Unsupervised hyperspectral stimulated Raman microscopy image enhancement: denoising and segmentation via one-shot deep learning.

Hyperspectral stimulated Raman scattering (SRS) microscopy is a label-free technique for biomedical and mineralogical imaging which can suffer from low signal-to-noise ratios. Here we demonstrate the use of an unsupervised deep learning neural network for rapid and automatic denoising of SRS images: UHRED (Unsupervised Hyperspectral Resolution Enhancement and Denoising). UHRED is capable of "one-shot" learning; only one hyperspectral image is needed, with no requirements for training on previously labelled datasets or images. Furthermore, by applying a k-means clustering algorithm to the processed data, we demonstrate automatic, unsupervised image segmentation, yielding, without prior knowledge of the sample, intuitive chemical species maps, as shown here for a lithium ore sample.

All normal dispersion nonlinear fibre supercontinuum source characterization and application in hyperspectral stimulated Raman scattering microscopy.

Hyperspectral stimulated Raman scattering (SRS) microscopy is a powerful label-free, chemical-specific technique for biomedical and mineralogical imaging. Usually, broad and rapid spectral scanning across Raman bands is required for species identification. In many implementations, however, the Raman spectral scan speed is limited by the need to tune source laser wavelengths. Alternatively, a broadband supercontinuum source can be considered. In SRS microscopy, however, source noise is critically important, precluding many spectral broadening schemes. Here we show that a supercontinuum light source based on all normal dispersion (ANDi) fibres provides a stable broadband output with very low incremental source noise. We characterized the noise power spectral density of the ANDi fibre output and demonstrated its use in hyperspectral SRS microscopy applications. This confirms the viability and ease of implementation of ANDi fibre sources for broadband SRS imaging.

Oblique angle transient-reflectivity laser-scanning microscopy for mineral imaging in natural ores.

The microscopic arrangement of different minerals in ores is of high interest for mine planning, mineral processing and extractive metallurgy. Many economically important, naturally occurring minerals are highly absorbing semiconductors. To characterize these materials, we have implemented pump-probe laser scanning microscopy (LSM) in a two-lens reflective configuration that offers efficient collection of signal light by using a combination of galvanometer and sample stage scanning. We show that the short-time (∼10 ps) pump-probe response of a material allows us to distinguish economically important sulfide minerals.

Pre-Clinical Translation of Second Harmonic Microscopy of Meniscal and Articular Cartilage Using a Prototype Nonlinear Microendoscope.

Previous studies using nonlinear microscopy have demonstrated that osteoarthritis (OA) is characterized by the gradual replacement of Type II collagen with Type I collagen. The objective of this study was to develop a prototype nonlinear laser scanning microendoscope capable of resolving the structural differences of collagen in various orthopaedically relevant cartilaginous surfaces. The current prototype developed a miniaturized femtosecond laser scanning instrument, mounted on an articulated positioning system, capable of both conventional arthroscopy and second-harmonic laser-scanning microscopy. Its optical system includes a multi-resolution optical system using a gradient index objective lens and a customized multi-purpose fiber optic sheath to maximize the collection of backscattered photons or provide joint capsule illumination. The stability and suitability of the prototype arthroscope to approach and image cartilage were evaluated through preliminary testing on fresh, minimally processed, and partially intact porcine knee joints. Image quality was sufficient to distinguish between hyaline cartilage and fibrocartilage through unique Type I and Type II collagen-specific characteristics. Imaging the meniscus revealed that the system was able to visualize differences in the collagen arrangement between the superficial and lamellar layers. Such detailed in vivo imaging of the cartilage surfaces could obviate the need to perform biopsies for ex vivo histological analysis in the future, and provide an alternative to conventional external imaging to characterize and diagnose progressive and degenerative cartilage diseases such as OA. Moreover, this system is readily customizable and may provide a suitable and modular platform for developing additional tools utilizing femtosecond lasers for tissue cutting within the familiar confines of two or three portal arthroscopy techniques.

Direct mineralogical imaging of economic ore and rock samples with multi-modal nonlinear optical microscopy.

Multi-modal nonlinear optical (NLO) microscopy, including stimulated Raman scattering (SRS) and second harmonic generation (SHG), was used to directly image mineralogical features of economic ore and rock samples. In SRS/SHG imaging, ore samples generally require minimal preparation and may be rapidly imaged, even in their wet state. 3D structural details, at submicron resolution, are revealed tens of microns deep within samples. Standard mineral imaging based on scanning electron microscopy (SEM), with elemental analysis via energy dispersive X-Ray spectroscopy, was used to independently validate the mineral composition of the samples. Spatially-resolved SRS from dominant Raman-resonant bands precisely maps the locations of specific minerals contained within the samples. SHG imaging reveals locally non-centrosymmetric structures, such as quartz grains. Competing absorption and nonlinear scattering processes, however, can reduce contrast in SRS imaging. Importantly, the correlation between standard electron microscopy and multi-modal NLO optical microscopy shows that the latter offers rapid image contrast based on the mineral content of the sample.

Amplitude and polarization modulated hyperspectral Stimulated Raman Scattering Microscopy.

We present a simple hyperspectral Stimulated Raman Scattering (SRS) microscopy method based on spectral focusing of chirped femtosecond pulses, combined with amplitude (AM) and polarization (PM) modulation. This approach permits the imaging of low concentration components with reduced background signals, combined with good hyperspectral resolution and rapid spectral scanning. We demonstrate, using PM-SRS in a Raman loss configuration, the spectrally resolved detection of deuterated dimethyl sulfoxide (DMSO-d6) at concentrations as low as 0.039 % (5.5 mM). In general, background signals due to cross-phase modulation (XPM), two-photon absorption (TPA) and thermal lensing (TL) can reduce the contrast in SRS microscopy. We show that the nonresonant background signal contributing to the SRS signal is, in our case, largely due to XPM. Polarization modulation of the Stokes beam eliminates the nonresonant XPM background, yielding high quality hyperspectral scans at low analyte concentration. The flexibility of our combined AM-PM approach, together with the use of variable modulation frequency and lock-in phase, should allow for optimization of SRS imaging in more complex samples.

Label-free hyperspectral nonlinear optical microscopy of the biofuel micro-algae Haematococcus Pluvialis.

We consider multi-modal four-wave mixing microscopies to be ideal tools for the in vivo study of carotenoid distributions within the important biofuel microalgae Haematococcus pluvialis. We show that hyperspectral coherent anti-Stokes Raman scattering (CARS) microscopy generates non-invasive, quantitative real-time concentrations maps of intracellular carotenoid distributions in live algae.

Stearoyl-CoA desaturase inhibition blocks formation of hepatitis C virus-induced specialized membranes.

Hepatitis C virus (HCV) replication is dependent on the formation of specialized membrane structures; however, the host factor requirements for the formation of these HCV complexes remain unclear. Herein, we demonstrate that inhibition of stearoyl-CoA desaturase 1 (SCD-1) halts the biosynthesis of unsaturated fatty acids, such as oleic acid, and negatively modulates HCV replication. Unsaturated fatty acids play key roles in membrane curvature and fluidity. Mechanistically, we demonstrate that SCD-1 inhibition disrupts the integrity of membranous HCV replication complexes and renders HCV RNA susceptible to nuclease-mediated degradation. Our work establishes a novel function for unsaturated fatty acids in HCV replication.

Hyperspectral multimodal CARS microscopy in the fingerprint region.

A simple scheme for multimodal coherent anti-Stokes Raman scattering (CARS) microscopy is based on the spectral focusing of ultrafast-oscillator-derived pump/probe light and synchronous photonic crystal fiber (PCF) fiber-generated broadband Stokes light. To date, such schemes allowed rapid hyperspectral imaging throughout the CH/OH high frequency region (2700-4000 cm(-1) ). Here we extend this approach to the middle (1640-3300 cm(-1) ) and fingerprint regions (850-1800 cm(-1) ) of the Raman spectrum. Our simple integrated approach to rapid hyperspectral CARS microscopy in the fingerprint region is demonstrated by applications to label-free multimodal imaging of cellulose and bulk bone, including use of the phosphate resonance at 960 cm(-1) .

Fluorescence lifetime imaging of alterations to cellular metabolism by domain 2 of the hepatitis C virus core protein.

Hepatitis C virus (HCV) co-opts hepatic lipid pathways to facilitate its pathogenesis. The virus alters cellular lipid biosynthesis and trafficking, and causes an accumulation of lipid droplets (LDs) that gives rise to hepatic steatosis. Little is known about how these changes are controlled at the molecular level, and how they are related to the underlying metabolic states of the infected cell. The HCV core protein has previously been shown to independently induce alterations in hepatic lipid homeostasis. Herein, we demonstrate, using coherent anti-Stokes Raman scattering (CARS) microscopy, that expression of domain 2 of the HCV core protein (D2) fused to GFP is sufficient to induce an accumulation of larger lipid droplets (LDs) in the perinuclear region. Additionally, we performed fluorescence lifetime imaging of endogenous reduced nicotinamide adenine dinucleotides [NAD(P)H], a key coenzyme in cellular metabolic processes, to monitor changes in the cofactor's abundance and conformational state in D2-GFP transfected cells. When expressed in Huh-7 human hepatoma cells, we observed that the D2-GFP induced accumulation of LDs correlated with an increase in total NAD(P)H fluorescence and an increase in the ratio of free to bound NAD(P)H. This is consistent with an approximate 10 fold increase in cellular NAD(P)H levels. Furthermore, the lifetimes of bound and free NAD(P)H were both significantly reduced--indicating viral protein-induced alterations in the cofactors' binding and microenvironment. Interestingly, the D2-expressing cells showed a more diffuse localization of NAD(P)H fluorescence signal, consistent with an accumulation of the co-factor outside the mitochondria. These observations suggest that HCV causes a shift of metabolic control away from the use of the coenzyme in mitochondrial electron transport and towards glycolysis, lipid biosynthesis, and building of new biomass. Overall, our findings demonstrate that HCV induced alterations in hepatic metabolism is tightly linked to alterations in NAD(P)H functional states.

Spatial-spectral coupling in coherent anti-Stokes Raman scattering microscopy.

Coherent anti-Stokes Raman scattering (CARS) microscopy is a third-order nonlinear optical technique which permits label-free, molecule-specific hyperspectral imaging. The interference between coherent resonant and non-resonant terms leads to well known distortions in the vibrational spectrum, requiring the use of retrieval algorithms. It also leads to spatial imaging distortions, largely due to the Gouy phase, when objects are smaller than the Rayleigh range. Here we consider that the focal position and spectral contributions to the nonlinear image formation are intrinsically coupled and cannot be corrected by conventional retrieval methods.

Nonlinear optical microscopy in decoding arterial diseases.

Pathological understanding of arterial diseases is mainly attributable to histological observations based on conventional tissue staining protocols. The emerging development of nonlinear optical microscopy (NLOM), particularly in second-harmonic generation, two-photon excited fluorescence and coherent Raman scattering, provides a new venue to visualize pathological changes in the extracellular matrix caused by atherosclerosis progression. These techniques in general require minimal tissue preparation and offer rapid three-dimensional imaging. The capability of label-free microscopic imaging enables disease impact to be studied directly on the bulk artery tissue, thus minimally perturbing the sample. In this review, we look at recent progress in applications related to arterial disease imaging using various forms of NLOM.

Forward-collected simultaneous fluorescence lifetime imaging and coherent anti-Stokes Raman scattering microscopy.

We demonstrate the simultaneous collection and separation of femtosecond-laser-based forward-collected coherent anti-Stokes Raman scattering (F-CARS) and two-photon-excitation-induced fluorescence lifetime images (FLIM) using time-correlated single photon counting (TCSPC). We achieve this in a nondescanned geometry using a single multimode fiber without significant loss of light, field of view, and most importantly, TCSPC timing fidelity. In addition to showing the ability to separate CARS images from FLIM images using time gating, we also demonstrate composite multimodal epicollected FLIM imaging with fiber-collected F-CARS imaging in live cells.

Single laser source for multimodal coherent anti-Stokes Raman scattering microscopy.

Short laser pulse technology has significantly contributed to biomedical research, especially via nonlinear optical microscopy. Coherent anti-Stokes Raman scattering (CARS) microscopy is a label-free, chemical-selective method that is growing in importance as improved methods and light sources develop. Here we discuss different approaches to laser source development for CARS microscopy and highlight the advantages of a multimodal CARS microscope, illustrated by selected applications in biomedical research.

Dynamics of lipid droplets induced by the hepatitis C virus core protein.

The hepatitis C virus (HCV) is a global health problem, with limited treatment options and no vaccine available. HCV uses components of the host cell to proliferate, including lipid droplets (LD) onto which HCV core proteins bind and facilitate viral particle assembly. We have measured the dynamics of HCV core protein-mediated changes in LDs and rates of LD movement on microtubules using a combination of coherent anti-Stokes Raman scattering (CARS), two-photon fluorescence (TPF), and differential interference contrast (DIC) microscopies. Results show that the HCV core protein induces rapid increases in LD size. Particle tracking experiments show that HCV core protein slowly affects LD localization by controlling the directionality of LD movement on microtubules. These dynamic processes ultimately aid HCV in propagating and the molecules and interactions involved represent novel targets for potential therapeutic intervention.

Multimodal nonlinear optical imaging of atherosclerotic plaque development in myocardial infarction-prone rabbits.

Label-free imaging of bulk arterial tissue is demonstrated using a multimodal nonlinear optical microscope based on a photonic crystal fiber and a single femtosecond oscillator operating at 800 nm. Colocalized imaging of extracellular elastin fibers, fibrillar collagen, and lipid-rich structures within aortic tissue obtained from atherosclerosis-prone myocardial infarction-prone Watanabe heritable hyperlipidemic (WHHLMI) rabbits is demonstrated through two-photon excited fluorescence, second harmonic generation, and coherent anti-Stokes Raman scattering, respectively. These images are shown to differentiate healthy arterial wall, early atherosclerotic lesions, and advanced plaques. Clear pathological changes are observed in the extracellular matrix of the arterial wall and correlated with progression of atherosclerotic disease as represented by the age of the WHHLMI rabbits.

Differentiating atherosclerotic plaque burden in arterial tissues using femtosecond CARS-based multimodal nonlinear optical imaging.

A femtosecond CARS-based nonlinear optical microscope was used to simultaneously image extracellular structural proteins and lipid-rich structures within intact aortic tissue obtained from myocardial infarction-prone Watanabe heritable hyperlipidemic rabbits (WHHLMI). Clear differences in the NLO microscopic images were observed between healthy arterial tissue and regions dominated by atherosclerotic lesions. In the current ex-vivo study, we present a single parameter based on intensity changes derived from multi-channel NLO image to classify plaque burden within the vessel. Using this parameter we were able to differentiate between healthy regions of the vessel and regions with plaque, as well as distinguish plaques relative to the age of the WHHLMI rabbit.

Multimodal CARS microscopy of structured carbohydrate biopolymers.

We demonstrate the utility of multimodal coherent anti-Stokes Raman scattering (CARS) microscopy for the study of structured condensed carbohydrate systems. Simultaneous second-harmonic generation (SHG) and spectrally-scanned CARS microscopy was used to elucidate structure, alignment, and density in cellulose cotton fibers and in starch grains undergoing rapid heat-moisture swelling. Our results suggest that CARS response of the O-H stretch region (3000 cm(-1)-3400 cm(-1)), together with the commonly-measured C-H stretch (2750 cm(-1)-2970 cm(-1)) and SHG provide potentially important structural information and contrast in these materials.

All-fiber CARS microscopy of live cells.

Using an all-fiber laser system consisting of a femtosecond Er/Yb fiber oscillator as the pump and an ultra-highly nonlinear fiber for Stokes generation, we demonstrate multimodal (TPF+SHG+CARS) non-linear optical microscopy of both tissue samples and live cells. Multimodal imaging was successfully performed with pixel dwell times as short as 4 micros at low laser powers (< 40 mW total).

Optimally chirped multimodal CARS microscopy based on a single Ti:sapphire oscillator.

We demonstrate high performance coherent anti-Stokes Raman scattering (CARS) microscopy of live cells and tissues with user-variable spectral resolution and broad Raman tunability (2500 - 4100 cm(-1)), using a femtosecond Ti:Sapphire pump and photonic crystal fiber output for the broadband synchronized Stokes pulse. Spectral chirp of the fs laser pulses was a user-variable parameter for optimization in a spectral focusing implementation of multimodal CARS microscopy. High signal-to-noise, high contrast multimodal imaging of live cells and tissues was achieved with pixel dwell times of 2-8 micros and low laser powers (< 30 mW total).