the structure and organisation of the reticular network. Due to its complex morphology, image analysis methods to quantitatively describe this organelle, and importantly any changes to it, are lacking.

Results

In this work we detail a methodological approach that utilises automated high-content screening microscopy to capture images of cells fluorescently-labelled for various ER markers, followed by their quantitative analysis. We propose that two key metrics, namely the area of dense ER and the area of polygonal regions in between the reticular elements, together provide a basis for measuring the quantities of rough and smooth ER, respectively. We demonstrate that a number of different pharmacological perturbations to the ER can be quantitatively measured and compared in our automated image analysis pipeline. Furthermore, we show that this method can be implemented in both commercial and open-access image analysis software with comparable results.

Conclusions

We propose that this method has the potential to be applied in the context of large-scale genetic and chemical perturbations to assess the organisation of the ER in adherent cell cultures.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04334-x.

Whole Slide Imaging (WSI) in Pathology: Current Perspectives and Future Directions

Whole slide imaging (WSI), ever since its first introduction about two decades ago, has been validated for a number of applications in the field of pathology. The recent approval of US FDA to a WSI system for use in primary surgical pathology diagnosis has opened avenues for wider acceptance and application of this technology in routine practice. The ongoing technological advances in digital scanners, image visualization methods, and the integration of artificial intelligence-derived algorithms with these systems provide opportunities of its newer applications. Its benefits are innumerable such as ease of access through internet, avoidance of physical storage space, and no risk of deterioration of staining quality or breakage of slides to name a few. Various barriers such as the high cost, technical glitches, and professional hesitation to adopt a new technology have hindered its use in pathology. This review article summarizes the technical aspects of WSI, its applications in diagnostic pathology, training, and research along with future perspectives. It highlights the benefits, limitations, and challenges delaying the use of this technology in routine practice. The review is targeted at students, residents, and budding pathologists to better acquaint them with the key aspects of state-of-the-art technology and enable them to implement WSI judiciously.

eG Coated Stents Exhibit Enhanced Endothelial Wound Healing Characteristics

Purpose

Despite their widespread use, a significant fraction of coronary stents suffer from in-stent restenosis and stent thrombosis. Stent deployment induces extensive injury to the vascular endothelium. Rapid endothelial wound closure is essential for the success of a stenting procedure. A recent study has demonstrated that the BuMA Supreme® sirolimus-eluting stent exhibits particularly attractive strut coverage characteristics. A unique feature of this stent is the presence of a thin brush layer of poly-butyl methacrylate (PBMA), covalently bonded to the stent’s cobalt-chromium frame via electro-grafting (eG™). The present study aimed to determine whether the PBMA coating has an effect on endothelial cell wound healing and stent strut coverage.

Methods

We used an in vitro coronary artery model whose wall consisted of an annular collagen hydrogel and whose luminal surface was lined with a monolayer of endothelial cells. Mechanical wounding of the endothelial lining was preformed prior to deployment of a bare cobalt-chromium stent either with or without the PBMA layer. The migration of fluorescently labeled endothelial cells was monitored automatically over a period of 48 h to determine endothelial wound healing rates.

Results

Quantitative assessment of endothelial wound healing rates within the simulated arterial model is achievable using automated image analysis. Wound healing is significantly faster (44% faster at 48 h) for stents with the PBMA eG Coating™ compared to bare metal stents.

Conclusion

The PBMA eG Coating™ has the effect of promoting endothelial wound healing. Future studies will focus on elucidating the mechanistic basis of this observation.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13239-021-00542-x.

Automated Quantitative Analysis of Airway Epithelial Cell Detachment Upon Fungal Challenge

Host-pathogen interactions involve a complex interplay between host and pathogen factors, resulting in either host protective immunity or establishment of disease. One of the hallmarks for disease progression is host tissue destruction. The first host surface to interact with the opportunistic respiratory fungal pathogen, Aspergillus fumigatus, is the airway epithelium. Unravelling the mechanisms involved in airway epithelial cell damage by A. fumigatus is essential to understanding the establishment and progression of infection in the host. Although host cell damage can be measured in vitro by indirect cell lysis assays, here, we describe an automated, simple, and low-cost assay to directly visualize and quantify epithelial cell line damage after challenge with A. fumigatus. We employ the previously characterized tissue noninvasive A. fumigatus ΔpacC mutant to demonstrate the quantitative difference in cell damage relative to its parental tissue invasive strain. This assay is easily scaled up for high-throughput screening of multiple Aspergillus mutants and can be adapted to suit diverse host cell lines, different time points of infection, challenge with other microbes, and drugs or novel compounds.

A simple Ca2+-imaging approach to neural network analyses in cultured neurons

BACKGROUND

Ca²⁺-imaging is a powerful tool to measure neuronal dynamics and network activity. To monitor network-level changes in cultured neurons, neuronal activity is often evoked by electrical or optogenetic stimulation and assessed using multi-electrode arrays or sophisticated imaging. Although such approaches allow detailed network analyses, multi-electrode arrays lack single-cell precision, whereas optical physiology generally requires advanced instrumentation that may not be universally available.

NEW METHOD

Here we developed a simple, stimulation-free protocol with associated Matlab algorithms that enables scalable analyses of spontaneous network activity in cultured human and mouse neurons. The approach allows analysis of the overall network activity and of single-neuron dynamics, and is amenable to screening purposes.

RESULTS

We validated the new protocol by assessing human neurons with a heterozygous conditional deletion of Munc18-1, and mouse neurons with a homozygous conditional deletion of neurexins. The approach described enabled identification of differential changes in these mutant neurons, allowing quantifications of the synchronous firing rate at the network level and of the amplitude and frequency of Ca²⁺-spikes at the single-neuron level. These results demonstrate the utility of the approach.

COMPARISION WITH EXISTING METHODS

Compared with current imaging platforms, our method is simple, scalable, accessible, and easy to implement. It enables quantification of more detailed parameters than multi-electrode arrays, but does not have the resolution and depth of more sophisticated yet labour-intensive methods, such as patch-clamp electrophysiology.

CONCLUSION

The method reported here is scalable for a rapid direct assessment of neuronal function in culture, and can be applied to both human and mouse neurons. Thus, the method can serve as a basis for phenotypical analysis of mutations and for drug discovery efforts.

PiQSARS: A pipeline for quantitative and statistical analyses of ratiometric fluorescent biosensors

Genetically encoded ratiometric fluorescent probes are cutting-edge tools in biology. They allow precise and dynamic measurement of various physiological parameters within cell compartments. Because data extraction and analysis are time consuming and may lead to inconsistencies between results, we describe here a standardized pipeline for•Semi-automated treatment of time-lapse fluorescence microscopy images.•Quantification of individual cell signal.•Statistical analysis of the data.First, a dedicated macro was developed using the FIJI software to reproducibly quantify the fluorescence ratio as a function of time. Raw data are then exported and analyzed using R and MATLAB softwares. Calculation and statistical analysis of selected graphic parameters are performed. In addition, a functional principal component analysis allows summarizing the dataset. Finally, a principal component analysis is performed to check consistency and final analysis is presented as a visual diagram. The method is adapted to any ratiometric fluorescent probe. As an example, the analysis of the cytoplasmic HyPer probe in response to an acute cell treatment with increasing amounts of hydrogen peroxide is shown. In conclusion, the pipeline allows to save time and analyze a larger amount of samples while reducing manual interventions and consequently increasing the robustness of the analysis.

Impact of blurs on machine-learning aided digital pathology image analysis

BACKGROUND

Digital pathology image (DPI) analysis has been developed by machine learning (ML) techniques. However, little attention has been paid to the reproducibility of ML-based histological classification in heterochronously obtained DPIs of the same hematoxylin and eosin (HE) slide.

AIM

To elucidate the frequency and preventable causes of discordant classification results of DPI analysis using ML for the heterochronously obtained DPIs.

METHODS

We created paired DPIs by scanning 298 HE stained slides containing 584 tissues twice with a virtual slide scanner. The paired DPIs were analyzed by our ML-aided classification model. We defined non-flipped and flipped groups as the paired DPIs with concordant and discordant classification results, respectively. We compared differences in color and blur between the non-flipped and flipped groups by L1-norm and a blur index, respectively.

RESULTS

We observed discordant classification results in 23.1% of the paired DPIs obtained by two independent scans of the same microscope slide. We detected no significant difference in the L1-norm of each color channel between the two groups; however, the flipped group showed a significantly higher blur index than the non-flipped group.

CONCLUSION

Our results suggest that differences in the blur - not the color - of the paired DPIs may cause discordant classification results. An ML-aided classification model for DPI should be tested for this potential cause of the reduced reproducibility of the model. In a future study, a slide scanner and/or a preprocessing method of minimizing DPI blur should be developed.

A systems approach to understand microplastic occurrence and variability in Dutch riverine surface waters

Assessment methods on data quality and environmental variability are lacking for microplastics (MP). Here we assess occurrence and variability of MP number concentrations in two Dutch rivers. Strict QA/QC procedures were applied to identify MP using Fourier-transform infrared (FTIR) microscopy followed by state of the art automated image analysis. For a series of randomly selected, yet ever smaller subareas of filters, we assessed how accurately MP numbers and polymer types are represented during partial filter analysis. Levels of uncertainty were acceptable when analysing 50% of a filter during chemical mapping, and when identifying at least a subset of 50 individual particles with attenuated total reflection (ATR)-FTIR. Applying these guidelines, MP number concentrations between 67 and 11532 MP m^-3 were detected in Dutch riverine surface waters. Spatial differences caused MP number concentrations to vary by two orders of magnitude. Temporal differences were lower and induced a maximum variation of one order of magnitude. In total, 26 polymer types were identified, the most common were polyethylene (23%), polypropylene (19.7%) and ethylene propylene diene monomer rubber (18.3%). The highest diversity of polymer types was found for small MPs, whereas MP larger than 1 mm was scarce and almost exclusively made of polyethylene or polypropylene. Virtually all sampling locations revealed MP number concentrations that are considerably below known effect thresholds for anticipated adverse ecological effects.

Light-Inducible CRISPR Labeling

CRISPR labeling is a powerful technique to study the chromatin architecture in live cells. In CRISPR labeling, a catalytically dead CRISPR-Cas9 mutant is employed as programmable DNA-binding domain to recruit fluorescent proteins to selected genomic loci. The fluorescently labeled loci can then be identified as fluorescent spots and tracked over time by microscopy. A limitation of this approach is the lack of temporal control of the labeling process itself: Cas9 binds to the g(uide)RNA-complementary target loci as soon as it is expressed. The decoration of the genome with Cas9 molecules will, however, interfere with gene regulation and-possibly-affect the genome architecture itself. The ability to switch on and off Cas9 DNA binding in CRISPR labeling experiments would thus be important to enable more precise interrogations of the chromatin spatial organization and dynamics and could further be used to study Cas9 DNA binding kinetics directly in living human cells.Here, we describe a detailed protocol for light-inducible CRISPR labeling. Our method employs CASANOVA, an engineered, optogenetic anti-CRISPR protein, which efficiently traps the Streptococcus pyogenes (Spy)Cas9 in the dark, but permits Cas9 DNA targeting upon illumination with blue light. Using telomeres as exemplary target loci, we detail the experimental steps required for inducible CRISPR labeling with CASANOVA. We also provide instructions on how to analyze the resulting microscopy data in a fully automated fashion.

Imaging rRNA Methylation in Bacteria by MR-FISH

Methylation of RNA is normally monitored in purified cell lysates using next-generation sequencing, gel electrophoresis, or mass spectrometry as readouts. These bulk methods require the RNA from ~10⁴ to 10⁷ cells to be pooled to generate sufficient material for analysis. Here we describe a method-methylation-sensitive RNA in situ hybridization (MR-FISH)-that assays rRNA methylation in bacteria on a cell-by-cell basis, using methylation-sensitive hybridization probes and fluorescence microscopy. We outline step-by-step protocols for designing probes, in situ hybridization, and analysis of data using freely available code.

Image-based Quantification of Direct Cell-to-cell Transmission of Bovine Viral Diarrhea Virus

Different viruses rely on direct cell-to-cell transmission to propagate infection within the infected host. Measuring this mode of transmission in cultured cells is often complicated by the contribution of cell free viruses to spread, and the difficulty to distinguish between primary infected cells that produce the virus and neighboring cells that are the target of spreading. Here, we present a protocol to quantify cell-to-cell transmission of the model pestivirus bovine viral diarrhea virus that is based on the co-culture of producer cells that are infected with a reporter virus expressing mCherry and target cells that stably express GFP. Spread of cell-free viruses is blocked by the presence of a neutralizing antibody in the cell culture medium, and cell-associated transmission is unequivocally quantified by numbering cells that are positive for both GFP and mCherry using automated analysis of fluorescence microscopy images.

Automated image-analysis method for the quantification of fiber morphometry and fiber type population in human skeletal muscle

Background

The quantitative analysis of muscle histomorphometry has been growing in importance in both research and clinical settings. Accurate and stringent assessment of myofibers’ changes in size and number, and alterations in the proportion of oxidative (type I) and glycolytic (type II) fibers is essential for the appropriate study of aging and pathological muscle, as well as for diagnosis and follow-up of muscle diseases. Manual and semi-automated methods to assess muscle morphometry in sections are time-consuming, limited to a small field of analysis, and susceptible to bias, while most automated methods have been only tested in rodent muscle.

Methods

We developed a new macro script for Fiji-ImageJ to automatically assess human fiber morphometry in digital images of the entire muscle. We tested the functionality of our method in deltoid muscle biopsies from a heterogeneous population of subjects with histologically normal muscle (male, female, old, young, lean, obese) and patients with dermatomyositis, necrotizing autoimmune myopathy, and anti-synthetase syndrome myopathy.

Results

Our macro is fully automated, requires no user intervention, and demonstrated improved fiber segmentation by running a series of image pre-processing steps before the analysis. Likewise, our tool showed high accuracy, as compared with manual methods, for identifying the total number of fibers (r = 0.97, p < 0.001), fiber I and fiber II proportion (r = 0.92, p < 0.001), and minor diameter (r = 0.86, p < 0.001) while conducting analysis in ~ 5 min/sample. The performance of the macro analysis was maintained in pectoral and deltoid samples from subjects of different age, gender, body weight, and muscle status. The output of the analyses includes excel files with the quantification of fibers’ morphometry and color-coded maps based on the fiber’s size, which proved to be an advantageous feature for the fast and easy visual identification of location-specific atrophy and a potential tool for medical diagnosis.

Conclusion

Our macro is reliable and suitable for the study of human skeletal muscle for research and for diagnosis in clinical settings providing reproducible and consistent analysis when the time is of the utmost importance.

Electronic supplementary material

The online version of this article (10.1186/s13395-019-0200-7) contains supplementary material, which is available to authorized users.

Quantifying Ca2+ signaling and contraction in vascular pericytes and smooth muscle cells

Vascular pericytes and smooth muscle cells surround many blood vessels of the body. Their primary roles include vessel stabilization and regulation of the blood flow. The high degree of heterogeneity among these cells is dictated by (1) differences in their developmental origin and (2) their location in the vascular bed. Phenotype switching contributes to this heterogeneity especially following in vitro culture. In the absence of distinguishing molecular markers, functional assays that capture their heterogeneity in vitro are needed. Spatiotemporal changes in intracellular Ca²⁺ levels and contraction in response to vasoconstrictors reflect the differences between vascular pericyte and smooth muscle cell. In order to capture this heterogeneity in vitro, large ensembles of cells need to be analyzed. Here we developed an automated image processing method to measure intracellular Ca²⁺ and contraction in large cell groups which in combination with a computational approach for integrative analysis allowed vascular pericytes and smooth muscle cells to be distinguished without knowledge of their anatomical origin.

Stromal cell ratio based on automated image analysis as a predictor for platinum-resistant recurrent ovarian cancer

Background

Identifying high-risk patients for platinum resistance is critical for improving clinical management of ovarian cancer. We aimed to use automated image analysis of hematoxylin & eosin (H&E) stained sections to identify the association between microenvironmental composition and platinum-resistant recurrent ovarian cancer.

Methods

Ninety-one patients with ovarian cancer containing the data of automated image analysis for H&E histological sections were initially reviewed.

Results

Seventy-one patients with recurrent disease were finally identified. Among 30 patients with high stromal cell ratio, 60% of the patients had platinum-resistant recurrence, which was significantly higher than the rate in patients with low stromal cell ratio (9.80%, P <  0.001). Multivariate logistic regression analysis revealed elevated CA125 level after 3 cycles of chemotherapy (P <  0.001) and high stromal cell ratio (P = 0.002) were the negative predictors of platinum-resistant relapse. The area under the curve (AUC) of receiver operating characteristic (ROC) curves of the models for predicting platinum-resistant recurrence with stromal cell ratio, normalization of CA125 level, and the combination of two parameters were 0.78, 0.79, and 0.89 respectively.

Conclusions

Our results demonstrated stromal cell ratio based on automated image analysis may be a potential predictor for ovarian cancer patients at high risk of platinum-resistant recurrence, and it could improve the predictive value of model when combined with normalization of CA125 level after 3 cycles of chemotherapy.

Accelerating single molecule localization microscopy through parallel processing on a high-performance computing cluster

Super‐resolved microscopy techniques have revolutionized the ability to study biological structures below the diffraction limit. Single molecule localization microscopy (SMLM) techniques are widely used because they are relatively straightforward to implement and can be realized at relatively low cost, e.g. compared to laser scanning microscopy techniques. However, while the data analysis can be readily undertaken using open source or other software tools, large SMLM data volumes and the complexity of the algorithms used often lead to long image data processing times that can hinder the iterative optimization of experiments. There is increasing interest in high throughput SMLM, but its further development and application is inhibited by the data processing challenges. We present here a widely applicable approach to accelerating SMLM data processing via a parallelized implementation of ThunderSTORM on a high‐performance computing (HPC) cluster and quantify the speed advantage for a four‐node cluster (with 24 cores and 128 GB RAM per node) compared to a high specification (28 cores, 128 GB RAM, SSD‐enabled) desktop workstation. This data processing speed can be readily scaled by accessing more HPC resources. Our approach is not specific to ThunderSTORM and can be adapted for a wide range of SMLM software.

Lay Description

Optical microscopy is now able to provide images with a resolution far beyond the diffraction limit thanks to relatively new super‐resolved microscopy (SRM) techniques, which have revolutionized the ability to study biological structures. One approach to SRM is to randomly switch on and off the emission of fluorescent molecules in an otherwise conventional fluorescence microscope. If only a sparse subset of the fluorescent molecules labelling a sample can be switched on at a time, then each emitter will be, on average, spaced further apart than the diffraction‐limited resolution of the conventional microscope and the separate bright spots in the image corresponding to each emitter can be localised to high precision by finding the centre of each feature using a computer program. Thus, a precise map of the emitter positions can be recorded by sequentially mapping the localisation of different subsets of emitters as they are switched on and others switched off.

Typically, this approach, described as single molecule localisation microscopy (SMLM), results in large image data sets that can take many minutes to hours to process, depending on the size of the field of view and whether the SMLM analysis employs a computationally‐intensive iterative algorithm. Such a slow workflow makes it difficult to optimise experiments and to analyse large numbers of samples. Faster SMLM experiments would be generally useful and automated high throughput SMLM studies of arrays of samples, such as cells, could be applied to drug discovery and other applications. However, the time required to process the resulting data would be prohibitive on a normal computer.

To address this, we have developed a method to run standard SMLM data analysis software tools in parallel on a high‐performance computing cluster (HPC). This can be used to accelerate the analysis of individual SMLM experiments or it can be scaled to analyse high throughput SMLM data by extending it to run on an arbitrary number of HPC processors in parallel. In this paper we outline the design of our parallelised SMLM software for HPC and quantify the speed advantage when implementing it on four HPC nodes compared to a powerful desktop computer.

Detection of herpesvirus capsids in transmission electron microscopy images using transfer learning

The detailed analysis of secondary envelopment of the Human betaherpesvirus 5/human cytomegalovirus (HCMV) from transmission electron microscopy (TEM) images is an important step towards understanding the mechanisms underlying the formation of infectious virions. As a step towards a software-based quantification of different stages of HCMV virion morphogenesis in TEM, we developed a transfer learning approach based on convolutional neural networks (CNNs) that automatically detects HCMV nucleocapsids in TEM images. In contrast to existing image analysis techniques that require time-consuming manual definition of structural features, our method automatically learns discriminative features from raw images without the need for extensive pre-processing. For this a constantly growing TEM image database of HCMV infected cells was available which is unique regarding image quality and size in the terms of virological EM. From the two investigated types of transfer learning approaches, namely feature extraction and fine-tuning, the latter enabled us to successfully detect HCMV nucleocapsids in TEM images. Our detection method has outperformed some of the existing image analysis methods based on discriminative textural indicators and radial density profiles for virus detection in TEM images. In summary, we could show that the method of transfer learning can be used for an automated detection of viral capsids in TEM images with high specificity using standard computers. This method is highly adaptable and in future could be easily extended to automatically detect and classify virions of other viruses and even distinguish different virion maturation stages.

High Content, Phenotypic Assays and Screens for Compounds Modulating Cellular Processes in Primary Neurons

High content, phenotypic screens offer a powerful approach to systems biology at the cellular level. The approach employs cells carrying fluorescently labeled molecules or organelles in 384- or 1536-well microplates, and an automated confocal screening microscope for capturing images from each well. Although some specifics vary according to the assay type, each will apply some degree of image processing and feature extraction followed by a data analysis pipeline to identify the perturbations (small molecules, etc.) of interest. We describe and discuss the advantages and limitations of high content assays and screens using the specific example of assaying mitochondrial dynamics in primary neurons. We provide a detailed description of our culturing methods, imaging and data analysis techniques and provide an open source, ready to use CellProfiler pipeline for high-throughput image segmentation and quantification tool for mitochondrial parameters.

Focused scores enable reliable discrimination of small differences in steatosis

Background

Automated image analysis enables quantitative measurement of steatosis in histological images. However, spatial heterogeneity of steatosis can make quantitative steatosis scores unreliable. To improve the reliability, we have developed novel scores that are “focused” on steatotic tissue areas.

Methods

Focused scores use concepts of tile-based hotspot analysis in order to compute statistics about steatotic tissue areas in an objective way. We evaluated focused scores on three data sets of images of rodent liver sections exhibiting different amounts of dietary-induced steatosis. The same evaluation was conducted with the standard steatosis score computed by most image analysis methods.

Results

The standard score reliably discriminated large differences in steatosis (intraclass correlation coefficient ICC = 0.86), but failed to discriminate small (ICC = 0.54) and very small (ICC = 0.14) differences. With an appropriate tile size, mean-based focused scores reliably discriminated large (ICC = 0.92), small (ICC = 0.86) and very small (ICC = 0.83) differences. Focused scores based on high percentiles showed promise in further improving the discrimination of very small differences (ICC = 0.93).

Conclusions

Focused scores enable reliable discrimination of small differences in steatosis in histological images. They are conceptually simple and straightforward to use in research studies.

An automated mapping method for Nissl-stained mouse brain histologic sections

BACKGROUND

Histologic evaluation of the central nervous system is often a critical endpoint in in vivo efficacy studies, and is considered the essential component of neurotoxicity assessment in safety studies. Automated image analysis is a powerful tool that can radically reduce the workload associated with evaluating brain histologic sections.

NEW METHOD

We developed an automated brain mapping method that identifies neuroanatomic structures in mouse histologic coronal brain sections. The method utilizes the publicly available Allen Brain Atlas to map brain regions on digitized Nissl-stained sections.

RESULTS

The method's accuracy was first assessed by comparing the mapping results to structure delineations from the Franklin and Paxinos (FP) mouse brain atlas. Brain regions mapped from FP Nissl-stained sections and calculated volumes were similar to structure delineations and volumes derived from corresponding FP illustrations. We subsequently applied our method to mouse brain sections from an in vivo study where the hippocampus was the structure of interest. Nissl-stained sections were mapped and hippocampal boundaries transferred to adjacent immunohistochemically stained sections. Optical density quantification results were comparable to those from time-consuming, manually drawn hippocampal delineations on the IHC-stained sections.

COMPARISON WITH EXISTING METHODS

Compared to other published methods, our method requires less manual input, and has been validated comprehensively using a secondary atlas, as well as manually annotated brain IHC sections from 68 study mice.

CONCLUSIONS

We propose that our automated brain mapping method enables greater efficiency and consistency in mouse neuropathologic assessments.

Laccase production and pellet morphology of Coprinopsis cinerea transformants in liquid shake flask cultures

Laccase production and pellet formation of transformants of Coprinopsis cinerea strain FA2222 of C. cinerea laccase gene lcc1 subcloned behind the gpdII-promoter from Agaricus bisporus were compared with a control transformant carrying no extra laccase gene. At the optimum growth temperature of 37 °C, maximal laccase yields of 2.9 U/ml were obtained by the best lcc1 transformant pYSK7-26 in liquid shake flask cultures. Reduction in temperature to 25 °C increased laccase yields up to 9.2 U/ml. The control transformant had no laccase activities at 37 °C but native activity at 25 °C (3.5 U/ml). Changing the temperature had severe effects on the morphology of the mycelial pellets formed during cultivation, but links of distinct pellet morphologies to native or recombinant laccase production could not be established. Automated image analysis was used to characterise pellet formation and morphological parameters (pellet area, diameter, convexity and mycelial structure). Cross sections of selected pellets showed that they differentiated in an outer rind and an inner medulla of loosened hyphae. Pellets at 25 °C had a small and dense outer zone and adopted with time a smooth surface. Pellets at 37 °C had a broader outer zone and a fringy surface due to generation of more and larger protuberances in the rind that when released can serve for production of further pellets.

Automated histological classification of whole-slide images of gastric biopsy specimens

BACKGROUND

Automated image analysis has been developed currently in the field of surgical pathology. The aim of the present study was to evaluate the classification accuracy of the e-Pathologist image analysis software.

METHODS

A total of 3062 gastric biopsy specimens were consecutively obtained and stained. The specimen slides were anonymized and digitized. At least two experienced gastrointestinal pathologists evaluated each slide for pathological diagnosis. We compared the three-tier (positive for carcinoma or suspicion of carcinoma; caution for adenoma or suspicion of a neoplastic lesion; or negative for a neoplastic lesion) or two-tier (negative or non-negative) classification results of human pathologists and of the e-Pathologist.

RESULTS

Of 3062 cases, 33.4% showed an abnormal finding. For the three-tier classification, the overall concordance rate was 55.6% (1702/3062). The kappa coefficient was 0.28 (95% CI, 0.26-0.30; fair agreement). For the negative biopsy specimens, the concordance rate was 90.6% (1033/1140), but for the positive biopsy specimens, the concordance rate was less than 50%. For the two-tier classification, the sensitivity, specificity, positive predictive value, and negative predictive value were 89.5% (95% CI, 87.5-91.4%), 50.7% (95% CI, 48.5-52.9%), 47.7% (95% CI, 45.4-49.9%), and 90.6% (95% CI, 88.8-92.2%), respectively.

CONCLUSIONS

Although there are limitations and requirements for applying automated histopathological classification of gastric biopsy specimens in the clinical setting, the results of the present study are promising.

Characterization of glycosylphosphatidylinositol biosynthesis defects by clinical features, flow cytometry, and automated image analysis

Background

Glycosylphosphatidylinositol biosynthesis defects (GPIBDs) cause a group of phenotypically overlapping recessive syndromes with intellectual disability, for which pathogenic mutations have been described in 16 genes of the corresponding molecular pathway. An elevated serum activity of alkaline phosphatase (AP), a GPI-linked enzyme, has been used to assign GPIBDs to the phenotypic series of hyperphosphatasia with mental retardation syndrome (HPMRS) and to distinguish them from another subset of GPIBDs, termed multiple congenital anomalies hypotonia seizures syndrome (MCAHS). However, the increasing number of individuals with a GPIBD shows that hyperphosphatasia is a variable feature that is not ideal for a clinical classification.

Methods

We studied the discriminatory power of multiple GPI-linked substrates that were assessed by flow cytometry in blood cells and fibroblasts of 39 and 14 individuals with a GPIBD, respectively. On the phenotypic level, we evaluated the frequency of occurrence of clinical symptoms and analyzed the performance of computer-assisted image analysis of the facial gestalt in 91 individuals.

Results

We found that certain malformations such as Morbus Hirschsprung and diaphragmatic defects are more likely to be associated with particular gene defects (PIGV, PGAP3, PIGN). However, especially at the severe end of the clinical spectrum of HPMRS, there is a high phenotypic overlap with MCAHS. Elevation of AP has also been documented in some of the individuals with MCAHS, namely those with PIGA mutations. Although the impairment of GPI-linked substrates is supposed to play the key role in the pathophysiology of GPIBDs, we could not observe gene-specific profiles for flow cytometric markers or a correlation between their cell surface levels and the severity of the phenotype. In contrast, it was facial recognition software that achieved the highest accuracy in predicting the disease-causing gene in a GPIBD.

Conclusions

Due to the overlapping clinical spectrum of both HPMRS and MCAHS in the majority of affected individuals, the elevation of AP and the reduced surface levels of GPI-linked markers in both groups, a common classification as GPIBDs is recommended. The effectiveness of computer-assisted gestalt analysis for the correct gene inference in a GPIBD and probably beyond is remarkable and illustrates how the information contained in human faces is pivotal in the delineation of genetic entities.

Electronic supplementary material

The online version of this article (doi:10.1186/s13073-017-0510-5) contains supplementary material, which is available to authorized users.

Automated quantification of steatosis: agreement with stereological point counting

BACKGROUND

Steatosis is routinely assessed histologically in clinical practice and research. Automated image analysis can reduce the effort of quantifying steatosis. Since reproducibility is essential for practical use, we have evaluated different analysis methods in terms of their agreement with stereological point counting (SPC) performed by a hepatologist.

METHODS

The evaluation was based on a large and representative data set of 970 histological images from human patients with different liver diseases. Three of the evaluated methods were built on previously published approaches. One method incorporated a new approach to improve the robustness to image variability.

RESULTS

The new method showed the strongest agreement with the expert. At 20× resolution, it reproduced steatosis area fractions with a mean absolute error of 0.011 for absent or mild steatosis and 0.036 for moderate or severe steatosis. At 10× resolution, it was more accurate than and twice as fast as all other methods at 20× resolution. When compared with SPC performed by two additional human observers, its error was substantially lower than one and only slightly above the other observer.

CONCLUSIONS

The results suggest that the new method can be a suitable automated replacement for SPC. Before further improvements can be verified, it is necessary to thoroughly assess the variability of SPC between human observers.

Automated image analysis of placental villi and syncytial knots in histological sections

INTRODUCTION

Delayed villous maturation and accelerated villous maturation diagnosed in histologic sections are morphologic manifestations of pathophysiological conditions. The inter-observer agreement among pathologists in assessing these conditions is moderate at best. We investigated whether automated image analysis of placental villi and syncytial knots could improve standardization in diagnosing these conditions.

METHODS

Placentas of antepartum fetal death at or near term were diagnosed as normal, delayed or accelerated villous maturation. Histologic sections of 5 cases per group were photographed at ×10 magnification. Automated image analysis of villi and syncytial knots was performed, using ImageJ public domain software. Analysis of hundreds of histologic images was carried out within minutes on a personal computer, using macro commands.

RESULTS

Compared to normal placentas, villi from delayed maturation were larger and fewer, with fewer and smaller syncytial knots. Villi from accelerated maturation were smaller. The data were further analyzed according to horizontal placental zones and groups of villous size.

DISCUSSION

Normal placentas can be discriminated from placentas of delayed or accelerated villous maturation using automated image analysis. Automated image analysis of villi and syncytial knots is not equivalent to interpretation by the human eye. Each method has advantages and disadvantages in assessing the 2-dimensional histologic sections representing the complex, 3-dimensional villous tree. Image analysis of placentas provides quantitative data that might help in standardizing and grading of placentas for diagnostic and research purposes.

Cell migration in schizophrenia: Patient-derived cells do not regulate motility in response to extracellular matrix

Schizophrenia is a highly heritable psychiatric disorder linked to a large number of risk genes. The function of these genes in disease etiology is not fully understood but pathway analyses of genomic data suggest developmental dysregulation of cellular processes such as neuronal migration and axon guidance. Previous studies of patient-derived olfactory cells show them to be more motile than control-derived cells when grown on a fibronectin substrate, motility that is dependent on focal adhesion kinase signaling. The aim of this study was to investigate whether schizophrenia patient-derived cells are responsive to other extracellular matrix (ECM) proteins that bind integrin receptors. Olfactory neurosphere-derived cells from nine patients and nine matched controls were grown on ECM protein substrates at increasing concentrations and their movement was tracked for 24h using automated high-throughput imaging. Control-derived cells increased their motility as the ECM substrate concentration increased, whereas patient-derived cell motility was little affected by ECM proteins. Patient and control cells had appropriate integrin receptors for these ECM substrates and detected them as shown by increases in focal adhesion number and size in response to ECM proteins, which also induced changes in cell morphology and cytoskeleton. These observations indicate that patient cells failed to translate the detection of ECM proteins into appropriate changes in cell motility. In a sense, patient cells act like a moving car whose accelerator is jammed, moving at the same speed without regard to the external environment. This focuses attention on cell motility regulation rather than speed as key to impairment of neuronal migration in the developing brain in schizophrenia.

Validation of adult height prediction based on automated bone age determination in the Paris Longitudinal Study of healthy children

BACKGROUND

An adult height prediction model based on automated determination of bone age was developed and validated in two studies from Zurich, Switzerland. Varied living conditions and genetic backgrounds might make the model less accurate.

OBJECTIVE

To validate the adult height prediction model on children from another geographical location.

MATERIALS AND METHODS

We included 51 boys and 58 girls from the Paris Longitudinal Study of children born 1953 to 1958. Radiographs were obtained once or twice a year in these children from birth to age 18. Bone age was determined using the BoneXpert method. Radiographs in children with bone age greater than 6 years were considered, in total 1,124 images.

RESULTS

The root mean square deviation between the predicted and the observed adult height was 2.8 cm for boys in the bone age range 6-15 years and 3.1 cm for girls in the bone age range 6-13 years. The bias (the average signed difference) was zero, except for girls below bone age 12, where the predictions were 0.8 cm too low.

CONCLUSION

The accuracy of the BoneXpert method in terms of root mean square error was as predicted by the model, i.e. in line with what was observed in the Zurich studies.

Bimodality of intratumor Ki67 expression is an independent prognostic factor of overall survival in patients with invasive breast carcinoma

Proliferative activity, assessed by Ki67 immunohistochemistry (IHC), is an established prognostic and predictive biomarker of breast cancer (BC). However, it remains under-utilized due to lack of standardized robust measurement methodologies and significant intratumor heterogeneity of expression. A recently proposed methodology for IHC biomarker assessment in whole slide images (WSI), based on systematic subsampling of tissue information extracted by digital image analysis (DIA) into hexagonal tiling arrays, enables computation of a comprehensive set of Ki67 indicators, including intratumor variability. In this study, the tiling methodology was applied to assess Ki67 expression in WSI of 152 surgically removed Ki67-stained (on full-face sections) BC specimens and to test which, if any, Ki67 indicators can predict overall survival (OS). Visual Ki67 IHC estimates and conventional clinico-pathologic parameters were also included in the study. Analysis revealed linearly independent intrinsic factors of the Ki67 IHC variance: proliferation (level of expression), disordered texture (entropy), tumor size and Nottingham Prognostic Index, bimodality, and correlation. All visual and DIA-generated indicators of the level of Ki67 expression provided significant cutoff values as single predictors of OS. However, only bimodality indicators (Ashman's D, in particular) were independent predictors of OS in the context of hormone receptor and HER2 status. From this, we conclude that spatial heterogeneity of proliferative tumor activity, measured by DIA of Ki67 IHC expression and analyzed by the hexagonal tiling approach, can serve as an independent prognostic indicator of OS in BC patients that outperforms the prognostic power of the level of proliferative activity.

Quantitative Functional Morphology by Imaging Flow Cytometry

This chapter describes advantages and limitations of imaging flow cytometry (IFC) based on Imagestream instrumentation using a hybrid approach of morphometric measurement and quantitation of multiparametric fluorescent intensities' distribution in cells and particles. Brief comparison is given of IFC with conventional flow cytometry and fluorescent microscopy. Some future directions of the IFC technology are described and discussed.

Automated image analysis reveals the dynamic 3-dimensional organization of multi-ciliary arrays

Multi-ciliated cells (MCCs) use polarized fields of undulating cilia (ciliary array) to produce fluid flow that is essential for many biological processes. Cilia are positioned by microtubule scaffolds called basal bodies (BBs) that are arranged within a spatially complex 3-dimensional geometry (3D). Here, we develop a robust and automated computational image analysis routine to quantify 3D BB organization in the ciliate, Tetrahymena thermophila. Using this routine, we generate the first morphologically constrained 3D reconstructions of Tetrahymena cells and elucidate rules that govern the kinetics of MCC organization. We demonstrate the interplay between BB duplication and cell size expansion through the cell cycle. In mutant cells, we identify a potential BB surveillance mechanism that balances large gaps in BB spacing by increasing the frequency of closely spaced BBs in other regions of the cell. Finally, by taking advantage of a mutant predisposed to BB disorganization, we locate the spatial domains that are most prone to disorganization by environmental stimuli. Collectively, our analyses reveal the importance of quantitative image analysis to understand the principles that guide the 3D organization of MCCs.

Summary: We develop an automated computational image analysis routine to quantify basal body organization, which detects subtle spatial phenotypes resulting from environmental and genetic perturbations.

A methodology for comprehensive breast cancer Ki67 labeling index with intra-tumor heterogeneity appraisal based on hexagonal tiling of digital image analysis data

Digital image analysis (DIA) enables higher accuracy, reproducibility, and capacity to enumerate cell populations by immunohistochemistry; however, the most unique benefits may be obtained by evaluating the spatial distribution and intra-tissue variance of markers. The proliferative activity of breast cancer tissue, estimated by the Ki67 labeling index (Ki67 LI), is a prognostic and predictive biomarker requiring robust measurement methodologies. We performed DIA on whole-slide images (WSI) of 302 surgically removed Ki67-stained breast cancer specimens; the tumour classifier algorithm was used to automatically detect tumour tissue but was not trained to distinguish between invasive and non-invasive carcinoma cells. The WSI DIA-generated data were subsampled by hexagonal tiling (HexT). Distribution and texture parameters were compared to conventional WSI DIA and pathology report data. Factor analysis of the data set, including total numbers of tumor cells, the Ki67 LI and Ki67 distribution, and texture indicators, extracted 4 factors, identified as entropy, proliferation, bimodality, and cellularity. The factor scores were further utilized in cluster analysis, outlining subcategories of heterogeneous tumors with predominant entropy, bimodality, or both at different levels of proliferative activity. The methodology also allowed the visualization of Ki67 LI heterogeneity in tumors and the automated detection and quantitative evaluation of Ki67 hotspots, based on the upper quintile of the HexT data, conceptualized as the "Pareto hotspot". We conclude that systematic subsampling of DIA-generated data into HexT enables comprehensive Ki67 LI analysis that reflects aspects of intra-tumor heterogeneity and may serve as a methodology to improve digital immunohistochemistry in general.

ROCK1 and ROCK2 inhibition alters dendritic spine morphology in hippocampal neurons

Communication among neurons is mediated through synaptic connections between axons and dendrites, and most excitatory synapses occur on actin-rich protrusions along dendrites called dendritic spines. Dendritic spines are structurally dynamic, and synapse strength is closely correlated with spine morphology. Abnormalities in the size, shape, and number of dendritic spines are prevalent in neurologic diseases, including autism spectrum disorders, schizophrenia, and Alzheimer disease. However, therapeutic targets that influence spine morphology are lacking. Rho-associated coiled-coil containing protein kinases (ROCK) 1 and ROCK2 are potent regulators of the actin cytoskeleton and highly promising drug targets for central nervous system disorders. In this report, we addressed how pharmacologic inhibition of ROCK1 and ROCK2 affects dendritic spine morphology. Hippocampal neurons were transfected with plasmids expressing fluorescently labeled Lifeact, a small actin binding peptide, and then incubated with or without Y-27632, an established pan-ROCK small molecule inhibitor. Using an automated 3D spine morphometry analysis method, we showed that inhibition of ROCK1 and ROCK2 significantly increased the mean protrusion density and significantly reduced the mean protrusion width. A trending increase in mean protrusion length was observed following Y-27632 treatment, and novel effects were observed among spine classes. Exposure to Y-27632 significantly increased the number of filopodia and thin spines, while the numbers of stubby and mushroom spines were similar to mock-treated samples. These findings support the hypothesis that pharmacologic inhibition of ROCK1 and ROCK2 may convey therapeutic benefit for neurologic disorders that feature dendritic spine loss or aberrant structural plasticity.

Automated image-based analysis of spatio-temporal fungal dynamics

Due to their ability to grow in complex environments, fungi play an important role in most ecosystems and have for that reason been the subject of numerous studies. Some of the main obstacles to the study of fungal growth are the heterogeneity of growth environments and the limited scope of laboratory experiments. Given the increasing availability of image capturing techniques, a new approach lies in image analysis. Most previous image analysis studies involve manual labelling of the fungal network, tracking of individual hyphae, or invasive techniques that do not allow for tracking the evolution of the entire fungal network. In response, this work presents a highly versatile tool combining image analysis and graph theory to monitor fungal growth through time and space for different fungal species and image resolutions. In addition, a new experimental set-up is presented that allows for a functional description of fungal growth dynamics and a quantitative mutual comparison of different growth behaviors. The presented method is completely automated and facilitates the extraction of the most studied fungal growth features such as the total length of the mycelium, the area of the mycelium and the fractal dimension. The compactness of the fungal network can also be monitored over time by computing measures such as the number of tips, the node degree and the number of nodes. Finally, the average growth angle and the internodal length can be extracted to study the morphology of the fungi. In summary, the introduced method offers an updated and broader alternative to classical and narrowly focused approaches, thus opening new avenues of investigation in the field of mycology.