This PDF file contains the front matter associated with SPIE Proceedings Volume 11647, including the Title Page, Copyright information, and Table of Contents.

From multi-photon to single molecule, the past several decades have witnessed a revolution in fluorescent microscopy. These techniques have revealed the inner working of cells and tissue and have relied on symbiotic advances in advanced molecular probes, light emitting molecules and particles, and novel instrumentation. More recently, researchers have begun to develop functional nanomaterials or materials that can response to their environment. In this talk, I will discuss some of our recent work in developing functional imaging agents for multi-wavelength and multi-photon live-cell imaging, focusing on recent molecular designs performed using density functional theory as well as in vitro studies.

Second harmonic generation (SHG) microscopy is a second-order nonlinear optical scattering process that can nondestructively image myosin filament microarchitecture up to 500 μm in depth due to its high spatial resolution, optical sectioning capabilities, and elastic scattering properties. The distribution, alignment, and spatial orientation of myosin filaments largely dictates the mechanical properties of skeletal muscle. Noninvasive quantification of these metrics in-vivo or in-vitro could be indicative of skeletal muscle function and viability. Here, we present a novel algorithm to quantitate sarcomere length with high spatial resolution and accuracy when compared to traditional approaches. Our approach runs in real-time and is less sensitive to instrumental noise and artifacts that shade regions of the image. We have tested this technique on ex-vivo SHG images of rat lumbrical muscles that have been damaged following standardized lengthening contraction protocols. Our approach shows accurate quantitation of sarcomere lengths within 1% compared to manual calculations. This algorithm can be used to reliably analyze SHG images in realtime to assess the structure and function of skeletal muscle tissue in a non-invasive, quantitative, and sterile manner.

Engineered brain tissue models with human derived cells are a promising platform for improving our understanding of brain function. Our study aims to develop a label-free, two-photon imaging focused approach that enables us to assess important morphological and functional changes that occur in such brain tissue models over time. We acquired spectral, intensity, and lifetime images of the same tissues over two months. Our results indicate that such dynamic monitoring of the cellular and matrix/scaffold components of such tissues is feasible, but complex because multiple fluorophores are present. Thus, a multi-modal, multi-wavelength approach is necessary to quantify meaningful functional changes.

Newts have the exceptional capability of regenerating the lens through their lifetime. The transparency of the anterior chamber makes OCT an idea imaging technology to track the entire process of the lens regeneration in vivo without interruptions. We demonstrated, for the first time, that OCT can capture not only essential morphological changes similar to the changes observed in histology but some fine structures, like zonular fibers, which are not visible in histology. Our initial results warrant the future research of tailoring OCT for dynamically imaging the lens regeneration in newts.

Reduced nicotinamide adenine dinucleotide (NADH) and oxidized flavin adenine dinucleotide (FAD) are coenzymes of cellular metabolism reactions, and their endogenous fluorescent signals are used to evaluate cell redox state and detect changes in cellular metabolism. Different cellular metabolic states can alter NADH and FAD fluorescence features. Here, a model is developed to determine T cell metabolic pathway utilization from autofluorescence lifetime imaging features. The model is trained and tested using cellular features extracted from NADH and FAD fluorescence lifetime images of activated and quiescent T cells with chemical inhibition of glycolysis, oxidative phosphorylation, glutaminolysis, and fatty acid synthesis. Feature analysis revealed the optical redox ratio (FAD intensity/ (NADH intensity + FAD intensity), the fluorescence lifetime redox ratio (fraction of bound NADH/fraction of bound FAD), and the fluorescence lifetime of free NADH are the highest weighed features for classification of T cells dependent on glycolysis versus oxidative metabolism. High classification accuracy is achieved for discrimination between quiescent and activated T cells, and modest classification accuracy is achieved for classification of T cells into metabolic subgroups. Autofluorescence features vary between cytoplasm and mitochondria and analyzing this difference can provide additional metabolic information. Altogether, these results demonstrate the potential for autofluorescence lifetime imaging features to classify T cell function and metabolic state.

3D models based on cells differentiated from patient-specific induced pluripotent stem cells (iPSCs) are widely used to identify disease phenotypes, to accurately analyze dysfunctions at the level of human tissues and organs, to screen new drugs, which makes them more promising tool for biomedical research tasks than monolayer cultures, which is associated with their proximity to in vivo. The metabolic activity with oxygenation level of cells, assessed by optical imaging methods, can be used as markers of cell viability, proliferative activity and the degree of differentiation in 3D culture conditions. In this paper we used fluorescence and phosphorescence lifetime imaging microscopies (FLIM and PLIM) to study the metabolic status and the oxygenation level of derived from iPSCs neural stem cells (NSC) cultured in 3D condition. An analysis of the fluorescence intensities and FLIM data showed that NSCs in monolayer and at the periphery of large spheroids have more glycolytic phenotypes, NSCs in the center of large spheroids and NSCs grouped into small spheroids have more oxidative state. For determination of the relative oxygen level in spheroids PLIM of BPTDM stained neurospheres was carried out. As it was supposed, oxygen transport in the spheroid depended on it size. In neurospheres with an average size 600 μm O2 distribution is radial, with the lowest concentration in the center. Thus, the metabolic status and oxygenation level of the NSC in the spheroid composition was assessed in a life-time and noninvasive manner.

Extracellular vesicles (EVs) are plasma-membrane formed particles released by cells, and range in diameter from 50 to 2000 nm. Interest in EVs is growing, and recent work has aimed to employ nonlinear optical microscopy techniques to better characterize the size, function, and biochemical makeup of EVs. Previous studies have shown that EVs can modulate gene expression and metabolism of cells that uptake them. Here, we use fluorescence lifetime imaging microscopy (FLIM) of reduced nicotinamide adenine dinucleotide and reduced nicotinamide adenine dinucleotide phosphate (NAD(P)H) to monitor the metabolic response of macrophages and other cells to native and foreign EVs and other known cellular activators.

Being the largest organelle in a cell, the nucleus houses the genetic code of an organism and primarily serves as the cell’s information center. Proper implementation and consolidation of physiological stress are intrinsic for proper growth and development. Studies reveal that in animals, under chronic stress, the nuclear size shows morphological plasticity. This paper proposes the chronic stress study on plant specimens, namely, Allium cepa root, upon induction on Rhodamine B fluorophore, using our in-house developed dual arm multi-level magnification light sheet microscopy (DMx-LSFM) system. Employing our home-built microscope, we can analyze the difference in girth of the nuclear membrane under different magnification levels simultaneously, which helps to inspect the minute details and structural changes occurring in the organelle. The study was then further extended to investigate the role of exposure time and concentration of the stress agent on the sample.

Photoacoustic imaging (PAI) has emerged as a powerful modality capable of providing accurate and quantitative 3D molecular and functional tissue information with high spatiotemporal resolution. However, widespread use of molecular PAI is severely limited by availability of validated and clinically translatable exogenous contrast agents. Here we will first present a set of requirements and specifications that need to be addressed in order to achieve a clinically translatable solution for molecular PAI. Then we will present a new class of molecular contrast agents for PAI that are based on antibody-targeted, liposome-encapsulated dye-aggregates. Through the use of tissue-mimicking phantoms, in vitro, and in vivo orthotopic ovarian cancer model, we will show that these agents, which are composed of all FDA approved components, can potentially address clinical needs for a photoacoustic (PA) contrast agent.

Cerebral microhemorrhages (CMHs) are associated with cognitive impairment and several conditions, diseases, and normal aging processes. Current histological methods manually identify and quantify Prussian blue-stained CMHs, which can take months to complete. To speed up this labor-intensive process, we developed a spectroscopic, semi-automated approach. We used the ratio of the red and green intensities relative to the blue intensity squared to discriminate CMH-pixels from background pixels. We calculated a sensitivity and specificity of 83.75% and 99.74%, respectively. The intraclass correlation coefficient was 0.992 (95% confidence interval: 0.989-0.995). Future studies are needed to test if this approach works in other CMH models.

Dual-clad fiber couplers allow combining coherent imaging and diffuse sensing within the same instrument. Since the technology's lab-to-market transfer, several groups reported on its application to OCT, confocal microscopy, and hyperspectral imaging, to name a few. From this initial design, our group has focused on other dedicated optical fiber couplers for biomedical imaging, namely: multimode circulators, bi-directional couplers, and modally-specific photonic lanterns. This presentation will review the application of these recent innovations in the context of biomedical imaging and sensing.

We present the first image sensor that, equipped with a matrix of 64x64 SPADs and a 4-layers distributed artificial neural network (ANN), can be ”reprogrammed” (by loading its weights) at run-time to perform any required task without redesigning the complete hardware. In general, SPAD sensors are specifically designed for only one type of application because systems that are multi-purpose usually have to face issues that are originated from the trade-off performance needed for each of those aimed applications. The next logical step is to design an image sensor that is application-independent. With the help of well-known high-level processing capabilities of ANNs, the system was designed to deliver qualitative information in micro seconds, such as ”car in front” or ”the vehicle is off the road” or ”sign ahead”. The system was synthesized for a high-speed clock of 100 MHz. The total processing time needed by a neuron is 320 ns. Considering that the neural network has three processing layers, the total time it takes to generate its outputs is 960 ns. However, the network is a pipelined system, meaning the throughput of information is that of the neuron (320 ns) with a latency of 960 ns. This prompt reaction, for example, could be used for driving assistance. We show the results of post-layout simulations for two different applications utilizing the same hardware model to prove this concept: Optical Character Recognition (OCR) and signal digitization. The recognition performance achieved is 96.47 %; it only lowers at 85 % with a given Signal-to-Noise Ratio (SNR) of 2

Hyper-Raman spectroscopy is a nonlinear optical variant of vibrational spectroscopy to acquire information about molecular structures. Hyper-Raman spectroscopy complements existing infrared and Raman spectroscopy techniques due to differences in the selection rules. Performing hyper-Raman spectroscopy at 532 nm results in the signal emission in the UV spectral range (266 nm – 296 nm), which benefits from near-resonant conditions for many biomolecules. Even operating in the electronic resonant enhancement regime, hyper-Raman spectroscopy requires high average and peak power, picosecond laser systems to achieve reasonable collection times (1 minute – 30 minutes). In this report, we explore applications of hyper-Raman spectroscopy to aromatic structures (L-phenylalanine and imidazole) that experience significant two-photon absorption and two-photon fluorescence which can obstruct measurements of the hyper-Raman spectra of these molecules. Since competing two-photon processes could significantly limit future UV hyper-Raman applications, we explore mitigating strategies to circumvent the fluorescence background of Lphenylalanine and imidazole by applying a quenching agent (hydrogen peroxide). We also outline a more general solution to alleviate two-photon absorption and fluorescence by proposing tailored laser configurations where the excitation wavelength could be tuned to avoid two-photon absorption resonances while remaining in the UV regime.

There is a demand for a robust and flexible illumination and screening instrument for preclinical light-sensitive drug development and optogenetic research. While there is a great selection of different types of commercial plate readers available on the market, these instruments do not provide enough versatility for high-throughput illumination experiments. In addition, plate readers typically utilize xenon flash lamps or LEDs for sample analysis, which have wider spectral output and lower excitation powers compared to lasers. To answer this unmet need, we have developed an automated, laser-based well-plate illuminator, the ML8500. It enables flexible setup of illumination parameters like wavelength, light irradiance and fluence well-by-well within a single experiment. The fluorescence monitoring possibility expands the applicability beyond sample illumination to support various fluorescence applications. The built-in incubator minimizes unspecific cellular stress and ensures consistent data even during long measurement cycles. The system is also Cloud-connected, supporting data collection and analysis, and enabling machine learning and AI based biomedical research in the future. The ML8500 can be a useful tool for many biomedical fields such as optogenetics, where the activation of light-responsive opsins and simultaneous fluorescence monitoring of sensor proteins enables spatiotemporally controlled, all-optical electrophysiology. Independent of the field of use, the ML8500 can reduce the cost of experimental labor while increasing the reproducibility and data throughput of experiments. In this presentation, we describe the key features of the ML8500, how it is operated, proof-of-concept testing results as well as present some application areas where the ML8500 is especially useful.

Cellular heterogeneity facilitating specialization of cells is essential to control normal physiological processes as well as drive pathogenesis of many diseases. While it remains common to measure population-averaged properties in biological studies, the assumption that all cells are identical can lead to incorrect or at least imprecise assessments. To better understand heterogeneous characteristics and responses of individual cells, single-cell analytic methods should be applied to profile individual cellular properties. Using micro-fabrication technology, we have developed high-throughput microfluidic platforms, that enable tracking of thousands of single cells on-chip. Building on the cell isolation capability, we innovated a photomechanical actuation method to selectively detach and retrieve target cells for downstream analysis. Irradiation of a nanosecond-pulsed laser is utilized to generate shear force for detaching target cells cultured on carbon nanotube–polydimethylsiloxane (CNT-PDMS) composite. This non-destructive cell retrieval method can preserve cell viability, membrane proteins, and mRNA expression levels, so we have successfully performed single-cell transcriptome analysis as well as functional studies on the retrieved target cells. In addition to cell tracking and retrieval, we applied machine learning techniques to correlate cellular morphological features with cellular functions including migration, cancer drug response, and tumor initiation. Machine made significantly better predictions than experienced researchers, and we found novel morphological features facilitating breakthroughs in mechanistic understanding. The integrated single-cell manipulation and analysis augmented with machine learning will change how we understand cell biology and ultimately improve how we treat diseases.

Lupus nephritis (LuN) is an inflammatory kidney disease characterized by the infiltration of immune cells into the kidney, including T-cells, B-cells, and dendritic cells. Here, we combine high-dimensional immunofluorescence microscopy with computer vision to identify and segment multiple populations of cells. A U-Net was trained to segment CD4+ T-cells in high-resolution LuN biopsy images and subsequently used to make CD4+ T-cell predictions on a test-set from a lower-resolution, high-dimensional LuN dataset. This produced higher precision, but lower recall and intersection over union for cells in the low-resolution dataset. Further application of U-Nets to immune cell segmentation will be discussed.

We propose a novel deep learning algorithm, denoted as Deep Optical Flow (DoF), capable of interpreting and predicting cell behaviors with high accuracy in time-lapse fluorescence images. DoF has dual pipeline networks, including 4D-Rank convolution operations. One classifies the behavior of induvial cells while generating Optical Flow for the cells, whereas another predicts the next few frames of cells. DoF was verified with our and public datasets for cell tracking, segmentation, and identification. The experimental results demonstrate that DoF outperformed other state-of-the-arts in the analysis of cell behaviors. Therefore, these suggested that DoF has the potentials to become a novel tool for a better understanding of cell behaviors.

Fluorescence lifetime microscopy (FLIM) and time-resolved cytometry (TRFC) are robust platforms that can resolve complex protein and cellular interactions. Flow cytometry has been a prominent staple in clinical and research for decades. In conventional form, flow cytometers can count cells and evaluate biophysical and biochemical attributes using fluorescence and inelastic scatter light. Cytometry has evolved beyond conventional paradigms are becoming polychromatic and mulitparametric apparatuses that can evaluate complex cellular interactions in real-time. A distancedependent technique known as Förster Resonance Energy Transfer, or FRET is a powerful quantitative tool that enables the ability to monitor binding interaction and morphological changes in the macro and microenvironment of cells. FRET measurements require sensitive instrumentation to capture and resolve subtle changes in biophysical and biochemical characteristics. TRFC captures a unique parameter known as fluorescence lifetime which is sensitive to microenvironmental changes. Past studies have demonstrated TRFC’s ability to resolve complex FRET interactions. Herein, we present the evolution of the TRFC modular platform that incorporates a microfluidic device. The microfluidic device in this contribution acoustically linearly focuses cells down the middle of the microcapillary, allowing for maximum optical excitation and optimizing optical geometries to maximize the capture of fluorescence.

Rare circulating tumor cells (CTCs) and circulating tumor cells clusters (CTCCs) have been shown to increase the metastatic potential of tumors, with CTCCs increasing metastatic potential by 23 to 50 times. Due to the high metastatic potential of CTCCs, there is a growing interest in the detection and isolation of these clusters to improve diagnosis, stratification, and treatment of cancer patients. This study aims to utilize a custom confocal back scatter and fluorescence flow cytometer (CBSFFC) that will leverage the high sensitivity of flow cytometers for in vitro and in vivo detection of CTCCs without the use of exogenous contrast agents.

small sampling volumes, minimal hydrodynamic focusing avoiding need for most sheath fluid for higher sample throughput, weight/power requirements using super-bright LEDs light sources, “top-hat” focusing optics minimizing cell-positioning requirements, small silicon photomultiplier sensors, with tightly-integrated electronics, all battery-powered and light-weight for portability. Smartphone technology includes CPU, touch-screen, digital speech recognition interface, telecommunications via phone and/or internet, GPS chip to mark the location of the patient using Motorola’s Moto Z3 Play, modular/open-architecture smartphone. The device is software-configurable from a Linux workstation running Android Studio IDE with Java and Python programming. Portable microfluidic-cytometry devices for measurements in the field requires a serious overall systems-level design to face the many engineering tradeoffs encountered for true portability, including: (1) sampling systems of small sample volumes with minimal need for sheath hydrodynamic focusing both to avoid the need for large amounts of sheath fluid and to enable higher volumes of actual sample, (2) weight/power requirements that dictate use of super-bright LEDs as light sources, integrated “top hat” focusing optics to minimize the need for cell positioning and very small silicon photomultiplier sensors, with tightly integrated electronics (3) powered by small batteries or regenerative power sources such as solar, and (4) light-weight and robust enough for portability in potentially extreme environments. Initial prototyping used Raspberry Pi credit-card sized microcomputers, but longer-term development is being geared toward smartphone technologies. Smartphone technology can provide a powerful CPU, touch-screen and digital speech recognition interface, telecommunications via phone and internet, with a GPS chip to mark the location of the patient in the field. We are now using Motorola’s new Moto Z3 Play, a dual 12-megapixel/5-megapixel rear camera with 8-megapixel front camera, modular and open-architecture smartphone which allows building custom applications using hot-swappable accessories via Moto Development Kit modules, including externally-mountable Li-battery packs to power linked cytometry optics and electronics modules and microSD card storage with 512GB addressable memory linkable to other special purpose modules via USB-C connections. The device is software-configurable from a Linux workstation running Android Studio IDE with Java and Python programming.

Cell-based therapy is an attractive strategy for the long-term management of various chronic diseases. Mesenchymal stem cells (MSCs) are a heterogeneous group of cells that have demonstrated clinically relevant therapeutic effects. The proliferative and therapeutic potential of MSCs can be characterized by the culture quality, which is reflected by their morphological phenotype. Morphological analysis has been a robust method for monitoring culture quality, but visual inspection is subjective and time-consuming. Our goal is to develop an automated algorithm to segment MSCs for an objective, non-invasive, and rapid cell assessment. We have built an algorithm to segment MSCs using U-Net architecture trained with 71 phase-contrast micro- graphs having 472 cells. MSC culture images are pre-processed and given as inputs to the trained U-Net model that provides a prediction map for cell segmentation. The U-Net output is then post-processed using morphological operations to get rid of false positive cell detections. Results were validated using visual inspection from MSC experts. Our independent test dataset of 36 images consisted of 186 cells. We obtained a sensitivity of 0.742 and a precision of 0.789 for cell detection and a Dice-Sorensen score of 0.823 ± 0.051 for segmentation. The proposed algorithm shows the potential to segment MSCs with accuracy and robustness higher than conventional U-Net. Automated cell segmentation would enable rapid quantification of cytomorphological features and may also drive stem cell quality control processes.

The movement of intracellular vesicle contains essential biomedical information, mediating drug delivery and virus transmission. However, due to the interaction between vesicles and cytoskeletal networks, the trajectories of vesicle transport are often too complicated to understand the details. Particularly, identifying active transport via cytoskeletal network from random motion requires time-consuming mathematical methods. In this paper, we propose a machine learning approach to categorize the vesicle transport into active transport and random movement, using the features computed from the vector analysis of 3D vesicle transport trajectories. This approach is expected to simplify the process for vesicle transport data analysis.

Systemic lupus erythematosus (SLE) is a complex, systemic autoimmune disease with many clinical presentations including lupus nephritis (LuN), or chronic inflammation of the kidneys. Current therapies for SLE are only modestly effective, highlighting the need to better understand networks of immune cells in SLE and LuN. In this work, we assess the performance of two convolutional neural network (CNN) architectures –Mask R-CNN and U-Net— in the task of instance segmentation of 5 immune-cell classes in 31 LuN biopsies. Each biopsy was stained for myeloid dendritic cells (mDCs), plasmacytoid dendritic cells (pDCs), B cells, and two populations of T cells, then imaged on a Leica SP8 fluorescence confocal microscope. Two instances of Mask R-CNN were trained on manually segmented images—one on lymphocytes (T cells and B cells), and one on DCs (pDCs and mDCs)—resulting in an average network sensitivities of 0.88 ± 0.04 and 0.82 ± 0.03, respectively. Five U-Nets, one for each of the five individual cell classes, were trained resulting in an average sensitivity of 0.85 ± 0.09 across all cell classes. Mask R-CNN yielded fewer false positives for all cell classes, with an average precision of 0.76 ± 0.03 compared to the U-Net object-level average precision of 0.43 ± 0.12. Overall, Mask R-CNN was more robust than the U-Net for segmenting cells in immunofluorescence images of kidney biopsies from lupus nephritis patients.

We demonstrate a deep learning-based offline autofocusing method, termed Deep-R, to rapidly and blindly autofocus a single-shot microscopy image captured at an arbitrary out-of-focus plane. Deep-R is experimentally validated using various tissue sections that were imaged with fluorescence and brightfield microscopes. Furthermore, snapshot autofocusing under different defocusing scenarios is demonstrated, including uniform axial-defocusing, sample tilting, cylindrical and spherical distortions within the field-of-view. Compared with other online autofocusing algorithms, Deep-R is significantly faster while having comparable image performance. Deep-R framework will enable high-throughput microscopic imaging over large fields-of-view, improving the overall imaging throughput, also reducing the photon dose on the sample.

High signal-to-noise ratio (SNR) images are necessary for analyzing sub-cellular features in biomedical images. Acquisition of such images may be limited by temporal or photon-budget-based imaging constraints. This study aims to use deep-learning-based image restoration methods to extract morpho-functional information from low-SNR, depth-resolved, label-free, two-photon images of human cervical tissue. A deep convolutional autoencoder model was trained using single-frame image inputs and multiple-frame averaged ground-truth image pairs. Automated analysis of restored reduced nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD) two-photon excitation fluorescence (TPEF) images extracts depth-dependent, morpho-functional information otherwise lost in single-frame images.

The ability to use a wide range of wavelengths for deep penetration is important in order to target or avoid absorption bands of the biological media. By utilizing the nonlinear optical effect in the scattering bio-soft-matter, we demonstrate the self-trapping and guiding of light in sheep red blood cell suspensions for a range of different wavelengths. By pump-probe type coupling, biological waveguides formed at one wavelength can effectively guide a wide spectrum of light at low power. Finally, we investigate propagation and guiding of non-Gaussian beams in biological suspensions.

Amyloid fibrils are highly stable and organized peptide or protein structures that on the one hand can cause partially severe diseases such as Alzheimer's disease and on the other hand play fundamental roles during a plethora of biological processes. Nevertheless, there are still plenty open questions concerning their formation. We present a thermophoretic trap which is able to confine the Brownian motion of single amyloid fibrils via temperature gradients. The time-resolved tracking of the fibrils' rotational diffusion coefficients in presence of monomers permits to extract their growth rates or to directly observe secondary growth processes as fragmentation.

We report the advancement of magnetomotive μOCT microrheology for measuring viscoelastic parameters of mucus in air-liquid interface cystic fibrosis (CF) cell cultures. We dispersed magnetic microbeads into the mucus and conducted a creep test by applying a constant magnetomotive force to the sample, while imaging the resulting mucus deformation using μOCT. The deformation of mucus compartments was then analyzed using ImageJ. The viscoelastic parameters were extracted by fitting the deformation to a power-law rheology model. Preliminary results showed that magnetomotive μOCT microrheology can probe the heterogeneous viscoelasticity of CF mucus in a spatially resolved manner.

To measure the metastatic or mechanical properties of cancer cells, various methods have been implemented. However, these methods have some shortcomings that they are time-consuming, require direct contact with cells, and are measuring accuracy is sensitive to boundary conditions of the sample. To overcome these shortcomings, in this study, we developed a cell mechanical property measurement system based on the quantitative phase imaging system and ultrasound stimulation system capable of the precise determination of the invasion potential of cancer cells. By using a developed system, it is possible to successfully measure the change in thickness due to the ultrasound stimulation of various types of cancer cells according to various acoustic pressure. In addition, the time to recover the original thickness after ultrasound stimulation was also measured with a high-speed camera to characterize the mechanical properties of the cancer cells.

I will present our recent development of high-throughput super-resolution microscopy for robust imaging and reconstruction of super-resolution images on a widely used type of clinical samples – formalin-fixed, paraffin-embedded (FFPE) tissue, referred to as PathSTORM. Its application to visualize disrupted higher-order chromatin folding in early carcinogenesis will also be discussed.

Photonics localization due to light scattering is an important probe to understand the molecular specific structural changes in brain cells due to diseases or abnormalities. Chronic alcoholism is associated with medical, behavioral, and psychological problems including physical damage in brain cells/tissues. The effects of chronic alcoholism on brain cells/tissue structures at the nanoscale are not well understood. On the other hand, probiotic treatment has shown some reversing effect in curing the abnormalities in an alcoholic’s brain. In this work, we study the effect of probiotic, Lactobacillus Plantarum treatment on chronic alcoholic brain cells/tissues using a mouse model. We evaluate the light localization properties of molecular specific spatial mass density fluctuations based on mesoscopic physics-based inverse participation ratio via confocal imaging of cells, confocal-IPR technique. Using the technique, we probe overexpression of astrocyte and microglial cells, and chromatin structures of different brain cells, by probing molecular specific overexpression by staining the cells with appropriate dye/proteins and then calculating the degree of spatial molecular structural disorder (Ld). The confocal-IPR analysis of alcoholic astrocytes, microglia, and chromatin of the mice brain cells show an increase in the structural disorder indicating that alcohol has an adverse effect on different brain cells and nuclei. Whereas the normalcy in the structural disorder of these brain cells happens when probiotics were fed simultaneously with alcohol, confirms the improvement in chronic alcoholic brain health. The potential application of this novel approach to diagnosing the alcohol effect and probiotic treatment in the alcoholic brain is explored.

In this work, we introduce a technique capable of recovering both the refractive index and thickness maps of a cell using a single measurement in the form of a color photograph of the sample. Our method exploits the appearance of thin-film interference colors on a cell when placed on top of a suitable surface. An inverse search algorithm is used to map these colors to the refractive index and thickness values of each pixel in the image. Experimentally, we show the technique can achieve a 10-2 RIU sensitivity, sufficient to differentiate between cellular organelles.

The quantitative measurement of nanoscale structural alterations in cells/tissues is important to understand their physical states. Molecular specific light localization technique and microscopic imaging are highly sensitive spectroscopic approaches for studying the structural abnormalities in brain cells under a sedative condition. Fetal alcohol syndrome and other neurological disorders are the severe, irreversible outcomes of fetal alcoholism. The alcohol consumed by a pregnant mother passes through the placenta to the growing womb and inhibits the growth of vital organs of the baby resulting in brain damage and other birth defects. This damage is initially at the nanoscale level in cells/tissue. We probe fetal alcoholic pup brain cells using dual spectroscopy approaches: 1) photonics localization method using inverse participation ratio via confocal imaging, confocal-IPR, to probe DNA and histone molecular spatial structural alterations; 2) a recently developed spectroscopic technique, partial wave spectroscopy (PWS), which combines mesoscopic physics with microscopic imaging and detects the nano to submicron scales alterations in pup’s brain cells/tissues. The molecular structural abnormalities calculated based on light localization properties show an increase in the degree of spatial molecular structural disorder in DNA and a decrease in histone. An increase in spatial disorder in DNA may suggest DNA unwinding while reduced structural disorder in histone may indicate the release of histone from the DNA and helps in the unwinding of the DNA and gene expression. This result is further supported by the PWS result which shows an increase in the degree of structural disorder in chronic alcohol-treated mice pup’s brain tissues.

As the number of cancers is steadily increasing, doctors are in need of automatic tools with better and faster analysis methods to help them with the diagnosis. One way to tackle this challenge is to propose label-free methods capable to analyze a large number of samples. Recent development in photonics components could enable to use infrared light to detect abnormal tissues and Mid-IR imaging can provide an unequivocal information about the biochemical composition of human cells. The combination of a set of Quantum Cascade Lasers (QCLs) and lensfree imaging with uncooled bolometer matrix will allow the biochemical mapping over a wide field of view. This experimental setup coupled to machine learning algorithms (Random Forest, Neural Networks, K-means) can help to classify the biological cells in a fast and reproducible way. Images from the frozen section tissue of nude mice bearing human orthotropic oral cavity tumors from the CAL33 cell line have been acquired and analyzed. Using amide and DNA absorption bands, we achieved up to 94% of successful predictions of cancer cells with a population of 325 pixels corresponding to muscle tissues and 325 pixels corresponding to cancer tissues. This work may lead to the development of an imaging device, that could be used for cancer diagnosis at hospital.

Our approach to multispectral fluorescence microscopy can non-invasively identify the biomolecular composition of cells and capture complex biological heterogeneity which is fundamentally important for biological research and medical diagnostics. We have applied this technology to demonstrate the embryo quality for chromosomal abnormalities (containing euploid and aneuploidy cells) and understanding the biochemical signatures of polycystic ovarian syndrome (PCOS) oocytes. We then explored oocyte quality following treatment with the NAD+ precursor NMN. These findings demonstrate the utility of our approach to the multispectral assessment of autofluorescence for the non-destructive, label-free assessment of clinically relevant problems.

The ability to generate 3D angiogenesis models is central for tissue engineering and drug screening applications. However, existing bio-fabrication technologies have yet to attain precise guidance of capillary networks in 3D. Here, we present our latest results in fabricating capillary networks using a novel laser-assisted bioprinting technique named Laser Induced Side Transfer (LIST). We found that LIST-printed human umbilical vein endothelial cells (HUVECs) present negligible loss of viability and maintain their abilities to migrate, proliferate and form intercellular junctions. Furthermore, we showed that LIST enabled the formation of capillary-like networks in 3D with high spatial precision (50 μm) over a large volume (1 cm3). Those networks were validated as angiogenesis assays for pro- and anti-angiogenic compounds. LIST could be widely adapted for applications requiring multiscale bioprinting capabilities, like the development of 3D drug screening models and artificial tissues.

Autofluorescence (AF) spectroscopy and imaging are used widely in the field of biomedicine for disease diagnosis and screening. Concentrations of many intrinsic fluorophores share a strict relationship with morphological and functional characteristics of tissue, making AF spectroscopy a powerful tool to directly monitor tissue health. One major challenge with AF imaging is maintaining high signal-to-noise ratios, as emission levels are low due to poor fluorophore quantum efficiencies and low illumination power levels. As a result, maximizing the throughput of the measurement system is critical to mitigate losses. Diffraction gratings are commonly used for spectroscopy for dispersion, but rarely exhibit efficiencies above 80%, limiting the system performance. Liquid crystal polarization gratings (LCPGs) are a relatively new technology that possess extremely high efficiency, typically over 90% for the design wavelength, and in some cases up to 99%, making it an attractive option for AF spectroscopy. However, with unpolarized autofluorescent light, the grating would split the light equally into two orders, only one of which could be collected with a standard detector array. Here, we present the first design and demonstration of a visible light spectrometer using a LCPG. To overcome the loss of 50% of incoming unpolarized light being split into separate orders, we report a novel prism system used to merge the two orders into a single spectrum with minimal degradation of spectral resolution. Our results indicate that that using LCPGs could increase signal levels by up to 20%, significantly improving the performance of spectrometers used for biomedical AF imaging.

Cell instance segmentation is a critical task to perform for the quantitative analysis of 3D live-cell images. Existing studies mostly apply a region proposal-based approach to instance segmentation of microscopy images. However, they often fail to detect cells in 3D live-cell images, which have complicated and heterogeneous shapes, often closely linked to the neighborhood cells. A different approach based on point proposal methods is more robust in handling complex shapes than the box proposal. These methods take an image and a proposed point in the form of its location (x; y) as input and generate a mask for an object that includes the point. They also show that the model can improve the prediction by utilizing negative point proposals chosen from false-positive areas. In this paper, we propose a novel cell instance segmentation approach based on point proposal for 3D cell imaging. Different from existing work, however, our model utilizes the nuclei of cells as point proposal and employ them as positive and negative point proposals. We constructed the 3D NIH3T3 dataset for training and evaluation, and examine the proposed model qualitatively on three independently gathered cells; HeLa, A549, and MDA-MB-231. Our model exhibits superior quantitative results; moreover, compared to previous methods, it properly predicts cell lines, which are not even well-annotated during training.

The monitoring of cardiopulmonary functions is essential to diagnose various heart diseases. Moreover, with the spread of the novel coronavirus which is particularly fatal to human life due to ‘silent hypoxia’, the monitoring of cardiopulmonary is being more and more important. We here demonstrate a multimodal necklace capable of monitoring blood oxygenation but also blood flow rate, velocity, and possibly blood pressure. The necklace consists of pulse oximeter sensors and single-element ultrasound transducers. The estimation of a pulse wave velocity (PWV), which is highly correlated with blood pressure, is investigated by exploiting the pulse oximeter sensor and ultrasound transducer in this study. The developed multimodal necklace was evaluated through in vivo study. The results showed that both the chemical and mechanical information of cardiopulmonary function can be monitored with low hardware complexity and resource using the multimodal necklace.

The importance of telemedicine using home healthcare system comes to the fore due to the COVID-19. Smartphone-based system can be a proper home healthcare tool for the diagnosis of early stages of diseases due to its portability, easy to use, and cost-effectiveness. Dental caries (DC) is one of the most common and serious diseases in dentistry. In the early stage of DC, white spot lesions (WSLs) can be observed. To detect WSLs, we demonstrate a smartphone-based multimode imaging system (SMIS) for the quantitative diagnosis of early DC. SMIS can offer RGB color, polarization, fluorescence, and multispectral imaging.