Quantitative microscopy deals with the extraction of quantitative measurements from samples observed under a microscope. Recent developments in microscopy systems, sample preparation and handling techniques have enabled high throughput biological experiments resulting in large amounts of image data, at biological scales ranging from subcellular structures such as fluorescently tagged nucleic acid sequences to whole organisms such as zebrafish embryos. Consequently, methods and algorithms for automated quantitative analysis of these images have become increasingly important. These methods range from traditional image analysis techniques to use of deep learning architectures. Many biomedical microscopy assays result in fluorescent spots. Robust detection and precise localization of these spots are two important, albeit sometimes overlapping, areas for application of quantitative image analysis. We demonstrate the use of popular deep learning architectures for spot detection and compare them against more traditional parametric model-based approaches. Moreover, we quantify the effect of pre-training and change in the size of training sets on detection performance. Thereafter, we determine the potential of training deep networks on synthetic and semi-synthetic datasets and their comparison with networks trained on manually annotated real data. In addition, we present a two-alternative forced-choice based tool for assisting in manual annotation of real image data. On a spot localization track, we parallelize a popular compressed sensing based localization method and evaluate its performance in conjunction with different optimizers, noise conditions and spot densities. We investigate its sensitivity to different point spread function estimates. Zebrafish is an important model organism, attractive for whole-organism image-based assays for drug discovery campaigns. The effect of drug-induced neuronal damage may be expressed in the form of zebrafish shape deformation. First, we present an automated method for accurate quantification of tail deformations in multi-fish micro-plate wells using image analysis techniques such as illumination correction, segmentation, generation of branch-free skeletons of partial tail-segments and their fusion to generate complete tails. Later, we demonstrate the use of a deep learning-based pipeline for classifying micro-plate wells as either drug-affected or negative controls, resulting in competitive performance, and compare the performance from deep learning against that from traditional image analysis approaches.

Electroporation-based treatments are motivated by the response of biological membranes to high- intensity pulsed electric fields. These fields rearrange the membrane structure to enhance the membrane's diffusive permeability, or the degree to which a membrane allows molecules to diffuse through it, is impacted by the structure, composition, and environment in which the cell resides. Tracer molecules have been developed that are unable to pass through intact cell membranes yet enter permeabilized cells. This dissertation investigates the hypothesis that the flow of such molecules may be used to quantify the effects of the electrical stimulus and environmental conditions leading to membrane electroporation. Specifically, a series of electrical pulses that alternates between positive and negative pulses permeabilizes cells more symmetrically than a longer pulse with the same total on-time. However, the magnitude of this symmetric entry decreases for the shorter alternating pulses. Furthermore, a method for quantitatively measuring the permeability of the cell membrane was proposed and validated. From data near the electroporation threshold, the response of cells varies widely in the manner in which cells become permeabilized. This method is applied to study the transient cell membrane permeability induced by electroporation and is used to demonstrate that the cell membrane remains permeable beyond 30 min following treatment. To analyze these experimental findings in the context of physical mechanisms, computational models of molecular uptake were developed to simulate electroporation. The results of these simulations indicate that the cell's local environment during electroporation facilitates the degree of molecular uptake. We use these models to predict how manipulating both the environment of cells during electroporation affects the induced membrane permeability. These experimental and computational results provide evidence that supports the hypothesis of this dissertation and provide a foundation for future investigation and simulation of membrane electroporation. / PHD / Electroporation is a biophysical process in which intense electric fields permeabilize bilayer membranes. The degree to which a membrane allows molecules to diffuse through it is called its diffusive permeability, and is impacted by the structure, composition, and environment in which the cell resides. This dissertation investigates the hypothesis that the flow of molecules into cells through their membranes may be used to quantitatively study the effects of the electrical stimulus and environmental conditions leading to membrane disruption. Here, I demonstrate that the cellular response to pulsed electric fields is affected by the waveform of the applied electrical stimulus. Specifically, a series of electrical pulses that alternates between positive and negative pulses permeabilizes cells more symmetrically than a longer pulse with the same total energized time. However, the total molecular uptake decreases for the shorter alternating pulses over the longer pulse. A method for quantitatively measuring the permeability of the cell membrane using a fluorescent tracer molecule is also developed and validated. This method is applied to show how cell membrane permeability changes following electroporation. To analyze these findings, computational models of molecular flow through the cell membrane are developed. These simulations indicate that the cell’s surrounding environment during electroporation dramatically impacts the degree of molecular uptake. We use these models to predict how manipulating both the environment of cells during electroporation affects the induced membrane permeability. These experimental and computational results provide a foundation for future investigation and simulation of membrane electroporation.

An understanding of the damage evolution prior to crack initiation in advanced structural materials is of vital importance to the fatigue community in both academia and industry. Features known as the Persistent Slip Bands (PSBs) play an integral role in this damage evolution. Therefore, PSBs have been the focus of a lot of science-based investigations over the years. However, most existing studies in this area are restricted to analysis of PSBs in single crystal face centered cubic (FCC) materials. Moreover, these studies lack a quantitative analysis of the evolution of the fatigue damage (or PSBs) as a function of the material microstructure. This is especially true for relatively modern materials such as the Advanced High Strength Structural (AHSS) steels that are gaining a lot of importance in the automotive sector. Accordingly, the objective of this research is to quantitatively characterize evolution of PSBs in three AHSS steels having different microstructures as a function of number of fatigue cycles and strain amplitude. For this purpose strain controlled interrupted fatigue tests have been performed on two dual phase steels (DP-590 and DP-980) having different relative amounts of tempered martensite and a ferritic high strength low alloy steel (HR-590). Digital image analysis and Stereology have been used for unbiased quantitative characterization of the evolution of global geometry of the PSB colonies as function of number of fatigue cycles and strain amplitude. Evolution of PSB colonies has been couched in terms of variation of PSB colony volume fraction and total surface area unit volume, and total surface area of individual PSBs per unit volume and three-dimensional angular orientation distribution of the PSBs. For this purpose, new stereological techniques have been developed for estimation of the three-dimensional angular orientation distribution. The stereological data reveal that during strain controlled in these AHSS steels, volume fraction of the PSB colonies varies linearly with the their total surface area per unit volume. Detailed analysis of the stereological data leads to a simple geometric model for evolution of the PSB colonies in the three AHSS steels, which accounts for all observed data trends.

La protéine DnaA, ou un homologue, est présente dans la plupart des organismes vivants parce qu'elle joue un rôle clé pour la réplication de l'ADN. Dans Escherichia coli, l'expression de DnaA, l'initiateur central de la réplication de l'ADN, est donc étroitement régulée. Des études antérieures ont montré qu'une forte surexpression de cette protéine conduit à une diminution de la viabilité cellulaire, alors que son absence induit l'arrêt de la division cellulaire. Même si ces conditions extrêmes sont bien étudiées, la transition entre l'arrêt de la division et la croissance normale n'a pas été analysée quantitativement.Nous avons modifié génétiquement Escherichia coli pour mettre l'expression de l'ARN polymérase et de DnaA sous le contrôle deux systèmes inductibles distincts. Pour contrôler l’expression de DnaA, nous avons utilisé un système d’induction se trouvant déjà dans la cellule, le système uhp. Le promoteur du gène uhpT est induit par le glucose-6-phosphate extracellulaire. Nous avons tout d’abord étudié les caractéristiques d’induction de ce système et ensuite caractérisé les phénomènes biologiques déclenchées par les variations de la concentration de DnaA. Les méthodes utilisées combinent des mesures sur une population des bactéries, avec celles en cellule unique, en utilisant la microscopie in vivo en temps réel et des systèmes microfluidiques. Les expériences de microscopie révèlent des phénomènes stochastiques en raison du faible nombre de molécules des composants du système d'induction uhp. En corrélant les observations de population et de cellules uniques, nous donnons une interprétation quantitative du comportement observé. Comme une application potentielle de notre système de contrôle, nous envisageons la possibilité d’arrêter la division cellulaire afin de transformer la cellule en un «sac d'enzymes» pour la production biotechnologique de métabolites. / The DnaA protein, or a homologue, is present in most living organisms because it plays a key role for DNA replication. In Escherichia coli, the expression of DnaA, the central initiator of DNA replication, is therefore tightly regulated. Previous studies have shown that a large overexpression of this protein leads to a decrease in cell viability, while its absence induces the arrest of cell division. Even though these extreme conditions are well studied, the transition from division arrest to normal growth has not been quantitatively analyzed.We genetically engineered Escherichia coli to put the expression of RNA polymerase and the expression of DnaA under the control of two distinct, inducible systems. For the control of DnaA expression, we used a regulatory system already present in the cell, the uhp system. The promoter of the uhpT gene is induced via extracellular glucose-6-phosphate. We characterized the induction characteristics of this system and studied the biological phenomena triggered by varying concentrations of DnaA, using population measurements and single cell, time lapse microscopy of microcolonies or cells grown in a microfluidics device. The microscopy experiments reveal stochastic phenomena due to the low number of molecules of components of the induction system and of DnaA. Confronting population and single cell observations we are able to give a quantitative interpretation of the observed behavior. As a potential application of our control system, we explored the possibility of freezing cell division in order to turn the cell into a “bag of enzymes” for the biotechnological production of metabolites.

Understanding biology paves the way for discovering drugs targeting deadly diseases like cancer, and microscopy imaging is one of the most informative ways to study biology. However, analysis of large numbers of samples is often required to draw statistically verifiable conclusions. Automated approaches for analysis of microscopy image data makes it possible to handle large data sets, and at the same time reduce the risk of bias. Quantitative microscopy refers to computational methods for extracting measurements from microscopy images, enabling detection and comparison of subtle changes in morphology or behavior induced by varying experimental conditions. This thesis covers computational methods for segmentation and classification of biological samples imaged by microscopy. Recent increase in computational power has enabled the development of deep neural networks (DNNs) that perform well in solving real world problems. This thesis compares classical image analysis algorithms for segmentation of bacteria cells and introduces a novel method that combines classical image analysis and DNNs for improved cell segmentation and detection of rare phenotypes. This thesis also demonstrates a novel DNN for segmentation of clusters of cells (spheroid), with varying sizes, shapes and textures imaged by phase contrast microscopy. DNNs typically require large amounts of training data. This problem is addressed by proposing an automated approach for creating ground truths by utilizing multiple imaging modalities and classical image analysis. The resulting DNNs are applied to segment unstained cells from bright field microscopy images. In DNNs, it is often difficult to understand what image features have the largest influence on the final classification results. This is addressed in an experiment where DNNs are applied to classify zebrafish embryos based on phenotypic changes induced by drug treatment. The response of the trained DNN is tested by ablation studies, which revealed that the networks do not necessarily learn the features most obvious at visual examination. Finally, DNNs are explored for classification of cervical and oral cell samples collected for cancer screening. Initial results show that the DNNs can respond to very subtle malignancy associated changes. All the presented methods are developed using open-source tools and validated on real microscopy images.

No description available.

No description available.

The methods of analysis with a focused ion beam, commonly termed as nuclear microscopy, include quantitative physical processes like PIXE and RBS. The element concentrations in a sample can be quantitatively mapped with a sub-micron spatial resolution and a sub-ppm sensitivity. Its fully quantitative and non-destructive nature makes it particularly suitable for analysing biological samples. The applications in biomedical research are manifold. The iron overload hypothesis in Parkinson\\\'s disease is investigated by a differential analysis of human substantia nigra. The trace element content is quantified in neuromelanin, in microglia cells, and in extraneuronal environment. A comparison of six Parkinsonian cases with six control cases revealed no significant elevation in iron level bound to neuromelanin. In fact, a decrease in the Fe/S ratio of Parkinsonian neuromelanin was measured, suggesting a modification in its iron binding properties. Drosophila melanogaster, or the fruit fly, is a widely used model organism in neurobiological experiments. The electrolyte elements are quantified in various organs associated with the olfactory signalling, namely the brain, the antenna and its sensilla hairs, the mouth parts, and the compound eye. The determination of spatially resolved element concentrations is useful in preparing the organ specific Ringer\\\'s solution, an artificial lymph that is used in disruptive neurobiological experiments. The role of trace elements in the progression of atherosclerosis is examined in a pilot study. A differential quantification of the element content in an induced murine atherosclerotic lesion reveals elevated S and Ca levels in the artery wall adjacent to the lesion and an increase in iron in the lesion. The 3D quantitative distribution of elements is reconstructed by means of stacking the 2D quantitative maps of consecutive sections of an artery. The feasibility of generating a quantitative elemental rodent brain atlas by Large Area Mapping is investigated by measuring at high beam currents. A whole coronal section of the rat brain was measured in segments in 14 h. Individual quantitative maps of the segments are pieced together to reconstruct a high-definition element distribution map of the whole section with a subcellular spatial resolution. The use of immunohistochemical staining enhanced with single elements helps in determining the cell specific element content. Its concurrent use with Large Area Mapping can give cellular element distribution maps.

The methods of analysis with a focused ion beam, commonly termed as nuclear microscopy, include quantitative physical processes like PIXE and RBS. The element concentrations in a sample can be quantitatively mapped with a sub-micron spatial resolution and a sub-ppm sensitivity. Its fully quantitative and non-destructive nature makes it particularly suitable for analysing biological samples. The applications in biomedical research are manifold. The iron overload hypothesis in Parkinson\\\''s disease is investigated by a differential analysis of human substantia nigra. The trace element content is quantified in neuromelanin, in microglia cells, and in extraneuronal environment. A comparison of six Parkinsonian cases with six control cases revealed no significant elevation in iron level bound to neuromelanin. In fact, a decrease in the Fe/S ratio of Parkinsonian neuromelanin was measured, suggesting a modification in its iron binding properties. Drosophila melanogaster, or the fruit fly, is a widely used model organism in neurobiological experiments. The electrolyte elements are quantified in various organs associated with the olfactory signalling, namely the brain, the antenna and its sensilla hairs, the mouth parts, and the compound eye. The determination of spatially resolved element concentrations is useful in preparing the organ specific Ringer\\\''s solution, an artificial lymph that is used in disruptive neurobiological experiments. The role of trace elements in the progression of atherosclerosis is examined in a pilot study. A differential quantification of the element content in an induced murine atherosclerotic lesion reveals elevated S and Ca levels in the artery wall adjacent to the lesion and an increase in iron in the lesion. The 3D quantitative distribution of elements is reconstructed by means of stacking the 2D quantitative maps of consecutive sections of an artery. The feasibility of generating a quantitative elemental rodent brain atlas by Large Area Mapping is investigated by measuring at high beam currents. A whole coronal section of the rat brain was measured in segments in 14 h. Individual quantitative maps of the segments are pieced together to reconstruct a high-definition element distribution map of the whole section with a subcellular spatial resolution. The use of immunohistochemical staining enhanced with single elements helps in determining the cell specific element content. Its concurrent use with Large Area Mapping can give cellular element distribution maps.

No description available.