Optical Methods for Inspection, Characterization, and Imaging of Biomaterials VI

The work describes deep learning-enabled holographic tomography for neuroblastoma cell processing, analysis, and diagnosis through three-dimensional (3-D) cell refractive index (RI) model. Deep learning-assisted approach is applied to execute effective segmentation of 3-D RI cell morphology for the different cellular states under normal, autophagy, and apoptosis. The biophysical parameters of 3-D RI cell morphology are analyzed and selected for learning-based classification to identify cell death pathways. The results show that the proposed approach achieve of 98% in identifying cell morphology through optimized biophysical parameters

In recent years, label-free microscopy has gained momentum over the well-established fluorescence microscopy, as it allows overcoming many important drawbacks related to the staining process. In particular, tomographic phase microscopy in flow cytometry has opened the route to the label-free, 3D, quantitative and high-throughput recording of live suspended cells. However, the lack of intracellular specificity due to the missing staining is limiting its promising applications. Here we show a computational method for the 3D stain-free segmentation of the nucleus inside the tomograms of flowing cells based on an ad hoc clustering of the intracellular voxels according to their statistical similarities.

We have developed different classes of quantitative phase microscope for applications ranges from material inspection to biomedical assays in cells and tissues. Here we focus on quantifying the conformations of many driven and active nematic fluids exhibit complex microstructures. We explore the structures emerging in a pressure-driven nematic lyotropic chromonic liquid crystal in a microfluidic channel. We show that twist-type topological defects spontaneously emerge under flow. Our single-shot quantitative polarization imaging method allows us to quantify the fluctuations of these defects, which we show to reflect the tumbling character of the liquid crystal. We report how the defect size is governed by the balance between nucleation and annihilation forces, a balance that can be tuned by the flow rate. Such control over the microstructure opens pathways for using these nematic materials in optical devices and to control assembly of biological systems. If time permits, we will further present a novel nanofabrication technology for creating novel meta-optical components with high refractive index contrast based on Implosion Fabrication and will describe the applications of quantitative phase imaging for manufacturing process control.

Quantitative phase imaging (QPI) is a powerful label-free imaging technique that enables high-resolution, three-dimensional imaging of unlabeled samples by exploiting refractive index (RI) distributions as intrinsic imaging contrast. In this talk, we present the latest developments in 3D QPI techniques for investigations of live cells, tissues, and organoids over a long period of time. We will introduce several reference-free QPI approaches with incoherent light sources and discuss the considerations of multiple light scattering into a tomographic reconstruction algorithm1-3. Additionally, we will present a method of measuring 3D dielectric tensor tomograms, which expands the applicability of QPI to 3D birefringent materials. These technological developments hold great promise for advancing 3D label-free high-resolution imaging and its application in various fields, particularly in photonics.

Deciphering live cell dynamics using time-lapse multiparametric assays is expected to provide new insights into cellular machinery, while fostering groundbreaking biomedical applications. In this work, we take on the challenge to measure the optical response of live cells upon electrical excitation. We build on recent advances in coupled EIS and quantitative phase imaging (QPI) to obtain quantitatively time series of cellular parameters with label-free imaging at high spatial and temporal resolution. We aim to quantitatively assess cellular dynamics (cell cycle progression) both under physiologic conditions and exposed to selected stimuli triggering a wide range of effects from gentle to lethal ones. Using tailored optoelectronic materials, we exploit the coupling of AC electric fields to the substrate for boosting the analytic capabilities of EIS and quantitative phase imaging. This novel multimodal investigation provides new capabilities for gauging subcellular structure and dynamics changes in response to electrical excitation. Among others, we used magnified image spatial spectrum (MISS) microscopy, a high-speed and sensitive QPI method, to study the distribution of both electrical and optical parameters of live cells. We also present an application of combined EIS-light microscopy concept to assess the alterations of bacterial cells dynamics, at bacterium level, when exposed to a model (bactericidal) antibiotic. This new type of time-lapse microscopy encompassing the dynamics of both structural and electrical cellular fingerprints can provide a rapid phenotypic approach likely to replace some currently used antimicrobial susceptibility/resistance testing assays based on lengthy microbiological methods.

Fiber-optic dual-beam trap is an emerging method for studying biophysical properties of biological cells and organoids. Nevertheless, current optical manipulation is often limited to single-axis cell rotation. We introduce an innovative dual beam trap, utilizes multicore fibers to enable precise control of cell rotation about all three axes. This is achieved through the development of a physics-informed neural network that generates tailored light fields in the trapping region via the multicore fiber, allowing real-time holographic control of the optical force. By leveraging the capability of 3D cell rotation, our system enables 3D optical diffraction tomography with full spatial frequency coverage, eliminating the missing cone problem for cancer cells.

We build up a quasi-common-path lateral-shearing holo-tomographic optical setup to reconstruct the 3D refractive index profile of live cells. This setup is robust to external vibrations and is scaled down in dimension, in the perspective of employing such systems outside of optical labs. We apply a reconstruction algorithm based on high-order total-variation to cope with the reduced quality of the measurements acquired. We numerically validate this reconstruction approach and apply it to experimental data. The assessment of the robustness of the proposed solution is allowed by evaluating some biophysical parameters of biological relevance over the reconstructed tomograms.

A normal bright-field microscope may be updated with coherent sensing capabilities easily, inexpensively, and compactly using phase imaging microscopy in the Gabor regime. The digital sensor records an in-line Gabor hologram of the target sample by incorporating coherent illumination into the regular microscope embodiment, generated by a small defocus distance of the sample at the input plane. This hologram is then numerically post-processed to obtain the quantitative phase information of the sample. However, when dealing with Gabor's regime, coherent noise and twin-image disturbances affect the recovered phase distribution. In this contribution, we describe a single-shot method for reducing these two error sources based on wavelength multiplexing. The sample is illuminated by a multi-wavelength laser source that utilized three diode lasers, and the wavelength-multiplexed Gabor hologram is to be captured on a digital color sensor using a traditional RGB color camera in a single exposure. The presented phase imaging method is completed by the implementation of a new algorithm built on a modified Gerchberg-Saxton kernel to obtain an enhanced quantitative phase image of a sample that has been improved in terms of coherent noise removal and twin-image reduction. Complex field filtering in terms of hologram's imaginary and real part numerical alteration is in place in an iterative fashion. Experimental validations are carried out in a off-the-shelf Olympus BX-60 upright microscope with a 20X 0.46NA objective lens employing static (resolution test targets) and dynamic (live spermatozoa) phase sample quantitative imaging.

Deep learning techniques create new opportunities to revolutionize tissue staining methods by digitally generating histological stains using trained neural networks, providing rapid, cost-effective, accurate and environmentally friendly alternatives to standard chemical staining methods. These deep learning-based virtual staining techniques can successfully generate different types of histological stains, including immunohistochemical stains, from label-free microscopic images of unstained samples by using, e.g., autofluorescence microscopy, quantitative phase imaging (QPI) and reflectance confocal microscopy. Similar approaches were also demonstrated for transforming images of an already stained tissue sample into another type of stain, performing virtual stain-to-stain transformations. In this presentation, I will provide an overview of our recent work on the use of deep neural networks for label-free tissue staining, also covering their biomedical applications.

In this talk I will give an overview of the computational steps in the analysis of a single cell-based large-scale microscopy experiments. First, I will present a novel microscopic image correction method designed to eliminate illumination and uneven background effects which, left uncorrected, corrupt intensity-based measurements. New single-cell image segmentation methods will be presented using differential geometry, energy minimization and deep learning methods. I will discuss the Advanced Cell Classifier (ACC), a machine learning software tool capable of identifying cellular phenotypes based on features extracted from the image. It provides an interface for a user to efficiently train machine learning methods to predict various phenotypes. For cases where discrete cell-based decisions are not suitable, we propose a method to use multi-parametric regression to analyze continuous biological phenomena. To improve the learning speed and accuracy, we propose an active learning scheme that selects the most informative cell samples. Our recently developed single-cell isolation methods, based on laser-microcapturing and patch clamping, utilize the selection and extraction of specific cell(s) using the above machine learning models. I will show that we successfully performed DNA and RNA sequencing, proteomics, lipidomics and targeted electrophysiology measurements on the selected cells.

Non-linear excitation microscopy in the NIR (800 nm -- 1600 nm) offers several advantages for in-vivo imaging compared to conventional confocal techniques [1,2]. However, tissue penetration can still be an issue due to scattering and spherical aberrations induced on focused beams by the tissue. The use of low numerical aperture objectives to pass through the outer layers of the skin, together with high dioptric power microlenses implanted in-vivo close to the observation volume, can be beneficial to the reduction of optical aberrations. Here, we fabricated by means of 2 photon laser polymerization [3] plano-convex microlenses to be used for non-linear imaging of biological tissue. The microlenses have been tested on fibroblast cell culture as single lenses or arrays. A thorough test of the lenses wavefront is discussed together with the modulation transfer function and wavefront profile. We could retrieve magnified fluorescence images through the microlenses coupled to a commercial two-photon excitation scanning microscope. We perform a careful study of the signal-to-noise ratio of the images acquired through the fabricated microlenses finding that the signal/noise and the resolution are comparable to the ones obtained on commercial setups. These microlenses, when implanted in tissue, should allow in-vivo imaging with reduced spherical aberrations. The focal plane and magnification can be adjusted by changing the relative position of the microlens array to the microscope objective and the immersion medium. These results are opening the way to the application of implanted micro-optics for optical in-vivo inspection of biological processes. Acknowledgments: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 964481. References. [1] Bakker G.J., Weischer S., Ferrer Ortas J, Heidelin J., Andresen V., Beutler M., Beaurepaire, E., Friedl P. eLife 2022, 11, e63776 [2] A. Diaspro, G. Chirico, M. Collini. Q Rev Biophys. 2005 May;38(2):97-166. [3] C. Conci, E. Jacchetti, L. Sironi, L., et al. Adv. Optical Mater. 2022, 10, 2101103.

Despite their key-role during the histopathological diagnosis, staining procedures are expensive and time-consuming. Label-free microscopy provides an alternative since it allows the visualization of endogenous proteins without the need of extrinsic dyes. SuperµMAPPS, a novel AI-based method, analyzes the Polarized Second Harmonic Generation signal from collagen to characterize its micro-architecture in terms of fibrils mean orientation θF and anisotropy γ, related to tumor development. After a proper validation on synthetic images, human breast cancer samples at different growth stages have been analyzed through SuperµMAPPS, highlighting its capability to detect tumorous tissue at early stages in a real clinical context.

In an overview, digital holographic microscopy (DHM) for usage in a biomedical laboratory environment and the application of DHM-based quantitative phase imaging (QPI) for quantification of inflammation and toxicity related effects in tissue sections, cell cultures and blood cells are presented. First, the capabilities to determine inflammation as well as nanomaterial induced pathological alterations in dissected ex vivo colon and lung tissues are illustrated. Then, results from cell culture-based time-lapse growth monitoring assays and perioperative observation of living primary human leucocytes demonstrate DHM as a cytometric tool for in vitro toxicity testing and temporal disease course characterization.

High-resolution images over wide field of view (FoV) is a highly sought outcome in conventional microscopy of tissue slides. Fourier Ptychography (FP) overcomes this trade-off in label-free mode. Here we show that FP is suitable for kidney physiology. FP phase-contrast images of stain-free tissue slides are compared to light-microscopy images of the same slides to discuss the capability of QPI in describing the morphology of kidney samples. Indeed, renal inner structures such as glomeruli and tubules are clearly visible without staining. We extensively compare FP phase images and light-microscopy images of stain-free and stained kidney slides of different thicknesses.

Holo-Tomographic Flow Cytometry is a new technology for single-cell analysis that combine Phase-Contrast Tomography and Flow Cytometry opening to a new approach in biomedical field by high-throughput, tri-dimensional imaging of unstained cell populations. Holo-Tomographic Flow Cytometry allows to retrieve the unique all-optical 3D fingerprint for each cell flowing into the field-of-view opening to a wide range of applications such as: (i) identification of inner subcellular compartments; (ii) recognition of nanoparticle uptake and (iii) phenotyping of different subclasses in heterogeneous populations. Future perspectives are presented in the fields of liquid biopsy, drug resistance and genetic disfunctions.

We propose an analysis of the polarimetric properties of biological tissue (dichroism, retardance and polarization) for characterizing the tissue inherent characteristics and its polarimetric response when interacting with light. These polarimetric parameters are obtained through the experimental measurement of the Mueller matrix of the sample. We study the suitability of different groups of metrics for the contrast enhancement and classification of biological tissues, highlighting some metrics and some physical parameters. The results suggest the convenience of certain parameters which may be of interest in multiple biomedical scenarios.

Meta-optical devices have emerged as promising candidates for all-optical image processing. These devices are of subwavelength size and have the potential to address limitations of current image processing methods including processing speed, energy requirements as well as form factor. We present experimental results demonstrating the use of thin-film absorbers and optical metasurfaces to real-time detection of edges in images and the visualisation of phase objects including human cancer cells. Furthermore, we discuss progress towards the use of meta-optics for ultra-compact wavefront recovery. The findings to be presented have potential for applications in biological live-cell imaging, ultra-compact medical diagnostic tools, and wavefront correction methods.

The precise and non-invasive control over single particles is key for an array of physical and bio-medical applications, such as microfluidics and biophysics. In particular, the three-dimensional manipulation of single rare-earth-doped luminescent particles is of great interest due to their biocompatibility and the sensitivity of their luminescent properties to environmental conditions, which stand out among other dielectric luminescent particles. The analysis of the damped rotation dynamics of an optically trapped microparticle is a novel and powerful tool that allows not only the controlled and remote manipulation of the sensor, but also an improved characterization of the medium and fast recording of its content. Here, an optically trapped and rotated rare-earth-doped β-NaYF 4 :RE 3+ microparticle is presented as a novel sensor to characterize the properties of a liquid medium at the microscale (temperature, viscosity and detection of bio-objects).

In lensfree microscopy, the sample is placed close to the image sensor without any imaging lenses in between. This configuration provides the benefits of low cost and compact hardware assemblies as well an ultra-large field of view and a high space-bandwidth product. Image focusing and reconstruction are performed computationally, relying on algorithms such as pixel superresolution and the angular spectrum method of propagation. We present recent progress on improving the resolution to characterize nanoscale materials, application to protein and COVID-19 sensing, ultrafine air pollution monitoring, and high resolution incoherent (fluorescent) imaging.

A fundamental challenge of biology is to understand the vast heterogeneity of cells, particularly how the spatial architecture of cells is linked to their physiological function. Unfortunately, conventional technologies such as fluorescence-activated cell sorting and the Coulter counter are limited in uncovering these relations. In this talk, I introduce Intelligent Image-Activated Cell Sorting, a technology that performs real-time, intelligent, image-based sorting of live cells at an unprecedented rate of >1000 cells per second (Nitta et al, Cell 2018; Isozaki et al, Nat. Protoc. 2019; Mikami et al, Nat. Commun. 2020). This technology integrates high-throughput optical imaging, cell focusing, cell sorting, and deep learning on a hybrid software-hardware data-management infrastructure, enabling real-time automated operation for data acquisition, data processing, intelligent decision-making, and actuation. I show a new class of applications in immunology, cancer biology, infectious disease, microbiology, and food science. Furthermore, I introduce recent developments such as Raman image-activated cell sorting that achieves image-based sorting of single live cells in a label-free manner (Nitta et al, Nat. Commun. 2020).

Holographic cytometry is introduced as an ultra-high throughput implementation of quantitative phase image based on off-axis interferometry capable of extracting information from millions of cells flowing through parallel microfluidic channels. The approach allows high quality phase imaging of a large number of cells greatly extending our ability to study cellular phenotypes using individual cell images. The large volume of individual cell imaging data provides suitable input for training sets to develop machine learning and deep learning algorithms. Here we present our findings on application of this technique to examining red blood cells and to distinguishing carcinogen-exposed cells from normal cells and cancer cells. A study of storage lesion, the degradation of red blood cells due to aging, is presented. By using logistic regression to analyze morphological cell features, high accuracy for discriminating cells by storage time is obtained. Further study of red blood cells shows the ability to detect sickle cell disease by implementing deep learning algorithms with careful selection of classifier training features, suggesting potential avenues for diagnosis and monitoring of this disease. Finally, studies of carcinogen exposed cells compared to cancer cell lines show distinct traits between cell populations. Use of deep learning algorithms enables high accuracy in detecting cell phenotype. This has potential application for environmental monitoring and cancer detection by analysis of cytology samples acquired via brushing or fine needle aspiration. The results of these studies demonstrate the potential of holographic cytometry as a diagnostic tool based on high throughput single cell imaging.

I will review our latest results in the field of quantitative imaging flow cytometry using off-axis holography. Flow cytometry has a great diagnosis potential, due to its ability to analyze a large number of cells during flow for samples obtained from body fluids. Since cellular morphology analysis plays an important role in various clinical diagnoses, such as screening cancer and various chronic diseases, flow cytometry is much anticipated to incorporate imaging capabilities, providing a more comprehensive analysis by presenting a detailed morphological structure image of individual cells. In addition, some erroneous analysis results, yielded in conventional flow cytometry, can be eliminated by acquiring and analyzing such cell images by clearly distinguishing between cells, debris, and clusters of cells. While conventional flow cytometry measures the integral intensity of fluorescent emission, fluorescence microscopy is able to yield the exact morphology of the cell and its organelles. Recent advances in imaging technologies, as well as the exponentially evolving computational capacity, have enabled imaging flow cytometry (IFC) by integrating fluorescence microscopy and conventional flow cytometry. However, the current-generation IFC remains highly inaccessible technology, due to its cost, requirement for operator expertise, lack of accuracy, and lack of objectiveness of data produced. We developed new approach for IFC, which is based on stain-free interferometric phase microscopy, a digital holographic microscopy technique. Using an external interferometric module and cutting-edge deep-learning analysis methods, we generate virtually stained cell images of a volume of cells, with a clear morphological discrimination between various cellular organelles. The module acquires, in a single camera frame, the digital hologram of the cell during flow, and our rapid reconstruction algorithms retrieve the complex wavefront of the cell, from which the optical path delay (OPD) topological map is calculated. This map represents, on each spatial point, the product of the cell physical thickness and its refractive index, accounting for both the cell morphology and contents. Such map is subsequently used as the input to a deep convolutional neural network that is pre-trained to transform the cell OPD into a 2-D image with stained-like organelles. The same IFC setup is also used for automatic cell classification. We demonstrate using the system for cancer detection in liquid biopsies.

Recently, advanced flow cytometry analysis technology based on digital holography has been extensively studied, which can meet various challenges in clinical diagnosis. Especially in liquid biopsy, it has incomparable advantages. Urothelial Holographic Flow Cytometry (HFC) microscopy can provide rich intracellular information by changing the cell’s intrinsic properties with label-free and high throughput. Carcinoma (UC) is the second most common malignancy in men. Urine cytology detection is the most convenient early cancer screening method for UC patients. Here, we developed HFC to identify the cancer cells in urine. Holographic microfluidic imaging was performed to obtain the phase images of different cells in simulation urine, including red blood cells, white blood cells, epithelial cells, and a small number of cancer cells. This study demonstrates that HFC can achieve high accuracy, high throughput, and label-free cancer cell identification in the urine.

The recent development of tomographic phase imaging flow cytometry has unlocked the possibility to achieve data throughput comparable to the state of the art imaging flow cytometry systems, but with the great advantages to be fully label-free and 3D. On the other hand, the huge amount of data to manage becomes one of the main computational problems to face with. Here we propose the use of the 3D version of Zernike polynomials to efficiently encode single-cell phase-contrast tomograms, demonstrating high data compression capability with negligible information loss. The experimental validation of the proposed methodology will be reported.

We will describe MaxwellNet, a machine learning method for analyzing and designing optical systems using Maxwell's equations as the loss function in the optimisation process. We will describe results of the application of MaxwellNet to nanophotonics, nonlinear optics and optical diffraction tomography

Fourier Ptychography (FP) bridges convenient optical features like high lateral resolution and wide field of view in label-free quantitative fashion. To make FP viable for clinical practice, non-ideal setup conditions should be coped with. We show how a blind multi-look FP algorithm eliminates phase artefacts for imaging cells and tissue slides. Then, we use the multi-look FP outcome as a ground truth to train a GAN, which returns the sample complex amplitude in real-time from severely misaligned setups. Applications in the fields of mechanobiology, cell analysis, and tissue physiology are reported.

Machine learning in combination with microscopy is a well-established paradigm for disease identification at single-cell level. Here we show that the combination of machine learning and holographic microscopy is an effective tool for allowing higher classification performances if compared to other standard microscopies, like brightfield or fluorescence imaging. Moreover, by exploiting a priori information about the samples, the classification performance can be further increased. We demonstrate this paradigm for the differential diagnosis of hereditary anemias, in which RBCs, imaged by holographic microscopy, are used to predict firstly if an anemia occurs, then which type of anemia among five phenotypes.

Our body is composed of many different types of cells, which can be seen as building blocks of the human body. By taking a closer look, we see that each cell class has its own set of biophysical properties such as range of size, structural components, and functions. In other words, cells contain a cell specific information, which distinguish them from each other. But cells are not homogenous objects, as they are composed of membranes, cytoplasm, nuclei, and various organelles to name the most important contributors. Their label-free classification in different classes and/or sates is highly challenging, but highly demanded for a wide range of diagnostic purposes. Besides, the mentioned heterogeneity of intrinsic cell properties, one crucial point for the automatic classification of different cell classes with Neural Network approaches is the handling of so called out of distribution cases. In fact, Neural Network based automatic cell recognition and classification plays a rising role in cell biology and medicine, generally using phase contrast or bright-field images, while light scattering pattern snapshots are understudied. Although light scattering pattern can provide invaluable biophysical cell properties, which can facilitate cost-effective individual cell class recognition, it can be considerate as nicheness technique. We want to sort things out and illustrate the enormous potential of pure light scattering snapshots for cell recognition application based on Neural Networks. However, a deep learning-based cell classification should not only produce accurate predictions of known cell property combinations, but also detect out of distribution cells and reject or classify them in a new class of cells. Today, most existing Neural Network based image classifiers are trained based on the closed-set assumption, where the test data is assumed to be drawn from the same distribution of training data. In fact, when cells out of distribution are classified, closed-set approaches are forced to choose a class label from one of the known classes, thus limiting their applicability in dynamic and ever-changing cell diagnosis scenarios. For instance, thresholding the classification score value for out of distribution cell detection in a closed-set scenario is performing impractical. Thereby, to proof our concept, human peripheral blood mononuclear cells (PBMCs), which can be easily isolated from blood of healthy donors or buffy coats are of valuable interest. Although PBMCs have a different composition, phenotype, and activation status than cells found in tissue, they can be seen as liquid biopsy of our body. The liquid biopsy has drastically revolutionized the field of clinical oncology, offering the possibility of continuous monitoring by repeated sampling. In fact, clinicians need the appearance of different cell classes and their status, to perform a precise diagnosis, which on the other hand can depend on single cell information. There is evidence in literature that pure biophysical cell properties can reveal distinct characteristics of cells. In this, the screening of label-free cell information has recently shown to distinguish fractions of cell class and/or states, especially when dealing with sparsely present cells. Beside the single cell recognition, the relative peripheral cell class count alterations, lower or higher than physiologic one, as well as anomalous cell shapes or cytoplasm complexity can be useful indicators for dysregulated responses to inflammation stimuli. Therefore, we present a simple microfluidic based single cell discrimination approach for by the mean of a label-free light scattering, together with a deep learning procedure for cell type prediction in liquid biopsies. Here optical signatures are recorded by a static light scattering apparatus, which continuously measure scattering profiles of passing living cells in microfluidic flow condition. Deep learning was trained with such scattering images as input to finally predict the searched for cell types in the testing phase. The prediction model was trained with human peripheral blood cells[29-33]. We tested our classification algorithms, regarding single cell type classification accuracy and for mixed cell samples to show the performance of our label-free classification approach. Furthermore, we show for the first time the ability to label-free predict out of distribution cells from scattering data using an open-set neural network approach, which significantly expand the application field of the presented single cell classification method. However, compared with standard flow cytometric approaches—which are known to have high instrumentation and service costs—our method is very simple and cost-effective, permitting a classification of cell subtypes without large numbers of cells and resource-intensive labelling. More importantly, measurements are realized using a lab-on-a-chip approach permitting the measurement of living cells in suspension, which furthermore are collectable and re-usable for other diagnostic investigations or therapeutic approaches.

H&E stained sections are the gold standard for disease diagnosis but, unfortunately, the staining process is time-consuming and expensive. In an effort to overcome these problems, here, we propose a virtual staining algorithm, able to predict an Hematoxylin/Eosin (H&E) image, usually exploited during clinical evaluations, starting from the autofluorescence signal of entire liver tissue sections acquired by a confocal microscope. The color and texture contents of the generated virtually stained images have been analyzed through the phasor-based approach to detect tumorous tissue and to segment relevant biological structures (accuracy>90% compared to the expert manual analysis).

The structural and microscopic examination of cartilage and tendon tissue is of central interest for the characterization of torn ligament healing or pathological tissue changes. Conventional microscopy or optical coherence tomography methods can only partially resolve tissue types and their structural orientation. Two- or multiphoton microscopy and second harmonic generation provide information about the material composition or alignment of nanoscaled structural proteins. In most systems, femtosecond lasers are used, which leads to high system costs and requires special system components. The aim of this work is to investigate a narrowband nanosecond laser with a pulse energy several orders of magnitude higher, reducing the required number of laser pulses to be averaged per pixel. Second harmonic generation and two photon excited fluorescence can be used to non-invasively examine deeper tissue structures. High lateral resolution was achieved by scanning the sample. Simultaneous real-time visualization of collagenous and cellular structures was attained. Defined aligned collagenous fibers of a sheep tendon were investigated. The anisotropy of the collagenous structures could be demonstrated. It was possible to realize a two-dimensional imaging method with a maximum point density of 5080 PPI and a numerical aperture of 0.15. The method allows the simultaneous separate observation of collagen fibers through the use of second harmonic generation and cellular tissue using a narrow-band nanosecond laser for two photon excited fluorescence.

We present a new AI-based method for the quantification of liver fibrosis in tissue sections stained with Picro Sirius Red which highlights collagen. The method segments and quantifies collagen, a marker of the fibrotic response, through a deep learning model trained on 20 whole-slide images. The results show a Dice score > 90% compared to manual annotations, demonstrating its potential aid during diagnosis. Furthermore, our approach can be extended to other staining protocols.

We propose a new label-free method for noninvasive and automated cell processing for classification of WBCs. This is done by acquiring off-axis holograms of each cell during flow and achieving its optical path delay (OPD) profile. Based on this map, we extract highly discriminative features used to detect, classify, and differentiate between distinctive cells using a deep convolutional neural network. This label-free method might bring to new analysis tools for blood test processing.

I will present a photonic sensor that can be used for remote sensing of many biomedical parameters simultaneously and continuously. The technology is based upon illuminating a surface with a laser and then using an imaging camera to perform temporal and spatial tracking of secondary speckle patterns in order to have nano metric accurate estimation of the movement of the back reflecting surface. The capability of sensing those movements in nano-metric precision allows connecting the movement with remote bio-sensing and with medical diagnosis capabilities. The proposed technology was already applied for remote and continuous estimation of vital bio-signs (such as heart beats, respiration, blood pulse pressure and intra ocular pressure), for molecular sensing of chemicals in the blood stream (such as for estimation of alcohol, glucose and lactate concentrations in blood stream, blood coagulation and oximetry) as well as for sensing of hemodynamic characteristics such as blood flow related to brain activity. The sensor can be used for early diagnosis of diseases such as otitis, melanoma and breast cancer and lately it was tested in large scale clinical trials and provided highly efficient medical diagnosis capabilities for cardiopulmonary diseases. The capability of the sensor was also tested and verified in providing remote high-quality characterization of brain activity.

The concentration dependent attenuation, group refractive index and the group velocity dispersion (GVD) of dense turbid media were determined with a Mach-Zehnder interferometer in the spectral domain for wavelengths between 400 nm and 930 nm. After calibration, all optical properties could be retrieved from a single measurement. Dependent scattering has only a small effect on the real part of the effective refractive index of a suspension. The higher sensitivity of the GVD compared to the group index allows us to test the particle concentration dependence of the real part of the effective refractive index of the medium. It was found that the interparticle correlations have a measurable effect on the GVD. The phase refractive index can be fitted to the concentration dependence of the group index. The combined measurement of the attenuation and refractive properties of the particle suspension allowed the estimation of the particle size distribution of the turbid medium through forward Mie calculations.

Integrated optics has been intensively developed the last decades for the production of components for telecommunications or sensors in metrology. The basic element constitutes the waveguide which transmits the information by guiding the light. The manufacturing processes take place in clean rooms with controlled atmosphere (pressure, temperature, dust) and with generally various deposition and coating machines. The production of these photonics devices remains relatively expensive since it is carried out in a clean room by series of manipulations which requires the intervention of a specialized engineer into a specific environment. More recently, techniques for manufacturing different mechanical parts of mostly research scientific devices have been developed using commercial 3D printers without the need of clean room environment. With such commercial machines, plastic materials are heated and then shaped by controlling their flow through a nozzle. This work concerns the study of the feasibility of realizing and shaping a kind of integrated optics by a simple 3D printer in a normal environment. It aims to understand and develop the fabrication of an integrated planar optics with polymer waveguides using a 3D printer and G-code programming. The coating processes and the G-codes of the 3D printer have been developed in order to produce the thermoplastic polyurethane (TPU) guiding structures onto a silicon/silica wafer. Finally, the characterization of this TPU organic (measurements of surface or energy tension, Raman analysis, ellipsometry measurements) and the optical injection made possible to validate this concept in terms of production and approach. The results concerning the shaping of various rib waveguides and obtaining a propagation of the light in organic thermoplastic polyurethane planar guides with fiber tapers by this simple and low-cost printing processes are positive. This work further proves that the development of integrated optics on thermoplastic polyurethane by 3D printing is possible; this opens the way to the realization of other optical patterns based on the 3D printer principle.

Preventive medicine is flourishing and is becoming increasingly important today [1]. One key parameter in this field is vascular stiffness, which depends on blood flow velocity [2,3]. If the velocity is over 12 m/s, the vascular is not healthy [4]. Blood flow velocity is typically calculated by measuring the time difference between pulse wave fluctuations at the ankle and wrist. However, there is a growing demand for non-contact which required smaller, non-contact systems to replace traditional large, contact-based methods [5]. Two of the most common non-contact vibration measurement methods are Doppler laser interferometry and shearing-speckle interferometry [6,7]. Doppler laser interferometry requires sweeping with a Galvano mirror, making it complicated and expensive [6]. On the other hand, shearing-speckle interferometry has a simple configuration, does not require sweeping, and is relatively inexpensive [7]. Furthermore, it can suppress the effect of body movement as it uses differential data. In this study, we aimed to develop a blood flow velocity measurement device using shearing-speckle interferometry. The experimental method involved the measurement of luminance distribution of arterial vascular variation at the wrist and the calculation of blood flow velocity from posture and cumulative time-directed integrals of the velocity along the X direction. A camera with 200 fps was used. The experimental results showed that the time difference between arbitrary vibration points was 6.4 m/s and the distance between the arbitrary vibration points was 36 mm. This resulted in a blood flow velocity of 5.6 m/s, which falls within the blood velocity range of 5-12 m/s for an adult male. The results indicate that blood flow velocity can be successfully estimated using shearing-speckle interferometry. In conclusion, this study has demonstrated the feasibility of using shearing-speckle interferometry composed of portable devices to estimate blood flow velocity in preventive medicine. This technology has the potential to improve the way we monitor and diagnose vascular stiffness and could have far-reaching implications for the field of preventive medicine. References [1] Razzak, M. I., Imran, M., & Xu, G. (2020). Big data analytics for preventive medicine. Neural Computing and Applications, 32, 4417-4451. [2] Jarrett, R. J. (1996). The cardiovascular risk associated with impaired glucose tolerance. Diabetic medicine: a journal of the British Diabetic Association, 13(3 Suppl 2), S15-9. [3] Borch-Johnsen, Knut, et al. "Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria." Lancet 354 (1999): 617-621. [4] Kizilova, N., Mizerski, J., & Solovyova, H. (2020). Pulse wave propagation along human aorta: a model study. Journal of Theoretical and Applied Mechanics, 58. [5] Taylor, W., Abbasi, Q. H., Dashtipour, K., Ansari, S., Shah, S. A., Khalid, A., & Imran, M. A. (2020). A Review of the State of the Art in Non-Contact Sensing for COVID-19. Sensors, 20(19), 5665. [6] Tabatabai, H., Oliver, D. E., Rohrbaugh, J. W., & Papadopoulos, C. (2013). Novel applications of laser Doppler vibration measurements to medical imaging. Sensing and Imaging: An International Journal, 14, 13-28. [7] GANESAN, A. R.; SHARMA, D. K.; KOTHIYAL, Mahendra P. Universal digital speckle shearing interferometer. Applied optics, 1988, 27.22: 4731-4734.

Precise manufacturing of volume scattering phantoms with direct control of the scattering and absorption coefficients opens up the possibility to examine the behavior of focus generation and the resulting enhancement behind these phantoms by applying wavefront shaping. Volume scattering phantoms allow measurements on samples with selected optical thicknesses. Specific combinations of scattering and absorption properties can mimic the interaction of coherent monochromatic light with biological tissue at a predetermined wavelength. Phantoms similar to muscle and fat tissue and solely scattering samples are being designed to characterize the achievable enhancement of the optimized focus intensity versus the mean intensity before optimization.

To measure optical properties of turbid fluorescent samples, an advancement of an inhouse developed integration sphere system was build up. This system based on a 3D printed integrating sphere uses photodiodes with lock-in technics for detection. The tuneable monochromatic illumination is realized by a laser pumped xenon light source with an integrated monochromator. A second monochromator on the detection side allows to discriminate between inelastically and elastically scattered light from the fluorescent media. By measuring the reflection and transmission from the investigated turbid sample, the wavelength dependent optical properties (absorption coefficient µa, effective scattering coefficient µs’) as well as the quantum efficiency (or concentration) of the fluorophore can be determinate using lookup tables generated from Monte Carlo simulations. Validation measurements are performed on turbid rhodamine phantoms.

Oral hygiene is one of the key measures recommended by WHO. The aim of this work is to assess the efficiency of different methods and materials for manufacturing dental prostheses. Several such prostheses are considered. Three manufacturing and techniques are utilized: (i) conventional, (ii) milling and (iii) printing. Two assessment methods are utilized: microbiology evaluations and optical coherence tomography (OCT) imaging. The latter involves an in-house Swept Source (SS) OCT system operating at 1300 nm, with a 15 μm axial resolution. Sampling method: cells are detached from the surface of the gingival mucosa and dental prostheses by scraping (like to Babes-Papanicolau cytological examination for the evaluation of the cervical mucosa). After sampling they are placed on glass slides for examination and are called smears. The presence of physiological or pathogenic microflora can be identified on the smears, as well as the local effect, which translates into the induction of the immune response. The smears are stained by using APT-Drăgan and Babeș-Papanicolau methods. APT-Drăgan method is a fast method that is performed in maximum one minute and is based on a single basic stain (blue methylene), which has the property that in an alkaline aqueous solution the color turns from blue to purple, pink or red, a phenomenon called metachromasia. Depending on the pH of the collected specimen, the nuclear or cytoplasmic structures show color variations. Babes-Papanicolau method is a trichrome staining, a combination of a nuclear stain, hematoxylin and two contrast stains Orange G and EA-50 (polychrome with light green). Orange G stains keratin, and EA-50 (a dual stain, eosin and azure) stains squamous epithelial cell cytoplasm, nucleoli, and erythrocytes. The advantage of the cytological examination consists in efficiency, safety, speed, simplicity and the fact that it is a non-invasive medical procedure. Smears obtained from the surface of the gingival mucosa and dental prostheses are examined under an optical microscope. Cellularity and bacterial flora were evaluated, comparing them on different types of prostheses, depending on the cellular and inflammatory elements on the mucosal surface. The cytodiagnostic identified cellular lesions of an inflammatory, allergic, and tumor nature. It is possible to recognize: specific inflammatory lesions (epithelioid cells, giant cells) and non-specific inflammatory lesions, the presence of inflammatory cells, bacterial flora, tumor cells. Conclusions are drawn regarding the efficiency of each of the considered manufacturing methods and materials. The capability of OCT to perform relevant investigations is assessed, as well, in comparison to the common microbiology evaluation.

A structured wave plate (SWP) is an optical element allowing the generation of optical vortices and vector fields with spatially variant polarization. The SWP has recently become a key tool in various optical experiments, including biological imaging. As the SWP has a birefringent structure, the polarization transformation meets the optimal performance only for monochromatic light of design wavelength. Our work focuses on polychromatic illumination, significantly expanding the practical utilization of the polarization microscope. We experimentally investigate the performance of the polarization microscope with the SWP when imaging a phase calibration target under polychromatic illumination.

In today's world, more and more emphasis is placed on non-invasive, label-free diagnostic types in order to avoid the destruction of tissue structures. One example is Flow cytometry, which allows the differentiation of single cells. In order to realize a spectrally and angularly resolved scattered light measurement setup, which allows both the differentiation of cell clusters and provides information about the cell state, a special multispectral light source in the visible/near infrared wavelength range was developed. For this purpose, single-mode fiber-coupled laser diodes of defined wavelengths are coupled into a polarization-maintaining fiber using a developed wavelength-selective coupler and an optical switch. The desired polarization is set by a polarization-maintaining fiber using paddles. A developed electronical circuit with integrated temperature control enables the selection of the wavelengths as well as the control of the laser diodes. In addition to that, the light source achieves the required modulated operation in the nanosecond range to generate short pulses of 600 ns with a peak pulse power of about 3 mW for time-resolved data acquisition. The fiber-based system can be flexibly integrated into a scattered light measurement setup, and principal component analysis was used to differentiate between the tissues of pig heart, pig liver, pig stomach, and sheep tendon based on the scattered light.

Digital holographic microscopy (DHM) combined with microfluidics enables high-throughput label-free imaging flow cytometry for acquisition of physical data from heterogeneous particle suspensions. In this study, the combination of DHM with a rectangular microfluidic channel with variable lateral hydrodynamic focusing capabilities was evaluated. We investigated the control of the focused sample stream width by setting different flow velocities as well as by variation of sample and sheath flow ratios. Moreover, the three-dimensional particle distribution in the sample flow was determined. In summary, our approach prospects flexible analysis of cell mixtures and particles like microplastics or urine components.

Probiotic bacteria are microbial species known to confer benefits to health. In order to act effectively as probiotics, microencapsulated bacteria have to maintain their viability during the gastro-intestinal transit and their motility to reach epithelial cells of the intestine. Here we use bio-speckle dynamic assays (BSDA) for rapidly testing the microencapsulation performance in experiments simulating gastro-intestinal conditions. Label-free samples are probed by coherent light to infer ensemble motility information. Then, we use Digital holography (DH) in transmission microscopy mode and 3D tracking as complementary tools to infer strain-specific locomotion profiles at the single cell level.

Recently, it has been demonstrated that nanographene oxide (nGO) particles can be exploited for drug-delivery applications thanks to their massive cellular internalization rate. Since internalized particles usually interact with intracellular organelles, modifying the life cycle of cells, the investigation of the whole cellular uptake process is an hot topic. Here we report the quantitative evaluation of the effect of nGO internalization in NIH-3T3 cells by using holographic microscopy to monitor in time the evolution of cellular uptake and the tomographic phase imaging in flow cytometry to visualize the 3D volumetric distribution of nGO within the cells cytoplasm.

In recent years, it has been discovered that lipid droplets (LDs) are involved in many pathologies, which diagnosis can be aided by detecting them within single cells. Moreover, it has been demonstrated that a suspended cell acts as a biological lens with specific focusing features. Here we show that the presence of intracellular LDs inside the cell changes its focalization features, measured through a holographic imaging flow cytometry system. The attained results open the route to the development of a fast, non-destructive, and high-throughput tool working in flow-cytometry mode for the diagnosis of LDs-related pathologies by exploiting the biolens’ signature.

We investigate a biological strategy for increasing the intracellular contrast inside epidermal onion cells to recognize their nuclei. By setting specific environmental temperature and humidity, we can induce dehydration, thus provoking the water evaporation from the vacuole and therefore increasing the nucleus-cytoplasm contrast. The reduction of the turgor pressure causes a rearrangement of the cytoskeleton, thus allowing nuclear roto-translations. We exploit an ad-hoc algorithm to estimate the nucleus rolling angles around the image plane. Then, we perform phase-contrast tomography to reconstruct the three-dimensional (3D) RI distribution of the plant cells’ nuclei by operating in complete reversible conditions.

We propose a strategy for forming free-standing thin liquid film under the monitoring of digital holography (DH): a customized iris diaphragm has been used to stretch the liquid droplet inside to a thin liquid film. Under the condition of quantitatively adjusting the opening speed and radius of the iris, the precise manufacturing of the desired thin film can be achieved. In this case, DH is implemented to provide the thickness distribution of the droplet during stretching; the real-time thickness mapping of thin film builds up a close loop controlling for fabrication process.

3D cell culture resembles tissues better than traditional monolayer cultures, which differ greatly from in vivo models. We use this technique to develop multicellular spheroids from two cell lines: MCF-7 (adenocarcinoma) and U87mg (glioblastoma astrocytoma), by forced-floating method. In this work, we research the spheroid behaviour through two optical techniques: photovoltaic tweezers and laser irradiation. We use photovoltaic tweezers to manipulate spheroids and to explore their electric charge. We also investigate their biological response to laser irradiation depending on wavelength and laser power. Finally, cell viability of the spheroids after undergoing each of these optical/optoelectric treatments has been quantified.

Cancer remains a significant global medical challenge, and the selection of effective treatment modalities is crucial for an optimistic prognosis. Photothermal therapy, being non-invasive and targeted, holds immense potential for future therapeutic developments. Due to their high biocompatibility, carbon nano-onions particles are frequently employed as photothermal materials. The investigation of the dynamic three-dimensional distribution of these nanoparticles within cancer cells is imperative for constructing an accurate photothermal conversion model. In this research, we employed digital holographic tomography to monitor the temporal changes in the three-dimensional distribution of onion-like carbon nanoparticles within colorectal cancer cells. We reconstructed the three-dimensional refractive index distribution of carbon nano-onions particles within cancer cells at different time points. Further, we quantified two morphological parameters, surface area and volume, of these nanoparticles within cancer cells and performed preliminary analysis of their temporal evolution. This methodology introduces a novel perspective to study the interaction between Carbon nano-onions particles and cancer cells, enhancing our understanding of the photothermal therapy mechanism.

Optical microscopy is an indispensable tool that is driving progress in biology and is still the only practical means of obtaining spatial and temporal resolution within living cells and tissues. In this context, staining samples with fluorescent labels provides a highly sensitive and specific method of visualizing biomolecules. However, fluorescence microscopy has various limitations including sample manipulation and staining artifacts, fluorophore photobleaching and associated phototoxicity. Therefore, much effort has been devoted to developing label-free optical microscopy techniques which are non-perturbing, photostable, and in turn offer quantitative capabilities unavailable with fluorescent methods. Our laboratory has been developing quantitative label-free optical microscopy set-ups featuring innovative excitation/detection schemes, with application ranging from synthetic lipid membranes [1] and nanoparticle materials [2-4] to living cells [5]. Specifically, we have demonstrated quantitative differential interference contrast microscopy [1], extinction microscopy [2], four-wave mixing imaging [3,4], and chemically-specific coherent Raman scattering (CRS) microscopy [5-7], including an interferometric CRS set-up which offers background-free image contrast, shot-noise limited detection, and phase sensitivity, enabling topographic imaging of interfaces [8]. We are also developing a new wide-field interferometric reflectometry method, aimed at monitoring single protein-lipid membrane interactions with unprecedented sensitivity. I will present our latest progress with these techniques and their applications to bioimaging. [1] Anal. Chem. 92, 14657 (2020). [2] Nanoscale 12, 16215 (2020). [3] Phys. Rev. X 7, 41022 (2017). [4] Nanoscale 12, 4622 (2020). [5] Analyst 146, 2277 (2021). [6] Nat. Nanotechnol. 9, 940 (2014). [7] Anal. Chem. 91, 2813 (2019). [8] APL Photonics 3, 092402 (2018).

We will describe the “digitization” of immunoassays based on the optical imaging of large numbers of single molecule labels in arrays or in flow. Digital methods have significantly improved the sensitivity of immuno-diagnostics, catalyzing improvements in the diagnosis of neurological diseases. We will describe digital bead assays (DBA)—based on the capture and labeling of proteins on beads, optically identifying “on” and “off” beads, and quantification using Poisson statistics—and other methods based on novel optical imaging and chemical labels. We will suggest how innovations in miniaturized optical and fluidic systems might drive broader adoption of digital protein detection methods.

Lysosomal storage diseases (LSDs) are a group of genetic disorders caused by defects in lysosomal function, which lead to the accumulation of undigested substrates and subsequent cell and tissue damage. Batten disease and Mucopolysaccharidosis are two neurodegenerative LSDs that have devastating consequences for affected individuals and their families. Despite significant research efforts, there is currently no cure for these diseases. Here, we present an innovative strategy to tackle LSDs that combines high-content imaging techniques with repurposing approved drugs to identify the correctors of these diseases. Our results show that several drugs that are already approved for other indications can effectively reduce lysosomal storage in different LSDs, including batten disease and Mucopolysaccharidosis. We further validated the most promising drug candidates in mouse models of these diseases and found that they can improve several disease phenotypes, including pathological storage, motor function, and neuroinflammation. In conclusion, our findings provide a promising starting point for the development of new treatments for these devastating diseases.

Human safety during space mission is one of the priorities of the international space agencies. Several efforts are in progress to study the effects induced on astronauts and in the next future on space tourists, diagnose pathological conditions and mitigate risks. There is a great demand of new reliable technologies to monitor health status of humans in space also in relation to long missions. One of the risks is connected to the exposure to ionizing radiation. Here we present a strategy based on the combination of microfluidics and stain-free imaging to monitor the effect on ionizing radiation on blood cells.

We have developed SHG microscope tools to probe all levels of collagen architecture organization in human high grade serous ovarian cancer (HSOC). We have found pronounced differences using machine learning classification of the fiber morphology as well as alterations in macro/supramolecular structural aspects through polarization analysis. We have used multiphoton excited fabrication to create SHG image-based orthogonal models that represent both the collagen morphology and stiffness of normal ovarian stroma and HGSOC. We found the fiber morphology of HGSOC promotes motility through a contact guidance mechanism and that stiffer matrix further promotes these same processes through a mechanosensitive mechanism. We have also developed a machine learning approach using generative adversarial networks (GANs) to optimize the scaffold design. Collectively, this data provides insight into disease etiology and suggests future diagnostic approaches.

Extracellular Vesicles (EV), and their subtype called exosomes, are biogenic nanoparticles released by almost any type of cell and carry a variety proteins, lipids, and nucleic acids. By transferring these biomolecules, EVs play important roles in intercellular communication, as such they are gaining increasing importance as potential biomarkers and therapeutic agents. This exciting area of research face big challenges due to EV small size, low refractive index, inherent heterogeneity and high sensitivity demand in detecting low abundant disease-specific sub-populations. Such need can be met by innovative affinity-probes and digital detection, namely capable to reach the single-molecule sensitivity. Our recent work has identified a class of membrane-sensing peptides (MSP) derived from Bradykinin protein as a novel class of molecular ligands for integrated small EV isolation and analysis. The membrane recognition and binding mechanisms are based on complementary electrostatic interactions between the peptide and the phospholipids on the outer membrane leaflet, that subsequently can lead to the insertion of hydrophobic residues into the membrane defects. Notably, small EVs present fairly distinctive lipid membrane features in the extracellular environment that could be considered as a ‘universal’ marker, alternative or complementary to traditional characteristic surface-associated proteins. Our MSP are therefore pan-specific, interspecies and interkingdom thus representing a multifarious class of ligands with additional advantages in terms of stability and synthetic versatility. Here we present the integration of MSP into different platforms for EV analysis and isolation such as microchips for SP-IRIS (Single Particle Interferometric Imaging Sensor) and beads for Single Molecule Immunoassays and their application into different workflows for liquid biopsy in urine and blood

On-chip microscopy is a major improvement toward portable, completely automated and high-throughput imaging of biological samples. Fabrication of such devices requires the integration of many different components that are not easy to manufacture with standard methods. Femtosecond laser micromachining (FLM) is rapidly becoming a major player in processing transparent materials. Being contactless, maskless, cost-effective and capable of three-dimensional (3D) structuring it proved to be a unique technology raising interest both in scientific as well as in industrial applications. The capability of FLM to directly inscribe optical waveguides in transparent materials has raised a significant interest for several applications and, in particular, for combining microfluidic channels with microlenses and optical waveguides. The former two components can be also fabricated by femtosecond laser irradiation followed by chemical etching (FLICE). This allows producing 3D microstructures of arbitrary shapes inside a glass substrate. This technology enabled the fabrication of light-sheet microscopes on an optofluidic chip for 3D reconstruction, by fluorescence imaging, of biological samples, from spheroids to single cells. In addition, by implementing structured light illumination on chip, super-resolution imaging has been demonstrated. These monolithic devices provide stable and accurate alignment of the components, avoiding any need for complicated assembly, offering ease of operation. Microfluidic delivery of the samples to the imaging area enable high-throughput and completely automated measurements. Basic concepts and experimental demonstrations will be discussed in the presentation.

Digital Holography in microscopy is an innovative imaging method with the advantages of label-free mode, a posteriori multiple refocusing and quantitative phase contrast images collection. Here, we propose to use this technology for the identification of the lysosomal compartment in normal cells and cellular models of two different lysosomal storage diseases (LSDs), Mucopolysaccharidosis IIIA (MEFs-MPS-IIA) and Niemann Pick C1 (NPC1-HeLa). Particularly, we demonstrate the application potential of the holographic method as discriminator between LSD phenotype and wild type (WT) populations of both cell lines by investigation of morphological variations and lysosomal aggregations.

We propose the combination of polyN-isopropylacrylamide (PNIPAM) particles and optical coherence tomography (OCT) to overcome the main limitations of current nanothermometry for medical purposes. We demonstrate that PNIPAM particles can behave as temperature-sensitive contrast agents in OCT thanks to their structural phase transition at 32 °C, resulting in changes in the refractive index that make their OCT contrast temperature-dependent. Simple experiments have been conducted to demonstrate the feasibility of this approach for three-dimensional imaging of phantom tissues subjected to photothermal processes. The results included in this work constitute an alternative route towards facile incorporation of nanothermometry into the clinical world.

The understanding and analyzing of solid particle behavior in a liquid is a challenge in numerous fields and engineering industry as the petroleum or the cosmetic one. It is indeed essential to know the behavior of soft matter process to avoid problems and ensures the product quality. This study presents the viscometer development working on a resonant optical signal principle by measuring the Free Spectral Range (FSR) parameter of a resonant optical mode during nanoparticles (NPs) sedimentation in a liquid which consists of a water/glycerol mixture. The photonic structure is composed of racetracks micro-resonators made of a UV210 polymer fabricated by deep-UV photolithography developed on an oxidized silicon layer to get a Si/SiO2 bi-layer. The chip is then integrated in an optical bench to track the evolution of the FSR during the complete sedimentation process. The resonant signal analyze established by an adapted signal processing of silica nanoparticles sedimentation in different water/glycerol concentrations allows us to determine stages and velocity rate of the sedimentation process to finally access to their viscosity. At the same time, measures are performed on a commercial mechanical rheometer so as to compare the dynamic evolution of their viscosity and their associated FSR. The plot of those data versus the glycerol concentration in water obviously shows a possible mathematical transformation between viscosity and FSR slope. There is therefore a good agreement between mechanical and resonant optical measures if we consider the dynamic evolution of both curves ; so this work proves the feasibility of an optical viscometer based on resonant signal.

In this work we present a phase-retrieval-based approach to quantifying the diffusion of drugs through biomembranes or biofilms. So far, the phenomenon was studied based on fringe orientation resulting from a refractive index gradient in the vicinity of the interface. The result is usually obtained with a Mach-Zehnder interferometer with an imaging system. This approach limits the spatial resolution of the method and does not allow observation of local changes in the diffusion. For this reason we propose to use a single-shot phase retrieval method by utilizing a polarization-sensitive CMOS sensor to obtain four phase-shifted interferograms in a polarization-modified version of the interferometer. Finally, we demonstrate the operation of the system by quantitative analysis of ampicillin diffusion through Pseudomonas aeruginosa biofilm formed on the polyethylene terepthtalate membrane.

The nanostructuring on titanium surfaces is studied by low-energy argon ion irradiation. The surfaces are analysed by EBSD for grain orientation mapping, SEM for surface imaging, WLI and AFM for topography characterization, XPS and ToF-SIMS for chemical surface analysis. Under normal incidence specific nanoripple structures are formed, whereas the morphology is defined by the crystallographic conditions only. A characteristic relation between grain orientation and ripple size is elaborated. Experiments on co-deposition with Al, C, Cu, Fe, and Si indicate that Fe impurities influence nanostructuring. The effect of inclined ion incidence shows an overlay between orientation-dependent and process geometry-related structure formation.

In this conference, we will discuss the advancements in Terahertz (THz) spectroscopy and imaging regarding biophotonics applications. In recent years, THz technology helped various fields, from fundamental physics to industrial process control. THz spectrometers offer a wide spectrum range and high dynamic range, enabling the analysis of semi-transparent materials and living tissues. THz imaging has shown potential in applications such as cancer detection, diabetes foot syndrome, burn wound analysis, and plant hydration assessment. Spectroscopy in the THz range is particularly promising for studying biological systems, including proteins and their compounds. However, challenges such as the strong absorption of water, temperature stability, and small sample sizes need to be overcome. Techniques like scanning microscopy, integrated photonics, and metallic broadband approaches have been explored to enhance THz analysis. Despite the challenges, THz technology has found industrial applications and efforts are being made to improve data processing and error evaluation. Additionally, a proposed method aims to enhance the resolution of THz systems, making them suitable for gas spectroscopy and the sensing of volatile organic compounds relevant to biology.

With observation of small objects, a precisely manipulation is also highly desirable, especially for a three-dimensional manipulation of nanoparticles or biomolecules with a size of less than 100 nm. Although optical tweezers have become powerful tools to manipulate microparticles and cells, they have limits when extended to the nanoscale because of the fundamental diffraction limit of light. The emergence of near-field methods, such as plasmonic tweezers and photonic crystal resonators, have enabled surpassing of the diffraction limit. However, these methods are usually used for two-dimensional manipulation and may lead to local heating effects that will damage the biological specimens. In this talk, I will introduce a near-field technique that uses a photonic nanojet, a highly focusing beam, from bio-microlenses to perform optical manipulation and imaging of sub-100-nm objects. With the photonic nanojet generated by a bio-microlens bound to an optical fiber probe, optical manipulation and super-resolution imaging were achieved for fluorescent nanoparticles, DNA molecules, subcellular structures and even viruses. Backscattering and fluorescent signals from the trapped targets were detected in real time with a strong enhancement. The demonstrated approach provides a potentially powerful tool for nanostructure assembly, biosensing and single-cell studies.

The authors introduced recently a vat photopolymerization (VP)-based bioprinting method named light sheet stereolithography (LS-SLA), and demonstrated the fabrication of centimeter-scale scaffolds with micrometer-scale features (> 13 μm) by using off-the-shelf optical compounds. This work proposes freeform optics to perform laser beam shaping in the LS-SLA device. The results show that rectangular beams are readily produced by freeform optics resulting in compact and energy efficient systems. Further considerations on the real laser output are necessary to deliver high beam uniformities. Tackling the design challenges of this work leads to energy efficient and high accuracy LS-SLA 3D printers.

We present the multimodal characterization of thin polymeric membrane by digital holography-based methods. Herein, two microscope techniques had been chosen to reveal the morphology of membranes, which are conventional off-axis digital holography (CDH) and space-time digital holography (STDH). The complementary features of the different methods allow for a bottom-up analysis of the related membranes. Meanwhile, the dynamic forming process of polymeric membrane at the air-water interface is revealed in real-time by CDH. By comparing the imaging results of different methods, the application range of different imaging methods is analyzed in detail.

Conventional microscopy is usually limited by the trade-off between field of view and resolution. To acquire highly resolved images at large fields of view, existing techniques typically record sequential images at different positions and then digitally stitch composite images. There are alternatives to this mechanical scanning procedure, such as structured illumination microscopy or Fourier ptychography that record sequential images at varying illuminations prevent mechanical scanning for high-resolution image composites. However, all of these approaches require sequential images and thus compromise speed, temporal resolution and experimental throughput. Here we present the Multi-Camera Array Microscope (MCAM), which is a microscope system that utilizes an array of many synchronized cameras, each with an individual imaging lens, for simultaneous image capture. The MCAM enables enhanced imaging capabilities and novel applications in various scientific and medical fields, by combining the images acquired from each individual camera-lens pair. The primary design of the MCAM includes 54 micro-cameras that are packed on a grid with 13.5mm center-to-center spacing. Other designs include up to 96 cameras. Beyond its obvious advantage to simultaneously image a large FOV at high resolution, the MCAM allows different configurations for different functionalities, which are presented in this talk and the respective proceedings manuscript.

Lab-on-a-Chip microfluidic devices represent an innovative and cost-effective solution in the current trend of miniaturization and simplification of imaging flow cytometry. Cell tracking is a fundamental technique for investigating a variety of biophysical processes, from intracellular dynamics to the characterization of cell motility and migration. The conventional target positioning based on holography is typically addressed by decoupling the calculation of the optical axis position and the transverse coordinates. The 2D positions of each cell are located based on the phase contrast. The axial position of the cell area is calculated by refocused external criterion in complex amplitude wavefront. Computing resources and time consumption may increase because all of frames need to be perform calculation in spatial frequency domain. To counteract the cumbersome processing time, we implement space-time self-assembly strategy on digital holograms. In this case, the horizontal positions of cells are revealed directly by morphological estimation, which is implemented on holograms. Meanwhile, the axial locations can be revealed accurately based on phase shifting paradigm. The proposed method has been proved by comparing with the conventional 3D tracking methodology. The proposed approach can be used in microfluidics to analyze objects flowing in microfluidics channels. We believe that this platform could open new perspectives for enhancing the throughput by 3D rapid volumetric imaging.

In digital holography (DH) modality for lab-on-chip applications, the cells passing through the field of view (FOV) of a microscope can be detected and analyzed even if they are flowing at different depths in a microfluidic channel. If the cells rotate while flowing along the channel, they can be probed by light beams from many different directions while they cross the holographic FOV, thus, it is possible to retrieve the 3D refractive index map of each flowing cell, i.e., a 3D phase-contrast tomogram. Since in biological samples many cells flow close to each other along the FOV, so giving the possibility of increasing the throughput of the system, it is important to establish how close the cells can be to avoid mutual disturbing effects on their rotation due to hydrodynamic interactions. Here, we investigate by means of direct numerical simulations the effects of the hydrodynamic interactions among several cells on their rotational behavior and mechanical deformation during the flow along a microfluidic channel, which are two essential aspects connected to the possibility of recovering the tomograms.

We propose a phase-retrieval process in STDH for optofluidics, which allows the quantitative phase imaging for flowing cells in different focus planes simultaneously. STDH allows the extension of the field-of view and an efficient data compression in QPI mode for long microfluidic experiments. Besides, it allows super-resolution along the scanning direction and cells’ velocimetry. Based on the proposed strategy, we show the 4D mapping of flowing cells in space-time domain; the velocities and 3D location of the cells, when they pass through the scanning position, are accurately revealed.

We evaluate the combination of a microfluidic system with a single-capture bright-field and digital holographic microscopy arrangement to achieve multimodal imaging flow cytometry for live cell analysis. The system simultaneously recovers both, bright-field and quantitative phase images (QPIs) of particles in flow from single-shot multiplexed digital holograms employing a common optical microscope. In focus QPIs of individual suspended cells in flow are achieved by object recognition procedures and holographic autofocusing, from which then biophysical cellular features like volume, refractive index, and dry mass are extracted. The platform is characterized by investigations on microspheres and evaluated for live cell analysis.

For ovarian cancer patients, paclitaxel remains to be primary chemotherapy drug. Once drug resistance is developed, it will lead to tumor progression and metastasis during chemotherapy. Many studies have shown that the development of drug resistance in cancer cells can cause morphological changes. Digital holographic microscopy is an interferometric imaging technique that can obtain 3D quantitative morphological information of label-free cells. Combining with microfluidics enables high-throughput holographic image acquisition of suspended cells. In this work, four kinds of epithelial ovarian cancer cells with different drug sensitivity, SKOV3 cells, SKOV3_Ta_2μM cells, SKOV3_Ta_8μM cells, and SKOV3_Ta_20μM cells were studied. Several machine learning algorithms were used to perform multi-classification on the extracted morphological features of four types of cells. Then, we employ the SHapley Additive exPlanations (SHAP) method to interpret the classification model. The SHAP value of each feature is calculated and sorted to obtain the important morphological features.

We discuss the use of machine learning in computational imaging for manufacturing process inspection and control. In a recent article [Qihang Zhang et al, Nat. Comm. 14:1159, 2023] we described a physics-enhanced auto-correlation based estimator (PEACE) for quantitative speckle. We derived an explicit forward relationship between the particle size distribution (PSD) and the speckle autocorrelation for particle sizes significantly larger than the wavelength (x100~x1,000). We subsequently trained a machine learning kernel to invert the autocorrelation and obtain the PSD, using the explicit forward model to reduce the number of experimentally acquired examples. In this talk, we present an expanded discussion of PEACE and its properties, including spatial and temporal sampling and accuracy, and more general applications.

The presence of microgravity and ionizing radiation during spaceflight missions causes excessive reactive oxygen species (ROS) production that contributes to oxidative cellular stress and multifunctional damage in astronauts(1) . This knowledge has underlined the importance of frequent monitoring of astronaut’s health to have early diagnoses. In this scenario, the biosensor diagnostic devices could offer the necessary analytical performance to study pathological astronaut conditions (2) . Herein, we propose an innovative biosensor for detecting highly diluted biomarkers at picogram level by using the pyro-electrohydrodynamic jet (p-jet) system (3). The detection limit of the system was confirmed using a model protein as the bovine serum albumin (BSA) by optimizing its deposition on different functionalized glass substrates through different chemical reactions starting with a manual procedure. Based on these results, the epoxy glass activated surface was chosen as the best slide for p-jet experiments. The characterization of the processes was performed through different spectroscopic techniques such as infrared-spectroscopy (IR) or confocal fluorescence. In the context of long-term human missions, our revolutionary approach could be extremely useful to monitor the astronaut health. References: (1). Gómez X, Sanon S, Zambrano K, Asquel S, Bassantes M, Morales JE, Otáñez G, Pomaquero C, Villarroel S, Zurita A, Calvache C, Celi K, Contreras T, Corrales D, Naciph MB, Peña J, Caicedo A. Key points for the development of antioxidant cocktails to prevent cellular stress and damage caused by reactive oxygen species (ROS) during manned space missions. NPJ Microgravity. 2021 Sep 23;7(1):35. doi: 10.1038/s41526-021-00162-8. PMID: 34556658; (2). Mehrotra, P., Biosensors and their applications - A review. J Oral Biol Craniofac Res 2016, 6m(2), 153-9. PMCID: PMC8460669. (3). Itri, S.; del Giudice, D.; Mugnano, M.; Tkachenko, V.; Uusitalo, S.; Kokkonen, A.; Päkkilä, I.; Ottevaere, H.; Nie, Y.; Mazzon, E.; Gugliandolo, A.; Ferraro, P.; Grilli, S., A pin- based pyro-electrohydrodynamic jet sensor for tuning the accumulation of biomolecules down to sub-picogram level detection. Sensing and Bio-Sensing Research 2022, 38, 100536.