Nuquantus: Machine learning software for the characterization and quantification of cell nuclei in complex immunofluorescent tissue images

A comparison of machine learning approaches for the quantification of microglial cells in the brain of mice, rats and non-human primates

Microglial cells are brain-specific macrophages that swiftly react to disruptive events in the brain. Microglial activation leads to specific modifications, including proliferation, morphological changes, migration to the site of insult, and changes in gene expression profiles. A change in inflammatory status has been linked to many neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease. For this reason, the investigation and quantification of microglial cells is essential for better understanding their role in disease progression as well as for evaluating the cytocompatibility of novel therapeutic approaches for such conditions. In the following study we implemented a machine learning-based approach for the fast and automatized quantification of microglial cells; this tool was compared with manual quantification (ground truth), and with alternative free-ware such as the threshold-based ImageJ and the machine learning-based Ilastik. We first trained the algorithms on brain tissue obtained from rats and non-human primate immunohistochemically labelled for microglia. Subsequently we validated the accuracy of the trained algorithms in a preclinical rodent model of Parkinson's disease and demonstrated the robustness of the algorithms on tissue obtained from mice, as well as from images provided by three collaborating laboratories. Our results indicate that machine learning algorithms can detect and quantify microglial cells in all the three mammalian species in a precise manner, equipotent to the one observed following manual counting. Using this tool, we were able to detect and quantify small changes between the hemispheres, suggesting the power and reliability of the algorithm. Such a tool will be very useful for investigation of microglial response in disease development, as well as in the investigation of compatible novel therapeutics targeting the brain. As all network weights and labelled training data are made available, together with our step-by-step user guide, we anticipate that many laboratories will implement machine learning-based quantification of microglial cells in their research.

Machine Learning in Tissue Engineering

Machine learning (ML) and artificial intelligence have accelerated scientific discovery, augmented clinical practice, and deepened fundamental understanding of many biological phenomena. ML technologies have now been applied to diverse areas of tissue engineering research, including biomaterial design, scaffold fabrication, and cell/tissue modeling. Emerging ML-empowered strategies include machine-optimized polymer synthesis, predictive modeling of scaffold fabrication processes, complex analyses of structure-function relationships, and deep learning of spatialized cell phenotypes and tissue composition. The emergence of ML in tissue engineering, while relatively recent, has already enabled increasingly complex and multivariate analyses of the relationships between biological, chemical, and physical factors in driving tissue regenerative outcomes. This review highlights the novel methodologies, emerging strategies, and areas of potential growth within this rapidly evolving area of research. Impact statement Machine learning (ML) has accelerated scientific discovery and augmented clinical practice across multiple fields. Now, ML has driven exciting new paradigms in tissue engineering research, including machine-optimized biomaterial design, predictive modeling of scaffold fabrication, and spatiotemporal analysis of cell and tissue systems. The emergence of ML in tissue engineering, while relatively recent, has already enabled increasingly complex analyses of the relationships between biological, chemical, and physical factors in driving tissue regenerative outcomes. This review highlights the novel methodologies, emerging strategies, and areas of potential growth within this rapidly evolving area of research.

The finite element method-based pattern recognition approach for the classification of patient-specific gunshot injury

Violence related injuries and deaths mostly caused by firearms are a major problem throughout the world. Understanding the factors that control the extent of hard-soft tissue wound patterns using computer imaging techniques, numerical methods, and machine learning algorithms may help physicians to diagnose and treat those injuries more properly. Here, we investigate the use of computational results coupled with the pattern recognition algorithms to develop an approach for forensic applications. Initially, computer tomography (CT) images of the patient whose leg was shot by a 9 × 19 parabellum bullet are used to construct the FE models of that patient’s femoral bone and the surrounding soft tissues. Then, Hounsfield units-based material properties are assigned to elements of the bone. To simulate the full range of loading conditions encountered in ballistic events, a constitutive model that captures the strain-rate dependent response is implemented. The entrance pathway vector of the bullet is directed in accordance with the patient’s wound and the simulations are deployed for the cases having various inlet velocities such as 150, 200, 250, 300, and 350 m/s. Once the FE results for each case are obtained, they are processed with supervised machine learning algorithms to classify the wound and inlet velocity correspondence. The results demonstrate that they can be diagnosed with a percent accuracy of 97.3, 97.5, and 98.3 for the decision tree (DT), k-nearest neighbors (kNN) and support vector machine (SVM) classifier, respectively. This approach may provide a useful framework in classifying the wound type, predicting the bullet impact velocity and its firing distance.

A review of risk prediction models in cardiovascular disease: conventional approach vs. artificial intelligent approach

Cardiovascular disease (CVD) is the leading cause of death worldwide and is a global health issue. Traditionally, statistical models are used commonly in the risk prediction and assessment of CVD. However, the adoption of artificial intelligent (AI) approach is rapidly taking hold in the current era of technology to evaluate patient risks and predict the outcome of CVD. In this review, we outline various conventional risk scores and prediction models and do a comparison with the AI approach. The strengths and limitations of both conventional and AI approaches are discussed. Besides that, biomarker discovery related to CVD are also elucidated as the biomarkers can be used in the risk stratification as well as early detection of the disease. Moreover, problems and challenges involved in current CVD studies are explored. Lastly, future prospects of CVD risk prediction and assessment in the multi-modality of big data integrative approaches are proposed.

Machine learning applications in radiation oncology

Machine learning technology has a growing impact on radiation oncology with an increasing presence in research and industry. The prevalence of diverse data including 3D imaging and the 3D radiation dose delivery presents potential for future automation and scope for treatment improvements for cancer patients. Harnessing this potential requires standardization of tools and data, and focused collaboration between fields of expertise. The rapid advancement of radiation oncology treatment technologies presents opportunities for machine learning integration with investments targeted towards data quality, data extraction, software, and engagement with clinical expertise. In this review, we provide an overview of machine learning concepts before reviewing advances in applying machine learning to radiation oncology and integrating these techniques into the radiation oncology workflows. Several key areas are outlined in the radiation oncology workflow where machine learning has been applied and where it can have a significant impact in terms of efficiency, consistency in treatment and overall treatment outcomes. This review highlights that machine learning has key early applications in radiation oncology due to the repetitive nature of many tasks that also currently have human review. Standardized data management of routinely collected imaging and radiation dose data are also highlighted as enabling engagement in research utilizing machine learning and the ability integrate these technologies into clinical workflow to benefit patients. Physicists need to be part of the conversation to facilitate this technical integration.

The Use of Quantitative Digital Pathology to Measure Proteoglycan and Glycosaminoglycan Expression and Accumulation in Healthy and Diseased Tissues

Advances in reagents, methodologies, analytic platforms, and tools have resulted in a dramatic transformation of the research pathology laboratory. These advances have increased our ability to efficiently generate substantial volumes of data on the expression and accumulation of mRNA, proteins, carbohydrates, signaling pathways, cells, and structures in healthy and diseased tissues that are objective, quantitative, reproducible, and suitable for statistical analysis. The goal of this review is to identify and present how to acquire the critical information required to measure changes in tissues. Included is a brief overview of two morphometric techniques, image analysis and stereology, and the use of artificial intelligence to classify cells and identify hidden patterns and relationships in digital images. In addition, we explore the importance of preanalytical factors in generating high-quality data. This review focuses on techniques we have used to measure proteoglycans, glycosaminoglycans, and immune cells in tissues using immunohistochemistry and in situ hybridization to demonstrate the various morphometric techniques. When performed correctly, quantitative digital pathology is a powerful tool that provides unbiased quantitative data that are difficult to obtain with other methods.

Classical and Non-Classical Progesterone Signaling in Breast Cancers

Much emphasis is placed on estrogen (E2) and estrogen receptor (ER) signaling as most research is focused on understanding E2 and ER’s ability to enhance proliferative signals in breast cancers. Progesterone (P4) is important for normal mammary gland development, function and menstrual control. However, P4 and its receptors (PRs) in breast cancer etiology continue to be understudied and its role in breast cancer remains controversial. The Women’s Health Initiative (WHI) clinical trial clearly demonstrated the importance of progestogens in breast cancer development. P4 has historically been associated with classical-signaling through nuclear receptors, however non-classical P4 signaling via membrane receptors has been described. Progestogens have the ability to bind to nuclear and membrane receptors and studies have demonstrated that both can promote breast cancer cell proliferation and breast tumor growth. In this review, we attempt to understand the classical and non-classical signaling role of P4 in breast cancers because both nuclear and membrane receptors could become viable therapeutic options for breast cancer patients.

CPNE1 predicts poor prognosis and promotes tumorigenesis and radioresistance via the AKT singling pathway in triple-negative breast cancer

Elevated expression of Copine 1 (CPNE1) has been observed in multiple cancers; however, the underlying mechanisms by which it affects cancer cells are unclear. We aimed to study the effect of CPNE1 on the tumorigenesis and radioresistance of triple‐negative breast cancer (TNBC). Quantitative real‐time polymerase chain reaction was used to detect the expression of CPNE1 in TNBC tissues and cell lines. Western blot, immunohistochemistry, and immunofluorescence were used to investigate the levels of CPNE1, p‐AKT, AKT, cleaved caspase‐3, cleaved PARP1, and γ‐H2AX. Cell viability and apoptosis were measured by CCK‐8 and flow cytometry, respectively. CPNE1 was overexpressed in TNBC tissues and cell lines and was associated with tumor size, distant metastases, and survival rates of patients with TNBC. Moreover, function study shows that CPNE1 promoted cell viability and inhibited cell apoptosis in vitro and inhibited the radiosensitivity of TNBC. Importantly, inactivation of AKT signaling inhibited the tumorigenesis and radioresistance mediated by CPNE1 in TNBC cells. In vivo xenograft study also shows that CPNE1 knockdown inhibited tumor growth and promoted cell apoptosis. Overall, our findings suggest that CPNE1 promotes tumorigenesis and radioresistance in TNBC by regulating AKT activation and targeted CPNE1 expression may be a strategy to sensitize TNBC cells toward radiation therapy.

Utilizing supervised machine learning to identify microglia and astrocytes in situ: implications for large-scale image analysis and quantification

BACKGROUND

The evaluation of histological tissue samples plays a crucial role in deciphering preclinical disease and injury mechanisms. High-resolution images can be obtained quickly however data acquisition are often bottlenecked by manual analysis methodologies.

NEW METHOD

We describe and validate a pipeline for a novel machine learning-based analytical method, using the Opera High-Content Screening system and Harmony software, allowing for detailed image analysis of cellular markers in histological samples.

RESULTS

To validate the machine learning pipeline, analyses of single proteins in mouse brain sections were utilized. To demonstrate adaptability of the pipeline for multiple cell types and epitopes, the percent brain coverage of microglial cells, identified by ionized calcium binding adaptors molecule 1 (Iba1), and of astrocytes, by glial fibrillary acidic protein (GFAP) demonstrated no significant differences between automated and manual analyses protocols. Further to examine the robustness of this protocol for multiple proteins simultaneously labeling of rat brain sections were utilized; co-localization of astrocytic endfeet on blood vessels, using aquaporin-4 and tomato lectin respectively, were efficiently identified and quantified by the novel pipeline and were not significantly different between the two analyses protocols. Comparison with Existing Methods: The automated platform maintained the sensitivity and accuracy of manual analysis, while accomplishing the analyses in 1/200^th of the time.

CONCLUSIONS

We demonstrate the benefits and potential of adapting an automated high-throughput machine-learning analytical approach for the analysis ofin situ tissue samples, show effectiveness across different animal models, while reducing analysis time and increasing productivity.

Obese mice exposed to psychosocial stress display cardiac and hippocampal dysfunction associated with local brain-derived neurotrophic factor depletion

INTRODUCTION

Obesity and psychosocial stress (PS) co-exist in individuals of Western society. Nevertheless, how PS impacts cardiac and hippocampal phenotype in obese subjects is still unknown. Nor is it clear whether changes in local brain-derived neurotrophic factor (BDNF) account, at least in part, for myocardial and behavioral abnormalities in obese experiencing PS.

METHODS

In adult male WT mice, obesity was induced via a high-fat diet (HFD). The resident-intruder paradigm was superimposed to trigger PS. In vivo left ventricular (LV) performance was evaluated by echocardiography and pressure-volume loops. Behaviour was indagated by elevated plus maze (EPM) and Y-maze. LV myocardium was assayed for apoptosis, fibrosis, vessel density and oxidative stress. Hippocampus was analyzed for volume, neurogenesis, GABAergic markers and astrogliosis. Cardiac and hippocampal BDNF and TrkB levels were measured by ELISA and WB. We investigated the pathogenetic role played by BDNF signaling in additional cardiac-selective TrkB (cTrkB) KO mice.

FINDINGS

When combined, obesity and PS jeopardized LV performance, causing prominent apoptosis, fibrosis, oxidative stress and remodeling of the larger coronary branches, along with lower BDNF and TrkB levels. HFD/PS weakened LV function similarly in WT and cTrkB KO mice. The latter exhibited elevated LV ROS emission already at baseline. Obesity/PS augmented anxiety-like behaviour and impaired spatial memory. These changes were coupled to reduced hippocampal volume, neurogenesis, local BDNF and TrkB content and augmented astrogliosis.

INTERPRETATION

PS and obesity synergistically deteriorate myocardial structure and function by depleting cardiac BDNF/TrkB content, leading to augmented oxidative stress. This comorbidity triggers behavioral deficits and induces hippocampal remodeling, potentially via lower BDNF and TrkB levels. FUND: J.A. was in part supported by Rotary Foundation Global Study Scholarship. G.K. was supported by T32 National Institute of Health (NIH) training grant under award number 1T32AG058527. S.C. was funded by American Heart Association Career Development Award (19CDA34760185). G.A.R.C. was funded by NIH (K01HL133368-01). APB was funded by a Grant from the Friuli Venezia Giulia Region entitled: "Heart failure as the Alzheimer disease of the heart; therapeutic and diagnostic opportunities". M.C. was supported by PRONAT project (CNR). N.P. was funded by NIH (R01 HL136918) and by the Magic-That-Matters fund (JHU). V.L. was in part supported by institutional funds from Scuola Superiore Sant'Anna (Pisa, Italy), by the TIM-Telecom Italia (WHITE Lab, Pisa, Italy), by a research grant from Pastificio Attilio Mastromauro Granoro s.r.l. (Corato, Italy) and in part by ETHERNA project (Prog. n. 161/16, Fondazione Pisa, Italy). Funding source had no such involvement in study design, in the collection, analysis, interpretation of data, in the writing of the report; and in the decision to submit the paper for publication.

Unsupervised quantification of tissue immunofluorescence in animal models of multiple sclerosis - Instructions for use

BACKGROUND

In the analysis of animal models of CNS diseases such as experimental autoimmune encephalomyelitis (EAE), immunostaining and histopathology are important readouts. However, the complex morphological features of a tissue staining are often reduced to a single measure which relies on tedious manual planimetry. Furthermore, the measure itself and co-variables such as the region being analysed are chosen in a human decision-making process, which introduces bias.

NEW METHOD

First aim of the present study is to provide an open-source workflow for the high-throughput, unsupervised quantification of different stainings in the spinal cord. We evaluate different EAE models, spinal cord regions and different time points of disease. By applying random forest classification, we compare different measures.

RESULTS

Exemplified for glial reactivity, we show that measures and variables interact and that their values are non-normally distributed, hampering the common use of parametric tests. Furthermore, we demonstrate that one-dimensional measures are insufficient descriptors for immunofluorescence data in EAE and thus need to be considered as partly invalid.

COMPARISON WITH EXISTING METHODS

We show in a systematic analysis of EAE studies that currently published immunohistological outcomes are highly incompatible regarding methodology and statistics. Furthermore, they lack the report of important information necessary for reproducibility and do not use unsupervised automatic analysis.

CONCLUSIONS

Our results discover relevant caveats in the currently used methods of immunofluorescence analysis. The provided step-by-step instructions and open-source code are intended to serve as a framework for sensitive, unbiased immunofluorescence analysis of tissue sections in translational research.

High-mobility group box-1 translocation and release after hypoxic ischemic brain injury in neonatal rats

Inflammation contributes to neonatal brain injury. Pro-inflammatory cytokines represent key inflammatory meditators in neonatal hypoxic-ischemic (HI) brain injury. The high mobility group box-1 (HMGB1) protein is a nuclear protein with pro-inflammatory cytokine properties when it is translocated from the nucleus and released extracellularly after stroke in adult rodents. We have previously shown that HMGB1 is translocated from the nucleus to cytosolic compartment after ischemic brain injury in fetal sheep. In the current study, we utilized the Rice-Vannucci model to investigate the time course of HMGB1 translocation and release after HI injury in neonatal rats. HMGB1 was located in cellular nuclei of brains from sham control rats. Nuclear to cytoplasmic translocation of HMGB1 was detected in the ipsilateral-HI hemisphere as early as zero h after HI, and released extracellularly as early as 6 h after HI. Immunohistochemical double staining detected HMGB1 translocation mainly in neurons along with release from apoptotic cells after HI. Serum HMGB1 increased at 3 h and decreased by 24 h after HI. In addition, rat brains exposed to hypoxic injury alone also exhibited time dependent HMGB1 translocation at 3, 12 and 48 h after hypoxia. Consequently, HMGB1 responds similarly after HI injury in the brains of neonatal and adult subjects. We conclude that HMGB1 is sensitive early indicator of neonatal HI and hypoxic brain injury.

Cortical Bone Stem Cell Therapy Preserves Cardiac Structure and Function After Myocardial Infarction

RATIONALE

Cortical bone stem cells (CBSCs) have been shown to reduce ventricular remodeling and improve cardiac function in a murine myocardial infarction (MI) model. These effects were superior to other stem cell types that have been used in recent early-stage clinical trials. However, CBSC efficacy has not been tested in a preclinical large animal model using approaches that could be applied to patients.

OBJECTIVE

To determine whether post-MI transendocardial injection of allogeneic CBSCs reduces pathological structural and functional remodeling and prevents the development of heart failure in a swine MI model.

METHODS AND RESULTS

Female Göttingen swine underwent left anterior descending coronary artery occlusion, followed by reperfusion (ischemia-reperfusion MI). Animals received, in a randomized, blinded manner, 1:1 ratio, CBSCs (n=9; 2×10⁷ cells total) or placebo (vehicle; n=9) through NOGA-guided transendocardial injections. 5-ethynyl-2'deoxyuridine (EdU)-a thymidine analog-containing minipumps were inserted at the time of MI induction. At 72 hours (n=8), initial injury and cell retention were assessed. At 3 months post-MI, cardiac structure and function were evaluated by serial echocardiography and terminal invasive hemodynamics. CBSCs were present in the MI border zone and proliferating at 72 hours post-MI but had no effect on initial cardiac injury or structure. At 3 months, CBSC-treated hearts had significantly reduced scar size, smaller myocytes, and increased myocyte nuclear density. Noninvasive echocardiographic measurements showed that left ventricular volumes and ejection fraction were significantly more preserved in CBSC-treated hearts, and invasive hemodynamic measurements documented improved cardiac structure and functional reserve. The number of EdU⁺ cardiac myocytes was increased in CBSC- versus vehicle- treated animals.

CONCLUSIONS

CBSC administration into the MI border zone reduces pathological cardiac structural and functional remodeling and improves left ventricular functional reserve. These effects reduce those processes that can lead to heart failure with reduced ejection fraction.