Biology subjects

Popescu, G.

Publications and source records attributed to Popescu, G..

At least 19 recordsLinked to original sources

bioRxiv2023

Machine learning assisted health viability assay for mouse embryos with artificial confocal microscopy (ACM)

The combination of a good quality embryo and proper maternal health factors promise higher chances of a successful in vitro fertilization (IVF) procedure leading to clinical pregnancy and live birth. Of these two factors, selection of a good embryo is a controllable aspect. The current gold standard in clinical practice is visual assessment of an embryo based on its morphological appearance by trained embryologists. More recently, machine learning has been incorporated into embryo selection "packages". Here, we report a machine-learning assisted embryo health assessment tool utilizing a quantitative phase imaging technique called artificial confocal microscopy (ACM). We present a label-free nucleus detection method with novel quantitative embryo health biomarkers. Two viability assessment models are presented for grading embryos into two classes: healthy/intermediate (H/I) or sick (S) class. The models achieve a weighted F1 score of 1.0 and 0.99 respectively on the in-distribution test set of 72 fixed embryos and a weighted F1 score of 0.9 and 0.95 respectively on the out-of-distribution test dataset of 19 time-instances from 8 live embryos.

bioengineering↗

bioRxiv2023

Baldur: Bayesian hierarchical modeling for label-free proteomics exploiting gamma dependent mean-variance trends

Due to its simplicity in sample preparation, label-free quantification has become de facto in proteomics research at the expense of precision. We propose a Bayesian hierarchical decision model to test for differences in means between conditions for proteins, peptides, and post-translation modifications. We introduce a novel Bayesian regression model to characterize local mean-variance trends in the data to describe measurement uncertainty and to estimate the decision model hyperparameters. Our model vastly improves over state-of-the-art methods (Limma-Trend and t-test) in several spike-in datasets by having competitive performance in detecting true positives while showing superiority by greatly reducing false positives.

bioinformatics↗

bioRxiv2022

GANscan: continuous scanning microscopy using deep learning deblurring

Most whole slide imaging (WSI) systems today rely on the "stop-and-stare" approach, where, at each field of view, the scanning stage is brought to a complete stop before the camera snaps a picture. This procedure ensures that each image is free of motion blur, which comes at the expense of long acquisition times. In order to speed up the acquisition process, especially for large scanning areas, such as pathology slides, we developed an acquisition method in which the data is acquired continuously while the stage is moving at high speeds. Using generative adversarial networks (GANs), we demonstrate this ultra-fast imaging approach, referred to as GANscan, which restores sharp images from motion blurred videos. GANscan allows us to complete image acquisitions at 30x the throughput of stop-and-stare systems. This method is implemented on a Zeiss Axio Observer Z1 microscope in brightfield mode, requires no specialized hardware, and accomplishes successful reconstructions at stage speeds of up to 5,000 m/s. We validate the proposed method by imaging H&E stained tissue sections. Our method not only retrieves crisp images from fast, continuous scans, but also corrects any defocusing that occurs during scanning. Using a consumer GPU, the inference runs at <20ms/ image.

bioengineering↗

bioRxiv2022

Circadian volume changes in hippocampal glia studied by label-free interferometric imaging

Complex brain functions, including learning and memory, arise in part from the modulatory role of astrocytes on neuronal circuits. Functionally, the dentate gyrus (DG) exhibits differences in the acquisition of long-term potentiation (LTP) between day and night. We hypothesize that the dynamic nature of astrocyte morphology plays an important role in the functional circuitry of hippocampal learning and memory, specifically in the DG. Standard microscopy techniques, such as differential interference contrast (DIC), present insufficient contrast for detecting changes in astrocyte structure and function and are unable to inform on the intrinsic structure of the sample in a quantitative manner. Recently, gradient light interference microscopy (GLIM) has been developed to upgrade a DIC microscope with quantitative capabilities such as single-cell dry mass and volume characterization. Here, we present a methodology for combining GLIM and electrophysiology to quantify the astrocyte morphological behavior over the day-night cycle. Colocalized measurements of GLIM and fluorescence allowed us to quantify the dry masses and volumes of hundreds of astrocytes. Our results indicate that, on average, there is a 25% cell volume reduction during the nocturnal cycle. Remarkably, this cell volume change takes place at constant dry mass, which suggests that the volume regulation occurs primarily through aqueous medium exchange with the environment.

neuroscience↗

bioRxiv2021

Label-free imaging of collagen fibers in tissue slices using phase imaging with computational specificity

Evaluating the tissue collagen content in addition to the epithelial morphology has been proven to offer complementary information in histopathology, especially in oncology tumor staging and prediction of survival in cancer patients. One imaging modality widely used for this purpose is second harmonic generation microscopy (SHGM), which reports on the nonlinear susceptibility associated with the collagen fibers. Another method is polarization light microscopy (PLM) combined with picrosirius-red (PSR) tissue staining. However, SHGM requires expensive equipment and provides limited throughput, while PLM and PSR staining are not part of the routine surgical pathology workflow. Here, we utilize phase imaging with computational specificity (PICS) to computationally infer the collagen distribution of unlabeled tissue, with high specificity. PICS utilizes deep learning to translate quantitative phase images (QPI) into corresponding PSR images with high accuracy and inference speed of 200 milisecond per forwardpass through the model once trained. We developed a multimodal imaging instrument that yields both Spatial light Inference Microscopy (SLIM) and polarized light microscopy (PLM) images from the same field of view. Our results indicate that the distributions of collagen fiber orientation, length, and straightness reported by PICS closely match the ones from ground truth as defined by KL-divergence.

pathology↗

bioRxiv2021

Cell cycle stage classification using phase imaging with computational specificity

Traditional methods for cell cycle stage classification rely heavily on fluorescence microscopy to monitor nuclear dynamics. These methods inevitably face the typical phototoxicity and photobleaching limitations of fluorescence imaging. Here, we present a cell cycle detection workflow using the principle of phase imaging with computational specificity (PICS). The proposed method uses neural networks to extract cell cycle-dependent features from quantitative phase imaging (QPI) measurements directly. Our results indicate that this approach attains very good accuracy in classifying live cells into G1, S, and G2/M stages, respectively. We also demonstrate that the proposed method can be applied to study single-cell dynamics within the cell cycle as well as cell population distribution across different stages of the cell cycle. We envision that the proposed method can become a nondestructive tool to analyze cell cycle progression in fields ranging from cell biology to biopharma applications. TeaserWe present a non-destructive, high-throughput method for cell cycle detection combining label-free imaging and deep learning.

bioengineering↗

bioRxiv2021

Label-free screening of brain tissue myelin content using phase imaging with computational specificity (PICS)

Inadequate myelination in the central nervous system is associated with neurodevelopmental complications. Thus, quantitative, high spatial resolution measurements of myelin levels are highly desirable. We used spatial light interference microcopy (SLIM), a highly sensitive quantitative phase imaging (QPI) technique, to correlate the dry mass content of myelin in piglet brain tissue with dietary changes and gestational size. We combined SLIM micrographs with an AI classifying model that allows us to discern subtle disparities in myelin distributions with high accuracy. This concept of combining QPI label-free data with AI for the purpose of extracting molecular specificity has recently been introduced by our laboratory as phase imaging with computational specificity (PICS). Training on nine thousand SLIM images of piglet brain tissue with the 71-layer transfer learning model Xception, we created a two-parameter classification to differentiate gestational size and diet type with an accuracy of 82% and 80%, respectively. To our knowledge, this type of evaluation is impossible to perform by an expert pathologist or other techniques.

bioinformatics↗

bioRxiv2020

Monitoring Reactivation of Latent HIV by Label-Free Gradient Light Interference Microscopy

Latent human immunodeficiency virus (HIV) reservoirs in infected individuals present the largest barrier to a cure. The first step towards overcoming this challenge is to understand the science behind latency-reactivation interplay. Fluorescence imaging of GFP-tagged HIV has been the main method for studying reactivation of latent HIV in individually infected cells. In this paper, we report insights provided by label-free, gradient light interference microscopy (GLIM) about the changes in measures including dry mass, diameter, and dry mass density associated with infected cells that occur upon reactivation. We discovered that mean cell dry mass and mean diameter of latently infected cells treated with reactivating drug, TNF-, are higher for cells with reactivated HIV as compared to those with latent disease. Results also indicate that cells with mean dry mass and diameter less than 10pg and 8{micro}m, respectively, remain exclusively in the latent state. Also, cells with mean dry mass greater than 23pg and mean diameter greater than 11{micro}m have a higher probability of reactivating. This study is significant as it presents a new label-free approach to quantify latent reactivation of a virus in single cells based on changes in cell morphology.

bioengineering↗

bioRxiv2020

Rapid SARS-CoV-2 Detection and Classification Using Phase Imaging with Computational Specificity

Efforts to mitigate the COVID-19 crisis revealed that fast, accurate, and scalable testing is crucial for curbing the current impact and that of future pandemics. We propose an optical method for directly imaging unlabeled viral particles and using deep learning for detection and classification. An ultrasensitive interferometric method was used to image four virus types with nanoscale optical pathlength sensitivity. Pairing these data with fluorescence images for ground truth, we trained semantic segmentation models based on U-Net, a particular type of convolutional neural network. The trained network was applied to classify the viruses from the interferometric images only, containing simultaneously SARS-CoV-2, H1N1 (influenza-A), HAdV (adenovirus), and ZIKV (Zika). Remarkably, due to the nanoscale sensitivity in the input data, the neural network was able to identify SARS-CoV-2 vs. the other viruses with 96% accuracy. The inference time for each image is 60 ms, on a common graphic processing unit. This approach of directly imaging unlabeled viral particles may provide an extremely fast test, of less than a minute per patient. As the imaging instrument operates on regular glass slides, we envision this method as potentially testing on patient breath condensates. The necessary high throughput can be achieved by translating concepts from digital pathology, where a microscope can scan hundreds of slides automatically. One Sentence SummaryThis work proposes a rapid (<1 min.), label-free testing method for SARS-CoV-2 detection, using quantitative phase imaging and deep learning.

bioengineering↗

bioRxiv2020

A 'Cell-Nucleus Segmentation' Script for Non-Invasive Nuclear Dry Mass Measurement

The nucleus is the largest organelle in cells carrying genetic materials that support genetic replication and transcription. It is likely that such genetic activities can influence the nuclear dry mass, but there is a lack of analytical tools enabling us to monitor dynamic changes in this quantity. To this end, this study demonstrates an image analysis script that allows us to quantify these changes in the nuclear dry mass. The script runs the cell-nuclei segmentation using Matlab. By using the fluorescent image as a template for the boundaries of cell nuclei and quantitative phase images for retrieving the dry mass density, the script recognizes nuclei of all cells in an image at a time and quantifies the nuclear dry mass. Using the "the cell-nucleus segmentation" script, this study reveals an interesting correlation between the nuclear dry mass and the filopodia protrusion of cervical epithelial cells. As the filopodia density and protrusion length increase, the nuclear dry mass increases. On the other hand, whenever the nuclear dry mass decreases, cells filopodia retract significantly. Taken together, the imaging script developed here will be useful to quantifying dynamic nuclear activities of a broad array of cells non-invasively.

bioengineering↗

bioRxiv2020

Label-free cell viability assay using phase imaging with computational specificity

Existing approaches to evaluate cell viability involve cell staining with chemical reagents. However, this step of exogenous staining makes these methods undesirable for rapid, nondestructive and long-term investigation. Here, we present instantaneous viability assessment of unlabeled cells using phase imaging with computation specificity (PICS). This new concept utilizes deep learning techniques to compute viability markers associated with the specimen measured by label-free quantitative phase imaging. Demonstrated on different live cell cultures, the proposed method reports approximately 95% accuracy in identifying live and dead cells. The evolution of the cell dry mass and projected area for the labelled and unlabeled populations reveal that the viability reagents decrease viability. The nondestructive approach presented here may find a broad range of applications, from monitoring the production of biopharmaceuticals, to assessing the effectiveness of cancer treatments.

bioengineering↗

bioRxiv2020

Quantifying myelin content in brain tissue using color spatial light interference microscopy (cSLIM)

Deficient myelination of the brain is associated with neurodevelopmental delays, particularly in high-risk infants, such as those born small in relation to their gestational age (SGA). New methods are needed to further study this condition. Here, we employ Color Spatial Light Interference Microscopy (cSLIM), which uses a brightfield objective and RGB camera to generate pathlength-maps with nanoscale sensitivity in conjunction with a regular brightfield image. Using tissue sections stained with Luxol Fast Blue, the myelin structures were segmented from a brightfield image. Using a binary mask, those portions were quantitatively analyzed in the corresponding phase maps. We first used the CLARITY method to remove tissue lipids and validate the sensitivity of cSLIM to lipid content. We then applied cSLIM to brain histology slices. These specimens are from a previous MRI study, which demonstrated that appropriate for gestational age (AGA) piglets have increased internal capsule myelination (ICM) compared to small for gestational age (SGA) piglets and that a hydrolyzed fat diet improved ICM in both. The identity of samples was blinded until after statistical analyses.

neuroscience↗

bioRxiv2019

Patterned substrates modulate growth and dynamics of 3D cellular systems

The development of 3D cellular architectures during development and pathological processes involves intricate migratory patterns that are modulated by genetics and the surrounding microenvironment. The substrate composition of cell cultures has been demonstrated to influence growth, proliferation, and migration in 2D. Here we study the growth and dynamics of mouse embryonic fibroblast (MEF) cultures patterned in a tissue sheet which then exhibits 3D growth. Using gradient light interference microscopy (GLIM), a label-free quantitative phase imaging approach, we explored the influence of geometry on cell growth patterns and rotational dynamics. We apply, for the first time to our knowledge, dispersion-relation phase spectroscopy (DPS) in polar coordinates to generate the radial and rotational cell mass-transport. Our data show that cells cultured on engineered substrates undergo rotational transport in a radially independent manner and exhibit faster vertical growth than the control, unpatterned cells. The use of GLIM and polar DPS provides a novel quantitative approach to studying the effects of spatially patterned substrates on cell motility and growth.

cell biology↗

bioRxiv2018

Tissue spatial correlation as cancer marker

We propose a new intrinsic cancer marker in fixed tissue biopsy slides, which is based on the local spatial autocorrelation length obtained from quantitative phase images. The spatial autocorrelation length in a small region of the tissue phase image is sensitive to the nanoscale cellular morphological alterations and can hence inform on carcinogenesis. Therefore, this metric can potentially be used as an intrinsic cancer marker in histopathology. Typically, these correlation length maps are calculated by computing 2D Fourier transforms over image sub-regions - requiring long computational times. In this paper, we propose a more time efficient method of computing the correlation map and demonstrate its value for diagnosis of benign and malignant breast tissues. Our methodology is based on highly sensitive quantitative phase imaging data obtained by spatial light interference microscopy (SLIM).

pathology↗

bioRxiv2017

Tissue disorder strength measured by quantitative phase imaging as intrinsic cancer marker

Tissue refractive index provides important information about morphology at the nanoscale. Since the malignant transformation involves both intra- and inter-cellular changes in the refractive index map, the tissue disorder measurement can be used to extract important diagnosis information. Quantitative phase imaging (QPI) provides a practical means of extracting this information as it maps the optical path-length difference (OPD) across a tissue sample with sub-wavelength sensitivity. In this work, we employ QPI to compare the tissue disorder strength between benign and malignant breast tissue histology samples. Our results show that disease progression is marked by a significant increase in the disorder strength. Since our imaging system can be added as an upgrading module to an existing microscope, we anticipate that it can be integrated easily in the pathology work flow.

bioengineering↗

bioRxiv2017

Real-time halo correction in phase contrast imaging

As a label-free, nondestructive method, phase contrast is by far the most popular microscopy technique for routine inspection of cell cultures. Yet, features of interest such as extensions near cell bodies are often obscured by a glow, which came to be known as the halo. Advances in modeling image formation have shown that this artifact is due to the limited spatial coherence of the illumination. Yet, the same incoherent illumination is responsible for superior sensitivity to fine details in the phase contrast geometry. Thus, there exists a trade-off between high-detail (incoherent) and low-detail (coherent) imaging systems. In this work, we propose a method to break this dichotomy, by carefully mixing corrected low-frequency and high-frequency data in a way that eliminates the edge effect. Specifically, our technique is able to remove halo artifacts at video rates, requiring no manual interaction or a priori point spread function measurements. To validate our approach, we imaged standard spherical beads, sperm cells, tissue slices, and red blood cells. We demonstrate the real-time operation with a time evolution study of adherent neuron cultures whose neurites are revealed by our halo correction. We show that with our novel technique, we can quantify cell growth in large populations, without the need for thresholds and calibration.

bioengineering↗

bioRxiv2017

Topography and refractometry of sperm cells using SLIM

Characterization of spermatozoon viability is a common test in treating infertility. Recently, it has been shown that label-free, phase-sensitive imaging can provide a valuable alternative for this type of assay. Here, we employ spatial light interference microscopy (SLIM) to decouple the thickness and refractive index information of individual cells. This procedure was enabled by quantitative phase imaging cells on media of two different refractive indices and using a numerical tool to remove the curvature from the cell tails. This way, we achieved ensemble averaging of topography and refractometry of 100 cells in each of the two groups. The results show that the thickness profile of the cell tail goes down to 150 nm and the refractive index can reach values of 1.6 close to the head.

cell biology↗

bioRxiv2017

Label-free quantitative screening of breast tissue using Spatial Light Interference Microscopy (SLIM)

Breast cancer is the most common type of cancer among women worldwide. The standard histopathology of breast tissue, the primary means of disease diagnosis, involves manual microscopic examination of stained tissue by a pathologist. Because this method relies on qualitative information, it can result in inter-observer variation. Furthermore, for difficult cases the pathologist often needs additional markers of malignancy to help in making a diagnosis. We present a quantitative method for label-free tissue screening using Spatial Light Interference Microscopy (SLIM). By extracting tissue markers of malignancy based on the nanostructure revealed by the optical path-length, our method provides an objective and potentially automatable method for rapidly flagging suspicious tissue. We demonstrated our method by imaging a tissue microarray comprising 68 different subjects - 34 with malignant and 34 with benign tissues. Three-fold cross validation results showed a sensitivity of 94% and specificity of 85% for detecting cancer. The quantitative biomarkers we extract provide a repeatable and objective basis for determining malignancy. Thus, these disease signatures can be automatically classified through machine learning packages, since our images do not vary from scan to scan or instrument to instrument, i.e., they represent intrinsic physical attributes of the sample, independent of staining quality.

pathology↗