Spatial distribution of single-photon and two-photon fluorescence light in scattering media: Monte Carlo simulation

Modeling nonlinear optical microscopy in scattering media, part II. Radiation from focal volume to far-field: tutorial

The development and application of nonlinear optical (NLO) microscopy methods in biomedical research has experienced rapid growth over the past three decades. Despite the compelling power of these methods, optical scattering limits their practical use in biological tissues. This tutorial offers a model-based approach illustrating how analytical methods from classical electromagnetism can be employed to comprehensively model NLO microscopy in scattering media. In Part I, we quantitatively model focused beam propagation in non-scattering and scattering media from the lens to focal volume. In Part II, we model signal generation, radiation, and far-field detection. Moreover, we detail modeling approaches for major optical microscopy modalities including classical fluorescence, multi-photon fluorescence, second harmonic generation, and coherent anti-Stokes Raman microscopy.

Modeling nonlinear optical microscopy in scattering media, part I. Propagation from lens to focal volume: tutorial

The development and application of nonlinear optical (NLO) microscopy methods in biomedical research have experienced rapid growth over the past three decades. Despite the compelling power of these methods, optical scattering limits their practical use in biological tissues. This tutorial offers a model-based approach illustrating how analytical methods from classical electromagnetism can be employed to comprehensively model NLO microscopy in scattering media. In Part I, we quantitatively model focused beam propagation in non-scattering and scattering media from the lens to focal volume. In Part II, we model signal generation, radiation, and far-field detection. Moreover, we detail modeling approaches for major optical microscopy modalities including classical fluorescence, multi-photon fluorescence, second harmonic generation, and coherent anti-Stokes Raman microscopy.

Scattering of Sculpted Light in Intact Brain Tissue, with implications for Optogenetics

Optogenetics uses light to control and observe the activity of neurons, often using a focused laser beam. As brain tissue is a scattering medium, beams are distorted and spread with propagation through neural tissue, and the beam's degradation has important implications in optogenetic experiments. To address this, we present an analysis of scattering and loss of intensity of focused laser beams at different depths within the brains of zebrafish larvae. Our experimental set-up uses a 488 nm laser and a spatial light modulator to focus a diffraction-limited spot of light within the brain. We use a combination of experimental measurements of back-scattered light in live larvae and computational modelling of the scattering to determine the spatial distribution of light. Modelling is performed using the Monte Carlo method, supported by generalised Lorenz-Mie theory in the single-scattering approximation. Scattering in areas rich in cell bodies is compared to that of regions of neuropil to identify the distinct and dramatic contributions that cell nuclei make to scattering. We demonstrate the feasibility of illuminating individual neurons, even in nucleus-rich areas, at depths beyond 100 μm using a spatial light modulator in combination with a standard laser and microscope optics.

Detection and analysis of tumor fluorescence using a two-photon optical fiber probe

The utility of a two-photon optical fiber fluorescence probe (TPOFF) for sensing and quantifying tumor fluorescent signals was tested in vivo. Xenograft tumors were developed in athymic mice using MCA207 cells expressing green fluorescent protein (GFP). The TPOFF probe was able to detect ex vivo fluorescence from excised tumors containing as little as 0.3% GFP-expressing cells. TPOFF results were similar to both flow-cytometric analysis of tumor cells after isolation and suspension, and fluorescence determined by microscope images of cryosectioned tumors. TPOFF was then used to measure GFP fluorescence from tumors in live mice. The fiber probe detected fluorescently-labeled Herceptin antibody targeted to HER2-expressing tumors in severe combined immunodeficient mice. Dendrimer nanoparticles targeted by folic acid and having 6-TAMRA as a fluorescent probe were also used to label KB cell tumors in vivo. The fiber probe documented a fourfold increase in tumor fluorescence in animals that received the targeted dendrimer. These results suggest TPOFF can be used as a minimally invasive system for identifying tumor markers and monitoring drug therapy.

Monte Carlo modeling of optical coherence tomography imaging through turbid media

We combine a Monte Carlo technique with Mie theory to develop a method for simulating optical coherence tomography (OCT) imaging through homogeneous turbid media. In our model the propagating light is represented by a plane wavelet; its line propagation direction and path length in the turbid medium are determined by the Monte Carlo technique, and the process of scattering by small particles is computed according to Mie theory. Incorporated into the model is the numerical phase function obtained with Mie theory. The effect of phase function on simulation is also illustrated. Based on this improved Monte Carlo technique, OCT imaging is directly simulated and phase information is recorded. Speckles, resolution, and coherence gating are discussed. The simulation results show that axial and transversal resolutions decrease as probing depth increases. Adapting a light source with a low coherence improves the resolution. The selection of an appropriate coherence length involves a trade-off between intensity and resolution.

Spatial localization of cardiac optical mapping with multiphoton excitation

Depth and radius of regions interrogated by cardiac optical mapping with a laser beam depend on photon travel inside the heart. It would be useful to limit the range of depth and radius interrogated. We modeled the effects of a condensing lens to concentrate laser light at a target depth inside the heart, and near infrared excitation to increase penetration and produce two-photon absorption. A Monte Carlo simulation that incorporated a 0.55-NA lens, and absorption and scattering of 1064- or 488-nm laser light in 3-D cardiac tissue indicated the distribution of excitation fluence inside the tissue. A subsequent simulation incorporating absorption and scattering of transmembrane voltage-sensitive fluorescence (wavelength 669 nm) indicated locations from which fluorescence photons exiting the tissue surface originated. The results indicate that mapping at depths up to 300 microm in hearts can provide significant improvement in localization over existing cardiac optical mapping. The estimated interrogation region is sufficiently small to examine cardiac events at a cellular or subcellular scale and may allow mapping at various depths in the heart.

Reply to comment on "Excitation with a focused, pulsed optical beam in scattering media: diffraction effects."

We address the issues that were raised by Tycho and Jørgensen [Appl. Opt. 41, 4709 (2002)] concerning our strategy [Appl. Opt. 39, 5244 (2000)] for incorporating the wave properties of light in the description of a propagating focused excitation beam in a highly scattering medium. We explain that the strategy is consistent with the Huygens-Fresnel principle and does not violate the energy conservation principle.

Three-dimensional observation of internal defects in semiconductor crystals by use of two-photon excitation

We present a method of three-dimensional observation of internal defects in semiconductor crystals for blue lasers by use of two-photon excitation. We excite photoluminescence by using a two-photon process. Since semiconductor materials have intrinsically high absorption in the short-wavelength region, the excitation light of photoluminescence is largely absorbed by the crystals. It is difficult to observe defects in deep regions. Two-photon excitation can overcome this limitation because near-infrared light can be used instead of blue light for excitation of photoluminescence, and the near-infrared light is absorbed at only the focused point. The excitation light can penetrate into the deep regions of the crystal. We succeeded in observing defects in a ZnSe crystal ~200 microm below the crystal surface. This is, as far as we know, the first demonstration of the observation of three-dimensional structures of defects in semiconductor crystals by the use of two-photon excitation.

Two-photon microscopy in brain tissue: parameters influencing the imaging depth

Light scattering by tissue limits the imaging depth of two-photon microscopy and its use for functional brain imaging in vivo. We investigate the influence of scattering on both fluorescence excitation and collection, and identify tissue and instrument parameters that limit the imaging depth in the brain. (i) In brain slices, we measured that the scattering length at lambda=800 nm is a factor 2 higher in juvenile cortical tissue (P14-P18) than in adult tissue (P90). (ii) In a detection geometry typical for in vivo imaging, we show that the collected fraction of fluorescence drops at large depths, and that it is proportional to the square of the effective angular acceptance of the detection optics. Matching the angular acceptance of the microscope to that of the objective lens can result in a gain of approximately 3 in collection efficiency at large depths (>500 microm). A low-magnification (20x), high-numerical aperture objective (0.95) further increases fluorescence collection by a factor of approximately 10 compared with a standard 60x-63x objective without compromising the resolution. This improvement should allow fluorescence measurements related to neuronal or vascular brain activity at >100 microm deeper than with standard objectives.

Excitation with a focused, pulsed optical beam in scattering media: diffraction effects

To gain a better understanding of the spatiotemporal problems that are encountered in two-photon excitation fluorescence imaging through highly scattering media, we investigate how diffraction affects the three-dimensional intensity distribution of a focused, pulsed optical beam propagating inside a scattering medium. In practice, the full potential of the two-photon excitation fluorescence imaging is unrealized at long scattering depths, owing to the unwanted temporal and spatial broadening of the femtosecond excitation light pulse that reduces the energy density at the geometric focus while it increases the excitation energy density in the out-of-focus regions. To analyze the excitation intensity distribution, we modify the Monte Carlo-based photon-transport model to a semi-quantum-mechanical representation that combines the wave properties of light with the particle behavior of the propagating photons. In our model the propagating photon is represented by a plane wave with its propagation direction in the scattering medium determined by the Monte Carlo technique. The intensity distribution in the focal region is given by the square of the linear superposition of the various plane waves that arrive at different incident angles and optical path lengths. In the absence of scattering, the propagation model yields the intensity distribution that is predicted by the Huygens-Fresnel principle. We quantify the decrease of the energy density delivered at the geometric focus as a function of the optical depth to the mean-free-path ratio that yields the average number of scattering events that a photon encounters as it propagates toward the focus. Both isotropic and anisotropic scattering media are considered. Three values for the numerical aperture (NA) of the focusing lens are considered: NA = 0.25, 0.5, 0.75.