This PDF file contains the front matter associated with SPIE Proceedings Volume 10890, including the Title Page, Copyright information, Table of Contents, Author and Conference Committee lists.

We present a label-free vibrational microscopy technique recently developed by us, which offers backgroundfree chemically-specific image contrast, shot-noise limited detection, and phase sensitivity enabling topographic imaging of interfaces. The technique features interferometric heterodyne detection of coherent anti-Stokes Raman scattering (CARS) in epi-geometry, as well as multi-modal acquisition of stimulated Raman scattering and forward-emitted CARS intensity in the same instrument. As an important biologically-relevant application, epi-detected heterodyne CARS imaging of individual lipid bilayers is demonstrated. We show that we can resolve a single lipid bilayer, distinct from a double bilayer, and measure the phase of its susceptibility, which provides information about the topography of the bilayer with nanometer resolution. As an additional application example, we show imaging of silicon oil droplets surrounded by an aqueous environment at the glass-water interface, where three different signal generation pathways are distinguished. Our epi-detected heterodyne CARS microscope setup thus paves the way to exciting new experiments pushing the sensitivity and resolution limits of vibrational microscopy to the nanoscale.

Many promising techniques proposed to monitor gamete developmental potential and quality are invasive and not realistically useful in clinical practise. Hence, there is increasing interest in the development of non-invasive imaging methods that can be applied to mammalian eggs and early embryos. Recent studies have shown that mammalian oocyte and embryo viability are closely associated with their metabolic profile, relying entirely on mitochondria as a source of ATP. Fatty acids, stored in intracellular lipid droplets, are an important source of ATP. We have recently demonstrated the use of Coherent Anti-stokes Raman Scattering (CARS) microscopy to allow chemically-specific, label-free imaging of lipid droplets, with high three-dimensional spatial resolution. Here, we summarize our main findings when using CARS to examine the number, size, and 3D spatial distribution of lipid droplets in mouse eggs and early embryos. Quantitative analysis showed statistically significant differences during oocyte maturation and early embryo development. Notably, CARS imaging did not compromise maturation or development. In mouse oocytes that had been subjected to alterations in mitochondrial metabolism we found that the spatial distribution pattern of lipid droplets was also altered. In addition, differences in the chemical composition of lipid droplets in living oocytes matured in media supplemented with different saturated and unsaturated fatty acids were detected using CARS hyperspectral imaging. We also imaged bovine oocytes, and found that lipid droplets appear to be larger and with less spatial aggregation than in mouse oocytes, possibly reflecting the fact that different species metabolise lipids differently. These data suggest that CARS microscopy is a promising non-invasive technique for assessing specific aspects of the metabolic profile of living mammalian eggs and early embryos, which could be potentially linked to their quality and viability.

Coherent anti-Stokes Raman scattering (CARS) microscopy utilises intrinsic vibrational resonances of molecules to drive inelastic scattering of light, and thus eradicates the need for exogenous fluorescent labelling, whilst providing high-resolution three-dimensional images with chemical specificity. Replacement of hydrogen atoms with deuterium presents a labelling strategy that introduces minimal change to compound structure yet is compatible with CARS due to an induced down-shift of the CH2 peak into a region of the Raman spectrum which does not contain contributions from other chemical species, thus giving contrast against other cellular components.

We present our work using deuterated oleic acid to optimise setup of an in-house-developed multimodal, multiphoton, laser-scanning microscope for precise identification of carbon-deuterium-associated peaks within the silent region of the Raman spectrum. Application of the data analysis procedure, factorisation into susceptibilities and concentrations of chemical components (FSC3), enables the identification and quantitative spatial resolution of specific deuterated chemical components within a hyperspectral CARS image. Full hyperspectral CARS datasets were acquired from HeLa cells incubated with either deuterated or non-deuterated oleic acid, and subsequent FSC3 analysis enabled identification of the intracellular location of the exogenously applied deuterated lipid against the chemical background of the cell. Through application of FSC3 analysis, deuterium-labelling may provide a powerful technique for imaging small molecules which are poorly suited to conventional fluorescence techniques.

Raman spectroscopic imaging can provide three-dimensional data set of samples, including two-dimensional spatial image and one-dimensional Raman spectral data. Currently, three strategies can be used to achieve Raman spectroscopic imaging, including point scanning, line scanning, and wide-field illumination. Point scanning method provides the best resolution but has low imaging speed. On the contrary, wide-field illumination can image fast but provides lower spatial resolution. To integrate the advantages of two methods, a new strategy for large-field Raman spectroscopic imaging was proposed, which uses the frequency modulation based spatially encoded light as the excitation. In this method, millions of single beams simultaneously illuminate on the sample to act as the wide-field illumination. Each beam illuminates on different positions of the sample, whose intensity are modulated with different frequencies. Thus, each excitation beam has its own modulation frequency and the excited Raman signal will carry the modulation information. At the detection end, a single point detector was used to collect the time series Raman signals carrying the unique modulation information. Using the sparse reconstruction based on demodulation strategy, the Raman image can be recovered effectively. The feasibility of the method was verified with numerical simulations. The results showed that it is feasible to conduct Raman spectroscopic imaging with high-resolution and high speed under the illumination of frequency modulation based spatially encoded light and the detection of single-point detector.

Raman tomography can provide quantitative distribution of chemicals in a three-dimensional volume with a non-invasive and label-free manner. In view of the problems of existing data collection strategy, a frequency modulation and spatial encoding based Raman tomography was proposed, which aims to improve the data collection scheme and reduce the data collection time. In this scheme, the laser beam was divided into several sub-beams to use as multipoint excitation light sources. These sub-beams were first modulated with different frequencies and then incident on the different points of sample surface simultaneously. Because the excited Raman signals would carry such modulation information, the Raman signals from which excitation position can be distinguished with the demodulation process. In detection end, the Raman scattering light first passed through a spatial-encoding mask and then was directed to the single photomultiplier tube. By changing the pattern of the mask and then performing recovery with sparse reconstruction, the distribution of the Raman signals on the sample surface can be obtained based on compressive sensing theory. Preliminary results showed that our scheme can recover the Raman images to the certain extent with a better signal-to-noise ratio, demonstrating the proposed scheme is feasible.

Flow cytometry is the standard tool for blood analysis which generates information gathered from the interaction of lasers with cell flowing in a stream to classify them based on their size, granularity and fluorescent emission from biomarkers used as labels. However, for many emerging applications the use of labels is undesirable because they alter the cell behavior through activation or inhibition of cellular functions and hinder downstream genetic studies. We have previously described an ultrahigh-throughput label-free imaging flow cytometer that analyzes cells using their biophysical features. Label-free ultrafast imaging in flow is implemented by photonic time stretch and the trade-off between sensitivity and speed is mitigated by using amplified time-stretch dispersive Fourier transform, a technique that was originally developed to enable realtime analog-to-digital data conversion with femtoseconds sampling resolution. In the application to time stretch to imaging, cells are illuminated by spatially dispersed broadband pulses, and the spatial features of the target are encoded into the short pulse spectrum. Both phase and intensity images are simultaneously captured, and this provides ample features such as the concentration of proteins, optical loss, and cell morphology, which are then used by a neural network to classify cells. However, image processing needed to extract these features from label-free images takes time and renders this technique unsuitable for cell sorting where decisions must be made in realtime before cells exist the fluidic stream. To eliminate this predicament we have a deep convolutional neural network that directly processes the raw time stretch waveforms from the imaging flow cytometer. Eliminating the requirement of an image processing pipeline prior to the classifier, the running time of cell analysis can be reduced significantly, and cell sorting decisions can be made in less than a millisecond, orders of magnitude faster than the previous state of the art.

Degradation of cartilage, occurring in osteoarthritis and other conditions leads to pain and reduced mobility. Current treatments beyond anti-inflammatories include intra-articular injections of hyaluronan or preparations based on adult mesenchymal stem cells (MSC), the latter shown to aid cartilage regeneration which requires the assessment of the cartilage, best on a molecular level and in a minimally invasive way. However, the conventional methods are invasive, destroying and can only provide a snapshot of a tissue structure and functional state on a sample-by-sample basis, while the continuous monitoring and high-throughput assays require low-invasive biopsy-free approach. As a first step to address this problem, in current work, we explored the potential of label-free multispectral imaging of endogenous tissue fluorescence to characterise the molecular composition, structure and functional status of ex vivo healthy bovine and osteoarthritic (OA) human knee articular cartilage followed by monitoring the effects of experimental treatment of OA cartilage performed ex vivo. However, strong autofluorescence of collagens (especially from collagen type II, which is the structural backbone of collagen fibrils) from various cartilage layers presents a challenge, because this signal tends to overpower the fluorescence from chondrocytes. We have managed to use Robust Dependent Component Analysis (RoDECA) to observe the detailed metabolic information with a proper account of intrinsic cellular heterogeneity, which signifies the sophisticated quantitative biochemical analysis. This work reports on the “signatures” of the healthy articular cartilage for superficial and transitional layer, define the “healthy range” of each fluorophore’s abundance and localization of chondrocytes non-invasively as well as identify the changes of the signatures in OA cartilage of real patients and observe the reaction of the OA cartilage on 2 types of experimental treatments.

Understanding light propagation in fibrous tissue plays a fundamental role in the development of novel and minimally invasive diagnosis techniques. For this purpose, we have developed a polarimetric microscope that operates in the backscattering geometry. Our apparatus has been thoroughly calibrated and verified with experiments and Monte Carlo simulations on well characterized colloidal suspensions. In this study, we have investigated the feasibility of retrieving structural information on multiply scattering, fibrous electrospun scaffolds fabricated of Poly(vinylidene fluoride-co-hexafluoropropylene) (PVDFhfp) nanofibers having diameters ranging from 500 to 1000 nm. These nanofibers display various degrees of structural alignment and the structural anisotropy manifests itself in optical birefringence. We probed these scaffolds with a focused near-infrared light beam at three pairs of cross-polarized states and recorded images of the Stokes vector elements of the light backscattered at the surfaces of the scaffolds. Our results demonstrate that it is possible to structurally differentiate the scaffolds by analyzing the spatial variations of the Stokes vectors/polarization ellipses as a function of the polarization state of the probing beam. Visualizing the rate of retardance induced by the birefringent fibers together with the distribution of the degree of polarization unveils the orientation of these fibers and their respective degree of organization, which was compared to results obtained by small angle x-ray scattering (SAXS). This study contributes to a better understanding of the interaction of the light with multiply scattering fibrous matter such as tissue, which is particularly challenging in the backscattering geometry but fundamental to make the diagnosis of cancer possible.

Automated and unbiased methods of non-invasive cell monitoring able to deal with complex biological heterogeneity are fundamentally important for biological research and medical diagnostics. Label-free cell imaging provides information about endogenous autofluorescent metabolites, enzymes and cofactors in cells. However, extracting high content information from autofluorescence imaging has been hitherto impossible. Here, we developed a multispectral fluorescence imaging technique which allows precise quantification of the native fluorophores in cells and tissues. With that approach we are now able to non-invasively image the aspects of biomolecular composition of cells and tissues; where many of these fluorophores (NADH, flavins, cytochrome C) are relevant to metabolism. We will discuss label-free detection of reactive oxygen species (ROS) and the cell cycle. Cell cycle and metabolism have a tight, bidirectional relationship, with the ability of the cell to commit to growth depending on the availability of metabolites, and the molecular mechanisms of the cell-cycle being linked to the regulation of metabolic networks. Cells entering the cell cycle increase glycolysis as they go from G1-phase into S-phase, this results in accumulation of the NADH relative to FAD which is also fluorescent. Moreover, metabolic dysregulation is common across the spectrum of diseases, this next-generation methodology is able to detect major health conditions including neurodegeneration and cancer. This work also reports on approaches for early diagnosis of motor neurone disease (MND) and localisation of cancer margins for ocular surface squamous neoplasia. Our optimal discrimination approach (extracted features for treatment monitoring in MND and melanoma) enables statistical hypothesis testing and intuitive visualisations where previously undetectable differences become clearly apparent.

While fluorescence image-guided surgery offers improved treatment outcomes for patients with cancer by permitting the identification of tumors during resection, it has been plagued by slow translation into clinical practice due to the lengthy and costly approval process for fluorescent molecular markers. Label-free approaches to image-guided surgery provide an alternative by discriminating between cancerous and noncancerous tissue based on differences in spectral reflectance and autofluorescence between the tumor microenvironment and the surrounding anatomy. Unfortunately, state-of-the-art hyperspectral imaging systems capable of monitoring spectral differences across the entire surgical site utilize complex optomechanical architectures that contribute to low image resolutions, low frame rates, and co-registration error that cannot be calibrated, making these instruments impractical during demanding surgical workflows. To provide label-free surgical guidance while addressing limitations with existing systems, we have developed a single-chip snapshot hyperspectral imaging system that provides 27 spectral bands from ~450 nm to ~750 nm. By monolithically integrating a stacked photodiode image sensor with pixelated interference filters, we have produced a highly compact imaging system that achieves a resolution of 1252-by-852 pixels at a rate of 17.2 frames per second while avoiding co-registration error. The system provides a signal-to-noise ratio of ~55 dB and a dynamic range of ~62 dB, and it can enable spectral discrimination under standard broadband surgical light sources. Preclinical images of human prostate tumor implants in a murine model have been examined and presented to demonstrate that the imaging system can differentiate between cancerous and noncancerous tissue and can discriminate between distinct cancer types.

New forms of cancer cell identification coupled with faster detection and better accuracy may enhance diagnostic capabilities. The purpose of this study is to improve recognition of minimal residual disease from peripheral blood samples. Spectral images are generated by optical microscopy using filtered broadband visible light elastically scattered from human blood and cancer cells. Exogenous tags, like CD markers may introduce a label bias and are not required. A training cell may be validated without detailed knowledge of intra-cellular spectra used to classify random cells. Spectral object classification is scalable to any number of cell types. Small samples of erythrocytes, leukocytes, Jurkat cancer and non-small lung cell adenocarcinoma are accurately classified and associated with unique spatial-spectral characteristics.

Despite its wide-spread use, the success rate of assisted reproductive technologies including IVF is less than 20% in Australia/New Zealand. Most early human embryos are mosaic for chromosome abnormalities, containing a proportion of normal and abnormal cells. The most common chromosomal abnormality is aneuploidy: incorrect number of chromosomes. This form of mosaicism is thought account for early pregnancy loss in IVF. Current single cell biopsies of embryos are not diagnostic for the proportion of cells that are aneuploid (degree of heterogeneity/mosaicism). Thus, development of a non-invasive tool to determine the proportion of aneuploid cells facilitating segregation of embryos with a low percentage of aneuploid cells would likely improve IVF success rates. In other cells, including cancer cells, aneuploidy results in altered cellular metabolism. In this study we utilised hyperspectral imaging as a means of non-invasively measuring cellular metabolism in the early embryo. We utilised a mouse model where we manipulated the ratio of aneuploid:normal cells. Aneuploid embryos were generated by treatment during division from 4 to 8 cells using a reversible spindle assembly check point inhibitor, reversine. Eight-cell aneuploid embryos were dissociated and joined with control/normal cells to generate 1:1 aneuploid:normal chimeras. Hyperspectral imaging of 1:1 chimeric embryos had a distinct spectral profile that varied dramatically from the control/normal embryos. Interestingly, entirely aneuploid embryos showed a spectral profile dissimilar from both normal and chimeric embryos. These data show hyperspectral imaging is capable of distinguishing between embryos with varying degrees of aneuploidy making it a promising tool in assessing embryo health.

Pulmonary X-ray imaging together with pulse oximetry are harmful and invasive techniques used to monitor and diagnose the clinical course of lung dysfunction in preterm born infants which most of the cases suffer Respiratory Distress Syndrome (RDS) [1]. Biophotonics@Tyndall is exploring Gas in Scattering Media Absorption Spectroscopy (GASMAS) [2] as a novel non-invasive technique to measure continuously absolute lung oxygen volume and concentration. This could assist and improve the assessment of lung function in neonates [3].

In this paper, we present results of bench-top measurements carried out in the preclinical phase of GASMAS studies. We start with a detailed explanation of the manufacturing process of multi-structure thorax phantoms with realistic geometry based on organ segmentation from anonymized DICOM images of neonates. After segmentation, the organs are 3D printed and used to create negative rubber molds. The tissue optical properties of heart, bone and muscle are assigned by mixing the silicone matrix with different concentrations of absorbers and scatters, the lung is kept as a gas content cavity and the thorax phantom is build up by placing all organs inside out immersed in the muscle structure.

The phantoms are used for quality control and validation of the system performance [4]. Oxygen gas absorption imprints are measured for different light source-detector remittance configurations and the results are used to define the potential and limitations of the GASMAS technology in the development of a bed-side clinical device.

Probiotics are microbial species that have been demonstrated to confer benefits to health. In recent years, the use of probiotics in food and health has increased enormously. A sufficient concentration of probiotics in the intestine acts against pathologies such as obesity, diabetes, etc. However, if probiotics are not able to maintain their viability during their transit through the gastro-intestinal apparatus, they cannot act to enhance the immune system. Hence, protection and preservation of probiotics are essential to both food industry and in pharmaceutics. Microencapsulation is one of the most common methods of preservation, applicable to several biological matrices, including probiotics. Whenever food products or pharmaceutical formulations contain microencapsulated probiotics, it is important to quantify the effectiveness of micro-encapsulation as a microbial protection system over the time, e.g. during the shelf life of a functional product containing encapsulated probiotics, conserved in the supermarket, and during gastro-intestinal transit. Here we use bio-speckle decorrelation as a tool for the rapid assessment of microencapsulation effectiveness. Although speckles are often thought as a noise to get rid of, they represent a precious source of information, increasing the sensitivity of image sensors based on coherent illumination. Such information is exploitable to characterize bacterial dynamics in a fast and simple way suitable for applications in food science and industry. Through bio-speckle decorrelation, we quantify the shelf-time of alginate-encapsulated Lactobacillus rhamnosus and Lactobacillus plantarum probiotic bacteria and their survival rate under simulated gastro-intestinal conditions.

As widely known, random diffraction effect due to refractive index inhomogeneity is considered as an annoying factor for propagating light through scattering media. Here, instead of overcoming the random diffraction, we utilize the natural randomness of strongly scattering media and their sensitivity to inhomogeneity of refractive index to develop a novel optical sensor. Unlike various sensing technologies with trade-off among complexity, sensitivity and stability, here we demonstrate a very simple sensing technique which uses scattering media to achieve super sensitivity, speedy response time and possibly high stability. In our sensing principle, a lasing beam passing through a turbid medium creates a speckle pattern on a camera due to interference of random refracted light off the scattering media. Light is scattered multiple times at multiple interfaces between ground glass and the surrounding environment creating a speckle pattern which is sensitive to the environment’s refractive index. The correlation of speckle patterns indicates the change of refractive index around the scattering medium. Simply placing the rough surface of ground glass in contact with sensing solutions, we are able to measure glucose, or sodium chloride concentration with sensitivity in the order of micro grams per liter. More interestingly, the sensitivity of the proposed approach could be improved simply by adding more scattering surfaces in contact with the target medium. Therefore, our simple technique could be very useful for prominent applications in refractive index sensing such as measuring solution concentration, distinguishing different gases, detecting pressure change and so on.

Stimulated Raman scattering spectroscopic optical coherence tomography (SRS-SOCT) is a recently developed molecular imaging technique that leverages the spatial and spectral multiplexing capabilities of SOCT with the molecular sensitivity and specificity of SRS for fast, tomographic, label-free molecular imaging. Here we present a theoretical framework for SRS-SOCT, discuss its advantages and limitations, and show experimental results from excised human adipose tissue. We also discuss our recent efforts in developing a novel light source, based on a coherently broadened laser, that is specifically tailored for SRS-SOCT. Implications of this label-free, molecular imaging technique for biomedical applications are discussed.

We present high-speed multicolor coherent Raman imaging (CRI) with rapid wavelength tuning within only 5 ms between successive images, enabled by a novel fiber optical parametric oscillator (FOPO). In CRI the limited tuning speed of conventional laser systems (at least several seconds per wavelength change) hinders the rapid acquisition of successive images at multiple vibrational frequencies and is a bottleneck for fast assessments of medical specimens or rapidly evolving samples. The output pulses of the presented FOPO are tunable across the vibrational spectrum between 865 and 3550 wavenumbers within only 5 ms. Therewith, the wavelength can be tuned in a frame-by-frame manner adequate for an image acquisition with up to 200 frames/s. Tuning of the FOPO is achieved by synchronized changes of both wavelength and repetition rate of the pump pulses, allowed by a novel fiber-integrated pump laser working without a mechanical delay. Based on this tuning mechanism, the FOPO resonator can be composed of all-spliced fiber components. Compared to previously presented FOPOs, the system exhibits short pulse durations of 7 ps and operates at a high repetition rate of 40 MHz to allow short pixel dwell times as low as 25 ns. The pump (<500 mW) and FOPO pulses (up to 200mW) exhibit equal durations and bandwidths below 12/cm. In comparison to previous approaches on multicolor CRI with multiple oscillators, requiring complex setups for synchronization, parallel laser amplifiers with limited emission bandwidth, or fragile mechanical delay lines, the realized FOPO system will dramatically simplify and improve CRI setups.

High resolution volumetric stimulated Raman scattering (V-SRS) imaging allows a precise measurement of chemical distribution in a three-dimensional (3-D) complex biological system. To compile a stack of multiplane images, current methods such as using piezo objective positioners or tunable lenses either yield low scanning speed, disturbance of specimen, or significant aberrations. Here, we develop a V-SRS microscope with a high-speed MEMS deformable mirror (DM) which has 140 actuators and a frame rate of 20 kHz using hardware-trigger. The DM conjugated with the objective pupil plane enables wavefront shaping at reflectance mode and remote focusing of both pump and Stokes beams on the sample. The depth scan range can reach tens of micrometers by using 40X and 25X objectives. Multiple 3-D cancerous cell images are obtained. We expect the V-SRS to have great potential to enable label-free studies of cell metabolism, brain function, and developmental biology.

Stimulated Raman scattering (SRS) microscopy has been increasingly used in biology and medicine in recent years due to its ability to image chemical bonds without labelling. Traditional SRS imaging uses Gaussian beams as the excitation sources, which can achieve high spatial and axial resolutions because of the tight focus of the Gaussian beam. However, the tight focus poses serious problems for observing the scattering media. The Gaussian beam would defocus after propagating through a small distance in scattering media. The SRS microscopy cannot work well in this case. Having the self-healing property, Bessel beams may bring solution to this problem. In our previous work, we applied the Bessel beams to the SRS and implemented three-dimensional SRS imaging with projection concept. Here, we simulated the propagation of Bessel beams and the generation of SRS signals with the beam propagation method (BPM). By adding glass beads on the beam propagation path to simulate scatters, the propagation of the Bessel beams and the generation of the SRS signals would change. We designed a series of simulations to investigate the influence of the size and position of the added glass beads to the generation of SRS signals. Simulation results demonstrated that the SRS signals can generate or be recovered at the certain depth in scattering media.

The development of a technology that allows for analyzing microscopic spatial distribution and dynamics of small gaseous molecules such as inhalational anesthetics and odors would advance our understanding of its biological activities in living cells. However, direct observation of such small molecules by optical microscopy is still challenging. We propose a new pump–probe stimulated Raman scattering (SRS) microscopy method for studying the localization, transport and metabolism of gaseous molecules in a living organism in a label-free manner. A technical challenge is how to detect the Raman signal of a small amount of drug molecules that is typically overwhelmed by unwanted nonlinear background, including nonresonant background and coherent Raman scattering of surrounding cells and tissues. In particular, the latter Raman-induced background is essentially inevitable in most standard coherent anti-Stokes Raman scattering (CARS) and SRS systems. We show that these background issues can be overcome by introducing a new pump–probe, time-resolved SRS detection approach coupled with a pair of spectrally-focused, asymmetrically shaped probe pulses (T. Ito et al. APL Photonics (2018)). In the pump–probe scheme, a long-lived vibration of the targeted molecules can be efficiently probed after short-lived vibrations of other background molecules such as water and fatty acids become silent. This unique lifetime-selective signal detection provides a significantly enhanced vibrational signal contrast. As a proof-of-concept experiment, we demonstrate that the passive transport of inhalational anesthetic molecules from aqueous solution to adipose cells can be monitored by time-lapse SRS imaging.

We report a novel concept of an advanced hyperspectral imaging system based on acousto-optical tunable filters (AOTFs) with optional phase imaging modality. Visualization of phase objects is performed using an additional adjustable liquid crystal amplitude mask. The mask shape is matched with the two-dimensional transfer function of the AOTF enabling bandpass spatial filtering. Label-free phase object visualization is demonstrated with unstained histological sections using diascopic incoherent illumination of a standard inverted light microscope.

In this work, we report a cost-effective supercontinuum (SC) laser-based multispectral photoacoustic spectroscopy (MSPAS) system for studying spectral characteristics of various analytes. We demonstrate an in vitro label-free monitoring of the analytes in the extended near-infrared (NIR) (1540–1840 nm) spectral range. We further demonstrate how a simple ratiometric analysis in conjunction with linear regression can be used for accurate prediction of glucose over commonly encountered physiological levels inside the human body (0–400 mg/dL). Looking forward, the proposed SC-based MSPAS system provides a framework for the development of label-free and non-invasive monitoring multiple bio-analytes accurately, with potential translation to clinical in vivo applications.

Now, a label-free biosensor has been defined as “to detect a whole biologically active molecule in real time”. Also, the specificity and sensitivity common to all assays must apply. To be precise, to be fast, to be small, to be integrated and less cost….. Is it possible? In this talk, several results for improving the sensitivity, the specificity and the dynamic range of a label-free optical fiber biosensor will be discussed. First, a tapered optical fiber refractometer and biosensor modifying their unclad region with the coating of SiO2 and gold nanospheres will be compared to show that the combination of nanosphere-induced absorption and scattering losses and multimode propagation of tapered fiber gives a way to increase its detection range. Second, the results of ultrasensitive label-free detection of cardiac troponin I with optical microfiber coupler is shown when it works around the turning point. The detection limit with a concentration as low as 2fg/mL can be achieved using the proposed immunosensor. This kind of immunosensor has great application potential for biomarkers detection due to its characteristics as simple scheme, fast response and ease to miniaturation.Finally, the Fano and some other extraordinary effects by tapered optical fiber coupled WGMs will be discussed. The functions of several special modes may help to take the challenges of selectivity for plasma detection.

Utilization of frequency locking for tracking optical resonances of microresonators is an emergent technique, which has relevance in label-free biosensing, owing to its extremely high sensitivity in detecting adsorption on a microcavity’s surface. In this study, we demonstrate the capabilities of a technique known as FLOWER (Frequency Locked Optical Whispering Evanescent Resonator) in a real-life problem: the detection of human chorionic gonadotropin (hCG) in urine samples. hCG, besides being a hormone secreted during pregnancy, is a performance enhancing agent prone to be abused by athletes, and is routinely investigated in doping laboratories. The gold standard in hCG detection is mass spectrometry. As the lowest limit of detection by mass spectrometry is close to the cut-off value for a positive result, various other measurement techniques such as optical interferometry, photoluminescence, or electrochemical sensors have been used to try to enhance measurement sensitivity. These methods, however, do not show a significant improvement in the limit of detection over mass spectrometry and they mostly have a narrow detection range. In addition, measurement selectivity is another issue. Here, we use microtoroidal optical resonators functionalized with antibodies against hCG, and track their resonances using frequency locking upon analyte infusion. Urine samples from pregnant women were measured using our setup. Our results demonstrate that FLOWER can detect the presence of hCG in a large concentration range from 1 fM to 10 nM.

Sensitive detection of specific chemicals on site can be extremely powerful in many fields. Owing to its molecular fingerprinting capability, surface-enhanced Raman scattering has been one of the technological contenders. In this paper, we describe the novel use of DNA topological nanostructure on nanoporous gold disk array (NPGDA) chip for chemical sensing. NPGDA features large surface area and high-density plasmonic field enhancement known as “hot-spots”. Hence, NPGDA chip has found many applications in nanoplasmonic sensor development. This technique can provide novel label-free molecular sensing capability with high sensitivity and specificity. In this paper, we introduce a new concept of multimodal signal amplification by exploring the synergy of catalytic multiplication and plasmonic intensification.

UV plasmonic materials such as aluminum and magnesium have drawn attention in recent years due to their potential applications in plasmonic enhanced label free sensing and UV light sources and detectors. In this paper, nanoapertures in Al and Mg thin film have been used to study the lifetime modification and net fluorescence intensity of the freely diffusing UV dye p-terphenyl. Previously, Jiao et al reported lifetime reduction of 3.5 times for Al nanoapertures with diameter 60nm. In this paper, lifetime reduction of 5.4 times is observed for Al nanoapertures with smaller diameter 40nm. For the firs time, lifetime modification in Mg nanoapertures is reported. For the same aperture geometry, Mg nanoaperture show higher Purcell factor and excitation enhancement due to lower loss of Mg in UV range . We observe 7.3 times lifetime reduction for Mg nanoapertures with 40nm diameter. We also studied numerically bowtie antenna for native fluorescent enhancement and have observed ~180x net enhancement, ~35x radiative rate enhancement. Those results suggest that label free sensing of amino acids and proteins are promising.

Detection of pre-cancerous (dysplastic) tissues in the clinic remains an ongoing challenge. Current methods often rely on systematic biopsies with no visual cues to guide the physician to areas likely to harbor dysplasia. We have developed novel spectroscopic techniques for assessing cell structure and diagnosing disease based on combining interferometry with light scattering. Angle-resolved low coherence interferometry (a/LCI) combines the depth gating of coherence imaging, as used in optical coherence tomography, with the high resolution structural information that can be obtained with light scattering spectroscopy. The a/LCI approach has been validated with in vitro measurements of phantoms and in vitro cells to recover sub-cellular structure with sub-wavelength accuracy and precision. Discrimination of pre-cancerous tissue states was initially demonstrated using animal models and ex vivo human tissues, by using light scattering and interferometry to measure the size of cell nuclei in selected tissue layers. We will discuss various light scattering models used in a/LCI, including Mie theory, T-matrix, and analysis methods based on wavelet transforms and machine learning, as they relate to their application to this clinical problem. To implement a/LCI for detection of pre-cancerous tissues in patients, we have developed a clinical a/LCI system, including a portable optical engine, and an endoscope compatible fiber probe for in vivo measurements. The capabilities of a/LCI will be shown with results from clinical studies to detect precancerous cells in both the esophagus and cervix during routine clinical examination. Further directions to enhance clinical utility and ergonomic ease of use will also be discussed.

Limited visualization of semi-transparent structures in the eye remains a critical barrier to improving clinical outcomes and developing novel surgical techniques. While increases in imaging speed has enabled intraoperative optical coherence tomography (iOCT) imaging of surgical dynamics, several critical barriers to clinical adoption remain. Our previous implementation of surgical microscope-integrated multimodal imaging technologies to address these limitations lacked the resolution and optical throughput for in vivo retinal imaging. Here, we present an optimized intraoperative spectrally encoded coherence tomography and reflectometry (iSECTR) system and demonstrate in vivo multimodality ophthalmic imaging. The ophthalmic surgical microscope objective was optically characterized to develop a lens model for system design. We estimated the glass thicknesses, refractive index, and intermediate curvature of the doublet objective lens using a novel combination of OCT, focus tracking, and computational ray tracing. The resulting lens model and corresponding optical aberrations were used to optimize the optical design for an iSECTR scan-head. In vivo ophthalmic imaging using iSECTR was performed on a healthy volunteer. En face spectrally encoded reflectometry (SER) and cross-sectional OCT images were acquired at 350 fps with 2560 x 1000 pix. (spectral x lateral). OCT volumes of 1000 B-scans were acquired in 2.86 s. Retinal and choroidal vasculature were readily visible on SER and provided complementary contrast to the OCT volume projection. We presented optimized designs for multimodal ophthalmic imaging with surgical microscope-integrated iSECTR. Clinical translation of iSECTR will benefit real-time instrument and FOV tracking for imaging of surgical dynamics and image-guided ophthalmic surgery.

In the absence of suitable label-free techniques many groups have developed and used labels such as fluorophores and nanoparticles for biological discovery and in vitro diagnostics. This has led to extraordinary advances in our understanding of fundamental biological processes. However, they are typically toxic and may interfere with the biological process being studied. Furthermore, since they can typically not be used in humans, a disconnect arises between biological discovery and clinical translation. Many researchers worldwide are working to address this problem. For example, we are developing label-free methods for structure and functional information in biological systems including animals and humans. Label-free methods offer many advantages including simpler protocols less ethical dilemmas, less opportunity for interference with the biological process and faster translation to the clinic. Optical coherence tomography has had the fastest uptake of any medical imaging modality in the history of medicine. We have developed technologies based on OCT to visualise the microcirculation to sense structure and structural change at the nanoscale and to provide super resolution in a simple and cost-effective manner. Extracting this information from the existing OCT signal allows us to add structural and functional information which can be overlaid on the basic OCT image. A significant advantage of such methods is that the new information is intrinsically coregistered with the OCT data. In this paper we will show several examples from our own work and label free imaging and sensing with OCT. Correlation mapping OCT relies on speckle dynamics due to moving blood cells which cause a decorrelation with previous or adjacent frames which can be exploited to efficiently generate microcirculation maps. Sub-voxel sensing is very exciting avenue in OCT research since the OCT signal contains much more information than the basic structure provided in a typical OCT image. For example structures smaller than the resolution of the system impact on the spectrum which is detected and since we use a broad spectrum for probing with OCT we can with the help of a high resolution spectrometer determine the sizes of the structures within the voxel. We can furthermore see how they change over time. We can do this very fast, more robustly with better noise performance because the information can be obtained from a single frame unlike phase based imaging. Nanosensitive OCT (nsOCT) is rather different to other methods in that it relies on the fact that the structural size is encoded in the spectrum detected. With appropriate equipment and protocols this can provide nanometres sensitivity to structures and structural changes. A novel approach to superresolution will also be presented where separation of voxels can be achieved by using the intensity or other variations in depth. Since we have already got this information in many microscopy and tomography techniques including OCT and confocal microscopy, we can use it to efficiently and cost-effectively provide superresolution. The superresolution can be achieved without breaking the rules of information theory because the extra information is in depth and can pass through the objective lens without being blocked by the diffractive limit. We will further discuss the applications of these techniques including an example of nanosensitive OCT being used to diagnose ottitis media in children.

Optical coherence tomography relies on the coherence of light that is backscattered from the sample with interference playing the important role of gating to allow an accurate determination of the origin of any retrieved signal. OCT images of highly-scattering samples suffer from multiple-scattering of light, which leads to speckle as an important noise contribution and dominates at higher depths obscuring information from deep within the sample. We present an approach a simple geometry, called spatially-offset optical coherence tomography (SO-OCT), allowing singly scattered photons from depth to be retrieved whilst suppressing the presence of multiply scattered photons (speckle). Therefore, it improves signal detection to enhance the image contrast at depth in the presence of strong scattering samples. An approximately two-fold enhancement in image contrast at depths can be observed in different biological samples, such as zebrafish and krill. This approach solely requires the translation of a single lens in the experimental OCT arrangement to achieve enhancement on both image quality and depth penetration.

Label-free methods provide image contrast without the need to introduce exogenous material that may alter the sample microstructure, perturb the microenvironment, or limit clinical translation. In addition to being label-free, it is increasingly important to consider modalities that work in reflection mode geometry to allow imaging intact tissue (in vivo or ex vivo) or 3D cell cultures too thick to image in transmission. However, reflection geometry precludes many useful contrast modalities, including transmitted phase and darkfield. Optical scattering in tissue limits contrast and depth of imaging, but scattering can also be exploited to quantify tissue structure and provide diagnostic markers. Here, we present another application of multiple scattering in tissue in which collection of scattered light is used to reconstruct images with contrast comparable to transmission imaging modalities. A Scattered Light Contrast (SLC) microscope collects a descanned image of the scattered light surrounding the confocal illumination point. The illumination point is raster scanned and the scattered light distribution is collected for each image pixel, providing 4D data. By analyzing the scattered light, a range of image contrast modalities can be reconstructed including reflectance confocal microscopy, transmitted phase contrast, and dark field microscopy. Computational Monte Carlo simulations relate the SLC signal to the phase gradient in the image plane and can be used to reconstruct the phase gradient of transmitted light in the focal plane. This modality enables label-free imaging of cells and structures in the retina that will advance diagnosis and monitoring therapies at the cellular level.

Mueller polarimetric imaging in dark-field observation shows a contrast enhancement between healthy and cancerous human colon tissue in some reports. We have developed a Mueller-matrix microscope system that combines a dark-field polarization illuminator with an imaging polarimeter to measure the polarization characteristics of scattered light from human colon tissue samples. A multichannel light source permits the acquisition of multispectral Mueller matrices of the sample. The wavelength and polarization state selections are automated, as is the Mueller matrix measurement. The imaging polarimeter permits the system to perform fast, stable measurements. Calibration allows us to reduce the error associated with the illumination and imaging optics in the microscope system. Our system indicates a clear difference between the average Mueller matrix measurements of healthy and cancerous human colon tissue, which agrees well with previously reported results.

Mesenchymal stem cells derived from adult adipose tissue possess the ability to differentiate into adipocytes, osteocytes, and chondrocytes which in turn can be developed into adipose tissues, cartilages, and bones. This regenerative characteristics has fueled the need to define improved stem-cell analysis protocol for enabling investigation of the differentiation process efficiently, economically, and non-invasively by start-of-the art imaging modalities. Here, we have demonstrated hyperspectral microscopy-based label-free imaging approach to study ASCs at a single-cell level. ASCs has been stimulated to become osteocytes using the growth media containing β –glycerophosphate, L-ascorbic acid 2-phosphate sesquimagnesium salt hydrate, and dexamethasone. Further, ASCs were stimulated to form adipocytes using the growth media containing biotin, pantothenate, bovine insulin, IBMX, penicillin, rosiglitazone, and dexamethasone.

In the present study, dark-field based hyperspectral Imaging (HSI) technique has been utilized to image single as well as multiple osteoblasts and adipocytes in salt media grown on the glass substrate. The spectral response of the cells at each pixel of the images were recorded in the visible-NIR range (400-900 nm). Response is stored in the three dimensional data-cube formed with two spatial dimensions and one spectral dimension. No special tagging or staining of the ASCs and derived osteoblasts, adipocytes has been done, as more likely required in traditional microscopy techniques. Incident light is diffracted at multiple angles and hence scattering response received after transmission is different even within the single cell due to sub-cellular heterogeneities present in the control and differentiating ASCs.

Based on dark-field images of control and differentiated sample, we found significant structural and spectral distinctiveness at day 14 onwards for differentiated osteoblasts and at day 6 onwards for adipocytes. Fourier filtering of images provides good visual inspection of structural modifications. Spectral data from the cellular surface and intracellular markers, and secreted molecules is stored to build the spectral libraries. Matrix-assisted laser deposition/ionization (MALDI) spectrometry technique is performed on control and differentiated cells to obtain insight of sub-cellular single molecules, mineral deposits, fats, proteins, and other biological mono-constituents. In the hyperspectral images, the entire spectrum is stored within each pixel as a vector where the number of spectral bands (wavelength range) equals vector dimension and the corresponding intensity signifies the component of the individual vector. Spectral signatures from the identified lipids are then matched to the in vitro stem-cells via spectral angle mapping (SAM) algorithms. By computing angle between two pixels, remarkable spectral similarity and dissimilarity are identified between control and differentiated stem cells. Pseudo-colored differentiating maps are produced by calibrating ‘match’ threshold. Secondary validation to the HSI is provided by evaluating optical images with template-match and edge-detection algorithms as well as second-harmonic generation microscopy to investigate osteoblasts.

Establishing this label-free protocol with minimum specimen preparation enables promising outcomes to overcome phototoxicity effect of traditional microscopy such as fluorescence/staining bleaching errors. The study would lead to high-throughput identification of patient specific derived cells for clinical use preventing mass rejection, and advance our understanding of the behavior of stem cellular clusters undergoing adipogenic and osteogenic differentiation.

Fluorescence lifetime imaging microscopy (FLIM) of metabolic coenzymes NAD(P)H and FAD monitors cancer treatment response and heterogeneity in in vitro and in vivo models. FLIM offers the potential to inform cancer patient treatment in a clinical setting, but requires further preclinical validation as an accurate predictor of patient response. Tissue banks offer readily available tumor samples with follow-up patient data that could be used for FLIM validation, but the effects of fixation and processing on autofluorescence have not been well characterized. This study aims to determine if trends seen with treatment in live tumor samples are conserved in formalin-fixed, paraffin-embedded tumors. Previously, fluorescence lifetime images of FaDu xenografts were acquired in vivo in mice before fixation and embedding. Here, corresponding lifetime images of the fixed FaDu tumors were acquired and compared with the in vivo data. The results demonstrate that while NAD(P)H lifetime values are generally conserved between in vivo and fixed tumors, FAD lifetime values are not. Additionally, the cancer response trends seen in vivo are positively correlated for most FLIM components in fixed tumors, but they vary in magnitude. Further investigation is required to determine the cause of the discrepancies.

Collagen is a main constituent of the extracellular matrix, and its content, organization, and crosslinking status affect significantly tissue mechanical properties. Cells continuously sense their local mechanical environment via interactions that impact significantly cell fate and the development and progression of numerous diseases. Non-linear imaging, especially through the combined use of second harmonic generation (SHG) and two-photon excited fluorescence (TPEF), has the potential to provide detailed information regarding collagen organization, content, and crosslinking with sub-micron level resolution in a non-invasive and label-free manner. However, the crosslink fluorescence properties that are correlated with tissue stiffness aren’t well understood. Thus, we characterized the TPEF emission over 680 to 920 nm excitation and 400 to 780 nm emission of three types of collagen gels: a) plain collagen gels, b) collagen gels with a 250 mM ribose solution for five days prior to gelation, and c) following gelation. Uniaxial tensile testing of gels created using the same protocols was performed by a custom-built testing system. We identified two components, which peak at 440 nm and 510 nm, that were significantly correlated with the exponential stiffening detected in collagen gels treated with ribose following gelation. Thus, TPEF information acquired at these emission ranges can be used in future studies and in combination with SHG measurements to develop detailed mathematical models that aim to predict tissue micromechanical properties. The latter will be useful in helping us to understand mechanosensitive cell behaviors that ultimately dictate the progress of diseases such as fibrosis and cancer.

Fluorescence lifetime images of the endogenous nicotinamide adenine dinucleotide (NADH), which is well known as autofluorescence chromophores, were obtained for rat normal fibroblast cells (WFB) and H-ras oncogene-transfected WFB cancer cells (W31) and for human normal lung fibroblast (MRC-5) and human lung large carcinoma (HCI-H661). In both cases, the average lifetime of the NADH autofluorescence was shorter in cancer cells than in normal cells, indicating that the difference in metabolism between healthy and cancer cells might alter the level of coenzymes such as NADH. It is also shown that application of nanosecond pulsed electric field (nsPEF) induces apoptosis in these cell, resulting in the morphological changes and lengthening the autofluorescence lifetime of NADH. Furthermore, we found that nsPEF more efficiently affected cancer cells than normal cells in cell viability, suggesting the possibility of drug-free cancer therapy by nsPEF.

The viscoelastic material properties of biological systems are increasingly recognized as important parts of signaling cascades involved in developmental and pathological processes. They are furthermore assumed to play a crucial role in surviving extreme environmental conditions for certain organisms, such as yeast cells. Confocal Brillouin microscopy gives access to the viscoelastic material properties of single cells and tissues in a contact- and label-free manner and with a high spatial resolution. In combination with quantitative phase imaging, it is then possible to determine the longitudinal modulus and the viscosity of the sample. In this study, we probed living zebrafish larvae in all anatomical planes, at different time points during development and after spinal cord injury. We could show, that confocal Brillouin microscopy detects the viscoelasticity of different anatomical structures without affecting the animal’s development. We furthermore observed a transiently decreasing Brillouin shift after spinal cord injury and a difference in Brillouin shift between in vivo and ex vivo measurements of the same sample region. Using quantitative phase imaging we additionally show, that the Brillouin shift of the probed tissues is mainly governed by their longitudinal modulus and viscosity. In conclusion, this work constitutes the methodical basis to identify key determinants of viscoelastic tissue properties during biologically important processes in vivo.

Colorectal cancer (CRC) ranks fourth in terms of newly diagnosed cases in the United States (135,430 in 2017); patients with locally advanced disease (Stage II and III) receive 5-fluorouracil (5-FU) and external beam radiotherapy-based neoadjuvant therapy (NAT) prior to surgical resection. However, there are no clinically accepted methods to assess in vivo therapeutic response to NAT.

Optical methods based on diffuse reflectance spectroscopy (DRS) have shown significant promise in predicting response to NAT in breast cancer, but the anatomy of the distal colon requires the use of endoscopically-deployable methods. We have developed a small-diameter (0.78 mm) multimodal optical imaging and diffuse reflectance spectroscopy (DRS) probe which can be deployed via the biopsy port of a commercial veterinary colonoscope (Karl Storz COLOView) to be used in a chemically-induced (azoxymethane (AOM)) orthotopic model.

Total diffuse reflectance measured by the probe was correlated with the reduced scattering (μ’_s(λ)) and absorption coefficients (μ_a(λ)) for λ = 450 – 800nm via a look-up table (LUT). Liquid phantoms were used to create the LUT and validate the measured μ’s and μa values. The LUT has a maximum total reflectance of 0.14 and ranges for μ_a and μ’_s are 0-10 cm^-1and 3-18 cm^-1, respectively. Error for μ’_s and μ_a has been 10.7±8.8% and 7.9±5.3%, respectively. For the imaging component, circular active area diameter is 325 μm and center-to-center fiber spacing of 3.3 μm.

Building on previous work this DRS approach enables quantification of total hemoglobin (Hb) content, oxygen saturation (SaO₂), estimates mean vessel diameter and scattering component, and allows for co-registered highresolution image data of superficial mucosa in vivo of tumor perfusion and microstructure, which can translate to the clinic to help physicians determine the response of tumors to therapy.

Optical microscopy and spectroscopy are widely used in multiple research areas relating to biology. Label-free spectroscopy and imaging are valuable tools that permit interrogation of biological samples without the need for exogenous labels, allowing for investigation of unperturbed biological systems. We demonstrate a coherent Raman technique called Doppler Raman (DR) spectroscopy which combines impulsive excitation with a novel frequency shift detection scheme for rapid, high sensitivity detection of low to medium frequency vibrational modes from 10-1800cm-1. Briefly, the DR spectroscope is a pump-probe system where the pump beam generates a time-varying index of refraction proportional to the Raman response of the sample. The time-delayed probe beam undergoes a frequency shift in the sample due to the time-varying index of refraction that is resolved using a novel high-sensitivity detection scheme. Other coherent Raman techniques such as Stimulated Raman Scattering (SRS) and Coherent Anti-Stokes Raman Spectroscopy (CARS) have been used to provide sensitive, label-free contrast for an array of biological targets, but their ability to detect low frequency vibrational modes is limited. Biologically significant targets like cytochrome c (740-760cm-1), DNA (782, 788, 1095cm-1), hydroxyapatite, and numerous pharmaceutical drugs exhibit rich Raman spectra across a range of low frequency modes below the well-known “fingerprint region”. Additionally, many proteins like hemoglobin, insulin, and bovine serum albumin have breathing modes below 50cm-1. Sensitive detection of low-frequency Raman vibrational modes unlocks a suite of potential biological and chemical dynamics like protein conformational changes and protein super complex formation.

Dementia is one of the main death leading causes worldwide and Alzheimer’s disease (AD) is its most common form. In postmortem examinations of AD brain tissue, extracellular deposits of proteins are observed, known as amyloid-beta (Aß) plaques. Aß plaques are characterized by their occurrence of beta-sheets and are, beside tau tangles, biological hallmarks in the postmortem diagnosis of AD. Little research on the detectability of Aß deposits in brain tissue using Raman spectroscopy has been published. Here, we examined formalin fixed, paraffin embedded tissue slices of AD and healthy control cases. The slices have been spectrally raster imaged with a step size smaller the size of a plaque using a commercial Raman spectroscope with a NIR laser source to obtain a hyperspectral map of the size of 0.5mm2. Specific band intensities including, among others, protein and lipid components were analyzed and afterward compared to the healthy control cases to study spectral differences. Further, Aß deposit locations could be precisely matched to the obtained spectral data by staining the same Raman imaged tissue slice with Thioflavin afterward. In addition, plaques can be co-localized by using histochemical stained adjacent tissue slices. In conclusion, we present new insights on spectral changes in the Raman fingerprint region of 950 to 1800cm-1 when analyzing the molecular composition of AD brain tissue.

Ovarian cancer is the fifth most deadly cancer among women in North America. Because this type of cancer is often diagnosed late, cytoreductive surgery is often the first therapeutic step. Currently, visual inspection of the surgical cavity is the only technique used to detect residual tumors. Therefore, there is a need for the development of new imaging techniques that can detect cancer tissue with high specificity and sensitivity during cytoreductive procedures. To address this unmet clinical need, we developed an intraoperative wide-field Raman spectroscopy (RS) imaging system to be used alongside tissue classification models trained to recognize cancer tissue using artificial intelligence techniques. The system can sequentially acquire up to 5 Raman bands in imaging mode over a macroscopic tissue area of more than 1-centimeter diameter. Preliminary analyses are presented demonstrating the ability of the system to recover the main Raman tissue bands in synthetic and biologic tissue. Two types of tissues in a biological sample can also be differentiated by the system. Moreover, cancer detection models are produced using a single-point RS probe based on ex vivo human measurements collected from 20 ovarian cancer patients. Using supervised machine learning techniques, it is demonstrated the model can detect tissue containing epithelial cancer cells with an accuracy higher than 90%. Based on this dataset, multivariate statistical analyzes were performed demonstrating the 5 features contributing the most to the classification. These studies pave the way to the development of a new generation wide-field Raman spectroscopy techniques for macroscopic tissue characterization during surgery.

Endometriosis is a condition in which the endometrium grows outside of the uterus. One of its severe effects is that it may cause infertility. Investigating only the symptoms of the disease is insufficient for the diagnosis. The correct and the exact diagnosis can be done by laparoscopy. During the disease the amount of molecules such as lipids, proteins and some antigens (i.e. CA125) in the blood serum changes. Raman spectroscopy is a vibrational spectroscopy technique which can analyze specific spectral patterns, substances and molecular changes with high specificity. In this study, Raman spectroscopy has been used as a tool for detecting the changes in blood serum contents for diagnosis of endometriosis. In our study, the Raman spectra of blood sera of 49 patients with 45 healthy women have been compared. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) methods are used in the analysis. The results indicated that 1003 cm-1 Raman band (phenylalanine) decreased for the women with endometriosis. In addition, a decrease in 1448 cm-1 band shows that the total amount of proteins decreased. The changes in the bands at 1155 and 1521 cm-1 imply that the amount of carotenoids diminished during the disease which is an expected effect of endometriosis. According to the analysis results, the sensitivity and the specificity of the Raman spectroscopy analysis were found to be 98.8% and 91.0% respectively for the detection of endometriosis. As a complementary method to laparoscopy, Raman spectroscopy can be used for diagnosis of endometriosis.

Optical spectroscopy offers unique opportunities for a label-free investigation of tissues at the molecular level to identify the variety of diseases. To transfer spectroscopic analysis from the scientific laboratories to clinical environment, fiber optic probes are required as optical bridges between the equipment and tissue. We developed single and combined fiber optic probes for the following set of spectroscopy methods: Mid IR-absorption, Raman scattering, Diffuse NIR-reflection, and auto-fluorescence. We benchmarked these methods and selected the optimal one (or their combination), that differentiate between healthy and malignant tissue, based on optical spectra. We tested cancer-normal tissue pairs of human body such as colon, kidney, brain as well as cartilages with and without injuries. Equines cartilage samples with and without osteoarthritis were tested as well. Obtained spectral data were evaluated by multivariate discrimination analysis to enable clear separation of malignant and normal tissues. Data fusion was revealed a synergic effect resulted in increasing of sensitivity, specificity and accuracy (up to 98% for kidney cancer).

Iron is an essential element required for human life. Iron is highly regulated in the body, as iron deficiency leads to many adverse health effects, such as anemias. Ferric iron (Fe3+) bound to serum transferrin (Tf) is internalized into cells via the transferrin receptor (TfR). Since the exact mechanisms of iron release in cells are not well known, a technique that allows detection of Tf bound iron inside intact human cells has been developed. Methods to determine when and where Tf releases iron inside a cell are required to better understand disease progression, including cancer. We have previously shown that Raman micro-spectroscopy is able to detect and quantify the Tf-bound iron in epithelial cells. In this work, we applied hyperspectral Raman imaging to visualize the spatial distribution of Tf-bound iron in human breast cancer T47D cells internalized with iron-loaded Tf, oxalate-Tf, a chemical Tf mutant unable to release iron, and iron-depleted Tf. We have also shown that Raman imaging can quantify the amount of iron under different times of Tf internalization (Tf uptake time), prior to fixation. Raman micro-spectroscopy provides a unique way to determine the amount of iron under different Tf internalization times by employing the Raman metric, which was used to quantify iron content in oxa- , apo-, and holo-Tf samples. Importantly, Raman microspectroscopy can be used to follow iron release from Tf in breast cancer cells. Determining the kinetics and location of iron release in cancer cells is key to further our understanding of iron metabolism during cancer progression.

Regenerative medicine encompasses the rebuilding or repairing of organs. We are developing bioengineered organoids that will differentiate when implanted in vivo to partially restore organ function. These complex organoids, derived from embryonic salivary gland cells, include both primary mesenchyme and epithelial progenitor cells. Noninvasive quality monitoring of tissue-engineered constructs is required before implantation of bioengineered constructs in vivo. Raman spectroscopy offers fast, simple, and, most importantly, non-invasive quantitative cell and tissue analysis that does not require elaborate sample preparation. We demonstrate the application of Raman micro-spectroscopy technique to in vitro monitoring of cell types within 3D cell clusters, with the ultimate goal of applying this technology in situ to monitor adult cell-derived organoids that are implanted in vivo. We have collected Raman spectra of epithelial and mesenchymal progenitor cells in vitro, and have shown that we are able to identify different Raman signatures corresponding to each cell type. In particular, we have observed Raman spectral differences which correspond to the C-C and C-N stretch in proteins, as well as in the Amide I and III envelopes. The embryonic mesenchyme cells are similar to mesenchymal stem cells, MSCs, which can differentiate into bone, cartilage, and other cell types. In addition to salivary gland tissue engineering applications, mesenchymal cells offer a great potential in repairing bone, cartilage, and damaged heart cells, and to treat inflammation and immune system diseases. In future studies, our Raman spectroscopy methods can be broadly applied to monitoring of organoids for application in many diseases.

We demonstrate a multi-color background-free coherent anti-Stokes Raman scattering (CARS) imaging system, using a low-cost, all-fiber, energetic, multi-wavelength time-lens source. The time-lens source generates picosecond pulse trains at three different wavelengths. The first is 1064.3 nm, the second is tunable between 1052 nm and 1055 nm, and the third is tunable between 1040 nm and 1050 nm. When the time-lens source is synchronized with a mode-locked Ti:Sa laser, two of the three wavelengths are used to detect different Raman frequencies for two-color on-resonance imaging, whereas the third wavelength is used to obtain the off-resonance image for nonresonant background subtraction. Mixed poly(methyl methacrylate) (PMMA) and polystyrene (PS) beads are used to demonstrate two-color background-free CARS imaging. Synchronization of the multi-wavelength time-lens source with a microscope enables pixel-to-pixel wavelength-switching. Simultaneous two-color CARS imaging of CH₂ and CH₃ stretching vibration modes with realtime background subtraction is demonstrated in ex vivo mouse tissue.

There is an increasing but unmet need for accurate, label-free and automated bio-aerosol sensing. To address this need, we developed a high-throughput, cost-effective and portable bio-aerosol sensor based on computational microscopy and deep-learning. Our device is composed of an impactor and a lens-less digital holographic on-chip microscope. It screens air at 13 liters per minute, and captures bio-aerosols on the impactor substrate. An image sensor then records the in-line holograms of these captured bio-aerosols in real time. Using these recorded in-line holograms, the captured bio-aerosols are analyzed within a minute, facilitated by two deep convolutional neural networks (CNNs): the first CNN simultaneously performs auto-focusing and phase-recovery to reconstruct both the amplitude and phase images of each bio-aerosol with sub-micron resolution; and the second CNN performs automatic classification of the reconstructed bio-aerosols into pre-trained classes and counting their densities in air. As a proof-of-concept, we demonstrated reconstruction and label-free sensing of five different types of bio-aerosols: Bermuda grass pollen, oak tree pollen, ragweed pollen, Aspergillus spore, and Alternaria spore. These bio-aerosols form some of the most common allergens in air. Using our mobile bio-aerosol sensor, we achieved ~94% precision and recall in differentiating these bio-aerosols without the use of any labeling. We also demonstrated successful sensing of oak tree pollens in the field using our mobile device. To the best of our knowledge, this is the first demonstration of automated label-free sensing of bio-aerosols using a portable device, which is enabled by computational microscopy and deep-learning. It has broad applications in label-free bio-aerosol sensing and air-quality monitoring.

Tomography is one of the most powerful imaging tools for analyzing biological samples, able to furnish complete mapping of the object in 3D. In particular, tomographic phase microscopy (TPM) exploits quantitative phase imaging (QPI) techniques to map the 3D refractive index (RI) of cells, by adopting laser beam deflection, direct mechanical rotation or holographic optical tweezers (HOTs) to probe the sample along a number of controlled directions. To date, all tomographic methods require a high-precision, opto-mechanical and/or opto-electronic device to acquire a set of many images by probing the sample along a large number of controlled directions. Here we report on a smart solution to obtain TPM of samples at lab-on-chip scale, by exploiting their tumbling inside microfluidic channels. This method, recently developed, presents the following advantages: (i) Permits to observe full 360° of rotating cells, this avoiding the limitation in the accuracy of tomograms; (ii) no mechanical contact neither holographic optical tweezers are needed to rotate the sample; (iii) it is suitable for application in flowing conditions with high-throughput performances. This would allow real microfluidic biomedical applications on a large scale. The results shown in a previous work for RBCs and diatoms are here extended to quasi-spherical cells, by exploiting a new algorithm for rolling angle recovery in TPM. In particular, we performed the 3D imaging of human breast adenocarcinoma MCF-7 cells, opening the way for the full characterization of circulating tumor cells (CTCs) in the new paradigm of liquid biopsy.

We present our latest advances in multi-wavelength label-free imaging using quantitative phase imaging. Basic QPI designs usually incorporate a single, coherent or partially coherent light source. The introduction of off-axis geometry allows for calculation of phase information from a single-exposure interferogram. For multi-wavelength imaging, it has been previously required to take sequential interferograms with different wavelengths, effectively reducing imaging throughput, or to use color cameras that have cross-talk between the color channels and artefacts caused by Bayer pattern. Alternatively, it was suggested to use spatial interferometric multiplexing in order to provide multi-wavelength data, where multiple interferograms of different off-axis angles are acquired in a single exposure. We introduce an external module for label-free QPI capable of multi-color imaging using spatial interferometric multiplexing of multiple color channels, with self-referencing that significantly reduces mechanical vibration noise. This technique has various applications such as interferometric spectroscopy of live cells and multiple wavelength phase unwrapping.

The differences in the swimming behavior of sperm cell populations that carry the opposite sex chromosome have been an important topic of research, aiming to shed more light on the seemingly random process of gender determination at conception. Earlier studies on human sperm cells resulted in a misconception that, Y-chromosome bearing sperm cells swim faster than X-chromosome bearing sperm cells as they carry a lighter payload. This has been clarified with more recent studies using modern computer-aided semen analysis (CASA) systems and improved sex-sorting techniques, showing that the velocity parameters of the two sperm populations exhibit similar values. CASA systems typically rely on conventional optical microscopes however, where the trade-off between spatial resolution and field-of-view and poor depth resolution necessitate confining the sperm cells into shallow chambers which limit their 3D motion. Alternatively, dual-view on-chip holographic imaging offers a unique capability to image free-swimming sperm cells across a large volume (~1.8 μl) and depth (~0.6 mm) in 3D. Operating our platform at 300 fps, we have comparatively analyzed the complete 3D motion characteristics of 235 X-sorted and 289 Y-sorted free-swimming bovine sperm cells, which include the head translation and spin as well as the 3D flagellar beating. While there was no significant difference in the velocity parameters, it was observed that the Y-sorted sperm had a stronger preference for helical trajectories compared to X-sorted sperm with a higher linearity. Comparatively studying the kinematic responses to the surrounding chemicals and ions could help better understand the reasons behind these observed differences.

Wide-field interferometric imaging systems can detect mechanical deformations of a cell during an action potential (AP), such as in quantitative phase microscopy, which is highly sensitive to the changing optical path length. This enables non-invasive optophysiology of spiking cells without exogeneous markers, but high-fidelity imaging of such deformations requires averaging of a large number of spikes synchronized by electrical recordings. We have developed new iterative methods for detecting single APs from quantitative phase microscopy of spiking cells, enabling an all-optical detection system with high accuracy and good temporal resolution. We demonstrate performance of the method across multiple preparations of spiking HEK-293 cells and compare the outcomes of the all-optical measurements with the ground truth detected on a multi-electrode array. We initially use a spike-triggered average, synchronized to an electrical recording, to measure deformations during the AP in spiking cells, which reach up to 3 nm (0.9 mrad) with a rise time of 4 ms and fall time of about 120 ms. Based on this knowledge of the AP dynamics, optical data analysis can provide reliable spike detection, within a standard deviation of 11.6 ms (9.7% of the length of the action potential) with an 8.5% false negative detection rate. The method is robust to natural variations between cells and can be modified to function without any prior knowledge of the AP dynamics. Such a system could achieve high-throughput measurements of network activity in culture and help identify the mechanisms linking cell deformations to the changes of transmembrane potential.

Even though the speed of sound (SoS) is non-homogeneous in biological tissue, most reconstruction algorithms for optoacoustic imaging neglect its variation. In addition, when heavy water is used as coupling medium to enable imaging of certain biological chromophores such as lipids and proteins, the SoS also differs significantly between couplant and tissue. While the assumption of uniform SoS is known to introduce visible deformations of features in single-wavelength optoacoustic images, the spectral error introduced by the assumption of uniform SoS is not fully understood. In this work, we provide an in-depth spectral analysis of multi-spectral optoacoustic imaging artifacts that result from the assumption of uniform SoS in situations where SoS changes substantially. We propose a dual-SoS model to incorporate the SoS variation between the couplant and the sample. Tissue-mimicking phantom experiments and in vivo measurements show that uniform SoS reconstruction causes spectral smearing, which dual-SoS modeling can largely eliminate. Due to this increased spectral accuracy, the method has the potential to improve clinical studies that rely on quantitative optoacoustic imaging of biomolecules like hemoglobin or lipids.

Imaging of biological specimen is one of the most important tools to investigate structures and functionalities in organic components. Improving the resolution of images into the nanometer range call for short wavelengths light sources and large aperture optics. Subsequently, the use of extreme ultraviolet light in the range of 2 nm to 5 nm provides high contrast and high resolution imaging, if it is combined with lensless imaging techniques. We describe important parameters for high resolution lensless imaging of biological samples and specify the required light source properties. To overcome radiation based damage of biological specimen, we discuss the concept of ghost imaging and describe a possible setup towards biological imaging in the extreme ultraviolet range.

A swept-source optical coherence tomography (OCT) system using mid-infrared light is demonstrated using commercially available light source and photodetectors. The light source is a tunable external cavity quantum cascade laser (QCL) with four laser modules and a wavelength range of 5.4μm – 12.8μm. The laser output is pulsed and has a maximum allowable duty cycle of 8%. A thermoelectrically cooled photovoltaic detector with a peak sensitivity at 8μm is configured in a Michelson interferometer setup. Noise from three sources are characterized: the analogue-to-digital converter (ADC), the detectors and the relative intensity noise (RIN). The ADC and the detector have a noise spectral density of 2.8 pW/√Hz and 21.4 pW/√Hz over a bandwidth of 100 MHz respectively; the value of RIN is -102 dB/Hz. The OCT system is used to collect the reflectivity profile of a mirror, which is measured with a signal to noise ratio (SNR) of 17.5 dB. Decreasing the time of sweep from 1s to 200ms has the effect of reducing the SNR by 4 dB.

An imaging device based on line-field confocal optical coherence tomography (LC-OCT) operating in two distinct spectral bands centered at 770 nm and 1250 nm is presented. A single supercontinuum light source and two different line-scan cameras are used. B-scans are acquired simultaneously in the two bands at 4 frames per second. In the 770-nm band, high resolution (1.3 μm x 1.2 μm, lateral x axial) imaging is achieved, while extended penetration (~ 700 μm) is obtained in the 1250-nm band. Greyscale fusion of the two images is performed to produce a single image with both high resolution in the superficial part of the image and deep penetration. A color representation is also used to highlight spectroscopic properties of the sample and to enhance contrast.

Early detection of cancer through medical imaging has a critical impact on patient survival rates. There are many efforts for detecting early cancer in situ using modalities other than traditional medical optical imaging, which contain additional information over conventional micrographs of surface morphology acquired without staining. We analyzed the Mueller matrix components of human colon tissue obtained with an imaging polarimeter microscope at an illumination wavelength of 442 nm by principal components analysis in order to separate the traditional non-polarized gray image and to investigate the structure of the parameter space of polarization transformation by tissue. We also analyzed Mueller matrix by mapping it to a coherent matrix and performed eigenvalue analysis. The 1st to 4th principal components contain 99% of the information present in the images; polarization information contributes less than 10% of the information in the Mueller matrix. In one individual, 80% of the cancer was detected, without the first components which contains traditional non-polarized gray image for traditional diagnosis. Microscopic fine structures were observed, particularly in the 3rd and 4th principal components’ score images. The entropy image of corrugated cancer tissue was smoother than that of the traditional gray image. There were several abnormal regions identified in adjacent regions of cancer, whose residues exceeded the noise level of the instrument used.

Febrile neutropenia (FN) is a common cause of hospitalization for cancer patients undergoing chemotherapy treatment. To screen for FN, patients require invasive blood draws and complete blood cell counts, which increases risk of nosocomial infection while in an immunocompromised state. There is a pressing clinical need for non-invasive, point-of-care technology to frequently screen for FN, which, if detected early, can be prophylactically managed. A promising approach to address this need is capillaroscopy, through which blood cells are imaged in capillaries non-invasively. Visualization of shadows caused by absorption of individual red blood cells is currently achievable, and correlation between the absence of optical absorption gaps and severe neutropenia has been observed. However, a completely accurate identification of the physical origin of these optical absorption gaps for conclusive neutropenia diagnosis remains an elusive task. Here we present scattering oblique plane microscopy as a means of imaging moving scattering particles within a turbid medium with the goal of eventually imaging and characterizing blood cells in vivo flowing in superficial capillaries. Our imaging system illuminates an oblique light sheet through a capillary bed and collects back-scatter using a single objective at frame rates of >200 Hz. To validate this system, we develop phantoms mimicking capillaries with 200 μm diameter lumens embedded deep in silicone doped with TiO₂ and India ink. Single 3 μm diameter polystyrene beads flowing through the capillaries are resolved with a signal to noise ratio of approximately 5:1 at a depth of 1 mean free path.

Skin lesions are the most common human cancer diseases, usually, is it diagnosed by clinical visual inspections followed by biopsy. Early detection of these diseases is critical, depending on an accurate and trained dermatologist and can increase the survival rate. Aiming for screening and early diagnose skin lesions many techniques are presented, however, optical techniques are highlighted since they are fast and noninvasive. In this context, fluorescence steady-state and lifetime imaging show potential by being able to image metabolic changes using endogenous contrast. Here it is demonstrated an in vivo label-free multispectral fluorescence lifetime imaging system to distinguish between two types of clinically similar lesions. A pulsed Nd:YAG laser emitting at 355 nm is used to excite the endogenous fluorophores and three channels of acquisition bands are used to imaging the skin. Preliminary results showed differences in the fluorescence lifetime between Bowen and Actinic Keratosis as well as the lesion and the skin around, demonstrating a potential tool to identify the lesion and its edges.

Coronary artery disease (CAD), or atherosclerosis, is responsible for nearly a third of all American deaths annually. Detection of plaques and differentiation of plaque stage remains a complicating factor for treatment. Classification of plaque before significant blockage or rupture could inform clinical decisions and prevent mortality. Current detection methods are either nonspecific, slow, or require the use of potentially harmful contrast agents. Recent advances in hyperspectral imaging could be used to detect changes in the autofluorescence of arteries associated with vessel remodeling and subsequent plaque formation and could detect and classify existing lesions. Here, we present data comparing spectral image characteristics of a mouse model designed to undergo vessel remodeling.

C57Bl/6 mice underwent ligation of three of four caudal branches of the left common carotid artery (left external carotid, internal carotid, and occipital artery) with the superior thyroid artery left intact under IACUC approved protocol. Vessels were harvested at a variety of timepoints to compare degrees of remodeling, including 4 weeks and 5 months post-surgery. Immediately following harvest, vessels were prepared by longitudinal opening to expose the luminal surface to a 20X objective. A custom inverted microscope (TE-2000, Nikon Instruments) with a Xe arc lamp and thin film tunable filter arrary (Versachrome, Semrock, Inc.) were used to achieve spectral imaging. Excitation scans utilized wavelengths between 340 nm and 550 nm in 5 nm increments. Hyperspectral data were generated and analyzed with custom Matlab scripts and visualized in ENVI. Preliminary data suggest consistent spectral features associated with control and remodeled vessels.

Photoacoustic microscopy (PAM) can be an effective imaging modality to visualize retinal vein occlusion during laserinduced photocoagulation on major retinal veins. Developments in the vessels could be observed before and after laser irradiation due to the change of the optical absorption spectrum of the target vessels. However, the suitable wavelength to achieve high contrast PAM images of occluded vessels is unclear. This study evaluates the effect of wavelength on PAM imaging to evaluate the photocoagulation lesions on the rabbit to optimize the wavelength for imaging. Retinal vein occlusion (RVO) was created using a 532 nm millisecond pulse duration green light with concurrent intravenous administration of Rose Bengal (5 mg/kg) in New Zealand rabbits. Imaging was acquired by the PAM system at various wavelengths ranging from 520 nm to 590 nm. In addition, the thermal lesion was also confirmed using optical coherence tomography (OCT). A group of 20 retinal veins was irradiated for 0.5 s at a laser fluence of 850 W/cm² (power = 150 mW, beam diameter = 75 μm). Twenty shots of the laser were applied to each major vein. PAM results showed that the thermal lesion was obviously visualized and exhibited lower contrast in comparison with untreated vessels posttreatment. Photoacoustic spectroscopy exhibited that the highest PA contrast of vessels treated with Rose Bengal laserinduced RVO occurred at a wavelength of 563, 570, and 578 nm, which was higher than the PA amplitude at lower and longer wavelengths respectively. The use of multi-wavelength PAM can provide a better method for visualization and evaluation of retinal vein occlusions.

Due to the chromatic dispersion properties inherent in all optical materials, even the best designed multi-spectral objective will exhibit residual chromatic aberration effect. Here we show that the aberration correction ability of Fourier Ptychographic Microscopy (FPM) is well matched and well suited for post-image acquisition correction of these effects to render in-focus images. We show that an objective with significant spectral focal shift (up to 0.02 μm/nm) and spectral field curvature (up to 0.05 μm/nm drift at off-axis position of 800μm) can be computationally corrected to render images with effectively null spectral defocus and field curvature. This approach of combining optical objective design and computational microscopy provides a good strategy for high quality multi-spectral imaging over a broad spectral range, and eliminating the need for mechanical actuation solutions.

We propose plasmonic nanogratings such that they have tunable plasmon resonances in far- and deep-UV wavelength ranges. These plasmonic nanostructures and nanogratings of different plasmonic materials have been simulated using Finite Difference Time Domain modeling and the Rigorous Coupled Wave Analysis. Application of these plasmonic nanostructures for surface plasmon resonance and localized surface plasmon resonance based sensing in the UV wavelength regime has been demonstrated by calculating the shifts in the plasmon resonance wavelength for changes in either the bulk refractive index or the localized refractive index of the medium next to the plasmonic structures.