Photorefractive holography is a whole-field, coherence-gating technique for 3-D imaging through turbid media that offers a unique mechanism to discriminate against a background of diffuse light. In contrast to the well-established technique of optical coherence tomography, it is a whole-field imaging technique and may be implemented with light sources of arbitrary spatial coherence, including low cost LEDs and broad-stripe, multimode diode lasers. One drawback of using broadband sources, such as LEDs, for off-axis holographic imaging is the 'walk-off' resulting from the short temporal coherence length that limits the field-of-view. Furthermore, the non-collinear geometry required for off-axis holography can introduce significant image aberration. In this paper we discuss these design considerations for various sources. We have addressed the issue of walk-off for sources of arbitrary bandwidth and have designed an off-axis holographic imaging system based on a Michelson interferometer with a collinear beam geometry that minimizes aberration. In this paper we review our work with high-powered LEDs and discuss these issues associated with spatially incoherent sources. Also, we present a novel, spatially coherent, broadband diode-pumped laser source that may also find application in OCT.

We explore in this paper a new method to perform time-resolved measurements of the diffuse light transmitted through a thick turbid medium. This method is based on the analysis of the speckle fluctuations due a wavelength modulated source. A time resolution of about 50 ps is already achieved, and we expect to improve this result soon. This method could allow the design of low cost setups to perform such measurements.

Time-efficient Monte Carlo models for fluorescence from layered tissue were developed. The computation time is reduced significantly by recognizing symmetry properties of the problem, and by reversing computation of the photon paths for the fluorescence light. Further reduction is obtained by using a white Monte Carlo approach, which enables scaling of the results to the desired optical properties after the simulation. The methods reduce computation time more than two orders of magnitude compared with conventional Monte Carlo code.

Presented here are methods of visualization of the retina, specifically the optic nerve, based on transversal OCT imaging and on the operation of a standalone dual channel OCT transversal/confocal system purpose built for the eye. We also demonstrate that enhanced measurement capabilities of parameters in the fundus are possible using these two channels. This is made possible owing to the unique guidance capability of the confocal channel along with the information it provides about the transversal eye movements. A large number of transversal OCT images are collected from the eye. Then, by software, longitudinal (B-scan) or transversal cuts (C-scan) may be made, post-examination, on the stack of transversal images. The software inferred B-scans are shown to have a similar level of resolution to the hardware generated images after movement errors are corrected. We hope that the system equipped with this 3D software processing could reduce the irradiation time required to collect the necessary data from the eye. We show that 3D visualization of transversal OCT images along with the simultaneous confocal imaging allows easier alignment of the patient eye than is possible with longitudinal OCT imaging alone. We also demonstrate that by this method of transversal imaging, direct comparison may be made between quantitative SLO acquired measurements made on the eye fundus and OCT measurements. For the first time, measurements of volumes and areas in the optic nerve area are demonstrated using transversal OCT imaging, similar to procedures utilized by scanning laser ophthalmoscope users.

We present an interference microscope for non-invasive 3D imaging of biological tissues with a resolution better than 2 micrometer in the 3 dimensions. Our microscope is based on a Linnik interferometer associated with a parallel lock-in detection on a CCD camera and produces en face tomographic images with full-field illumination. We describe the performances of our microscope and we present images obtained from various biological tissues.

We describe an improved setup for differential phase contrast OCT. Phase differences between transversally adjacent beams are derived from the phase functions of the interferometric signals. The accuracy of the system is demonstrated by imaging pure phase objects consisting of chromium layers containing steps of approximately 100 - 200 nm height. By imaging through a scattering layer of 3 mean free path thickness (single pass), phase imaging is still possible with phase noise of only +/- 2 degrees.

Dynamic tissue analysis using real-time measurements can lead to advanced diagnostics, particularly where processes evolve with time. We introduce a no moving parts optical coherence tomography (OCT) system that rapidly acquires sample data at a microseconds/data point sampling rate. The basic principle relies on using acousto-optic or Bragg cell devices. The proposed OCT system has attractive features of an acousto- optic scanning heterodyne interferometer and an acousto-optic variable optical delay line (ODL) operating in reflective mode. Experiments described demonstrate our high speed acousto-optically tuned optical scanning probe concept where optical delay lines can be selected a sub-microsecond speeds.

The correlation diffusion equation is used to calculate laser Doppler spectra for semi-infinite homogeneous and semi- infinite two-layered turbid media at different distances from the incident source. The influence of the optical coefficients on the laser Doppler spectra is show using the semi-infinite geometry. The two-layered geometry is applied to investigate the influence of the first layer (for example: the layers above the muscle in the extremities) and second layer (for example the muscle) on laser Doppler measurements. The possibility to extract the root-mean-square velocity of the blood cells in the second layer (muscle) using a semi-infinite homogeneous model is discussed.

The light propagation in the head models has been investigated to deduce the volume of the brain tissue sampled by NIR instrument. The light propagation in the adult head is strongly affected by the presence of low scattering cerebrospinal fluid (CSF) layer. Although the brain surface is folded with sulci filled with the CSF, the brain surface in the previous head models has had simple geometry such as a flat or curved surface. In this study, the light propagation in a realistic adult and neonatal head models of which geometry is generated from MRI scan of the head is predicted by Monte Carlo method. The adult neonatal head models consist of small square elements to represent complex geometry and each element specified its scattering and absorption coefficients. The delta-scattering algorithm is implemented to keep the computation time within reasonable limit. The spatial sensitivity profiles for various source-detector spacing are predicted and the difference in the effect of the CSF on the spatial sensitivity profiles in the adult and neonatal head models is discussed. The low scattering CSF affects the light propagation in the adult head model while the distortion in spatial sensitivity profiles in the neonatal head models is not significant.

A new spectroscopic OCT technique is introduced, spectroscopic frequency-domain OCT, and its application to measure depth resolved spectral absorption is described. The crucial parameters of this method like transversal, depth, and spectral resolution and their relations are discussed. As preliminary test of the feasibility of this method, a simple absorbing object, in the present case an IR filter glass plate is investigated. Since the filter attenuates shorter wavelengths stronger than longer ones, one expects a shift of the spectral weight to the IR domain, which is absent in the case of a non-absorbing object like a BK7 glass plate. This effect is demonstrated by the measurement. The results are then compared to theoretical calculations based on the well known characteristics of the filter glass plate.

The coherently emitted energy flux of thermal lamps is limited less to than 1 (mu) W and, therefore, is hardly useful in OCT. We describe a technique which enables the use of spatially incoherent light sources, like thermal light sources and transversal multi-mode lasers in the standard time-domain OCT technique. The beam power available for OCT depth scans is increased by combining several mutually incoherent OCT channels. This technique has advantages particularly in multiplex or parallel OCT: There is no coherent cross-talk between adjacent depth scans and the power available per pixel is not dependent on the number of parallel depth scans.

Optical Coherence Tomography (OCT), a new optical bioimaging technique was used to evaluate the state of mucosa in the urinary bladder. The state of mucosa of the bladder was evaluated in patients with prostatic adenoma (11 male patients) during the course of prostatectomy operation via a resection cytoscope. An OCT probe was inserted into the biopsy channel of a cystoscope. The sites to be imaged by OCT were determined visually and, after OCT study, underwent excisional biopsy and subsequent histological examination. Children (9 girls) were examined during diagnostic cystoscopy. Our analysis of diagnostic capabilities of OCT in urology relies on the comparison of OCT information on normal and morphologically altered tissues. OCT is able to provide objective data concerning the structure of mucosa of the bladder due to the difference in optical properties of different layers in tissue. The epithelium and the layers of connective tissue, both in norm and pathology, are clearly visualized in the tomograms. Our OCT study of healthy mucosa of the urinary bladder has demonstrated that the epithelium appears in the tomograms as an upper highly backscattering layer. An underlying optically less transparent layer, much greater in size than the previous one, corresponds to the connective tissue of the mucosa. Inside this layer, elongated poorly backscattering formations with clear contours are seen; they do not alter the longitudinal structure of the submucosal layer. These formations are blood vessels. Optical patterns characteristic of chronic inflammation are obtained. They correspond, as confirmed histologically, to liquid accumulation, cellular infiltration of mucosal layers, hypervascularization, and fibrosis. OCT information on proliferative processes, such as papillomatosis of the urinary bladder and squamous cell carcinoma, is analyzed. It is shown that OCT can reliably reveal edema of the mucous membrane of the bladder and identify the character of appearing elements, such bulla, granule and polyp. OCT can provide information on the structure of tissue by characterizing its thickness and scattering properties.

It is not known to what extent effects in extracerebral tissue influence non-invasive near infra-red optical measurement of cerebral arterial oxygenation saturation. Measurements were made at different positions on the forehead of six healthy adult male volunteers with arterial saturation near to 100%. The optical ratios between the pulse heights at different wavelengths were as expected from the spectral characteristics of hemoglobin, but showed an unacceptably large spread: the mean ratio between the 770 and 905 nm pulse heights was 0.69 (SD 0.08, range 0.50 - 0.95). We consider that this was due to pulsation of large extracranial arteries.

Multi-channel NIR system can obtain the topographic image of brain activity. Since the image is reconstructed from the change in optical density measured with the source-detector pairs, it is important to reveal the volume of tissue sampled by each source-detector pair. In this study, the light propagation in three-dimensional adult head model is calculated by hybrid radiosity-diffusion method. The model is a layered slab which mimics the extra cerebral tissue (skin, skull), CSF and brain. The change in optical density caused by the absorption change in a small cylindrical region of 10 mm in diameter at various positions in the brain is calculated. The greatest change in optical density can be observed when the absorber is located in the middle of the source and detector. When the absorber is located just below the source or detector, the change in optical density is almost half of that caused by the same absorber in the midpoint. The light propagation in the brain is strongly affected by the presence of non-scattering layer and consequently sensitive region is broadly distributed on the brain surface.

Spectroscopic investigations of the VIS-NIR range allow the objective determination of pigmentation, blood microcirculation and water content of human skin. Non- contacting in vivo measurements of the human skin of 50 volunteers reflect the clinical skin type well. Our correlation analysis yields that the red/infrared spectral range can be used for a determination of skin type. The observed strong spectral variations within the same group of skin type are likely based on the high biological variability of human skin and subjective clinically observed skin type. Therefore it can be useful to measure the full spectral range and to calculate a non-observed skin score with multivariate spectral methods. By multivariate analysis a correct classification of remittance spectra can be obtained. Time- depending spectral variations of dermal microcirculation can be measured at defined locations of the body, for instance the dynamics of oxygenation or blood volume in the skin of the fingertip. The cardial, pulmonal and vasomotoric waves of the micro- and macrocirculation are clearly visible at different wavelengths. The spectroscopic informations are important as an objective measure for the skin type evaluation, the penetration behavior of pharmaca, laser surgery, and therapy.

We present a versatile and compact fiber optic probe for real- time determination of the absorption and the reduced scattering coefficients from spatially resolved continuous wave diffuse reflectance measurements. The probe collects the diffuse reflectance at six distances in the range 0.6 - 7.8 mm at four arbitrary wavelengths, which were 660, 785, 805, and 974 nm in these experiments. The maximum sampling rate for one cycle of measurements including all four wavelengths is about 100 Hz. The absorption and the reduced scattering coefficients are extracted real-time from the probe measurements using multivariate calibration methods based on multiple polynomial regression and Newton-Raphson algorithms. The system was calibrated on a 6 X 7 matrix of Intralipid/ink phantoms with optical properties within typical biological ranges, e.g. at 785 nm, the ranges of the absorption and the reduced scattering coefficients, were 0 - 0.3/cm and 6 - 16/cm, respectively. Cross-validation tests showed that the mean prediction error, relative to the ranges of absorption and the reduced scattering coefficients were 2.8% and 1.3%, respectively.

A new method of ultrasound-modulated optical tomography is reported, in which the buried object in optical turbid media is imaged directly with frequency-domain signals by real-time FFT. Tunable near-infrared laser is modulated by focused ultrasound. 2D tomographic images are obtained through scanning and detecting the ultrasound-modulated optical signal. Multi-wavelengths laser and diverse frequency ultrasounds are used to study image quality.

We present a preliminary study, in vitro and in vivo, with a novel device for near-infrared tissue oximetry. The light sources used are two quasi-continuous-wave LEDs, emitting at 656 and 851 nm, and the detector is a photodiode. The data are acquired in back-scattering configuration, thus allowing the non-invasive characterization of thick tissues. Stability tests were performed by placing the optical probe on a tissue- like phantom and acquiring data for periods of time ranging from 5 to 40 minutes. No significant drifts in the DC signal were observed after a warm-up period of no more than 10 minutes. We performed reproducibility tests by repositioning the optical probe on the phantom for a number of times. We found a reproducibility better than 5% in the DC signal. We also present the results of a preliminary study conducted in vivo, on the calf muscle of human subjects. We report a comparison of the results obtained with the near-infrared oximeter with the values of blood oxygenation ctO₂ measured with conventional chemical tests.

A Monte Carlo simulation program with graphic user interface has been developed under Windows environment by Visual C++. The parameters of media and certain type of source-detector geometry can be set flexibly for different simulation conditions and purposes. Through graphical user interface, Monte Carlo modeling of photon migration process in tissue is displayed during simulation. Also the computation results can be visualized directly after simulation. The aims of work is to develop an integrated Monte Carlo Simulation environment which makes the computation of the complex process of photon-tissue interaction more intuitive and understandable.

Recently, it has been shown that clear regions within diffusive media can be accurately modelled within the diffusion approximation by means of a novel boundary condition or by an approximation to it. This can be directly applied to the study of light propagation in brain tissue, in which there exist clear regions, and in particular, the cerebro spinal fluid (CSF) layer under the skull. In this work, we present the effect that roughness in the boundary of non-diffusive regions has on the measured signal, since, in practice, the CSF layer is quite rough. The same conclusions can be extended to any diffusive medium which encloses rough non-diffusive regions. We will demonstrate with numerical calculations that the roughness statistics of the interfaces although not their actual profile must be known a priori in order to correctly predict the shape of the measured signal. We show that as the roughness increases, the effect of the non-diffusive region diminishes until it disappears, thus yielding data similar to those of a fully diffusive region.

Adequate modeling of light propagation in the complex and heterogeneous tissue of the human head is very important for quantitative near infrared spectroscopy and optical imaging. The presence of a clear and non-scattering CSF layer around the brain has been previously shown to strongly affect light propagation in the head. However the CSF layer is not totally filled with a non-scattering fluid and quite a few fine arachnoid trabeculae are actually present in the layer. In this study light propagation in an adult head model with discrete scatterers distributed within the CSF layer has been predicted by Monte Carlo simulation in order to investigate the effect of scattering caused by the arachnoid trabecula in the CSF layer. Results show that the presence of the arachnoid trabeculae affect the total optical path length, a parameter which can be directly measured by time-resolved measurement. However, the partial optical path length in the brain tissue, which relates the sensitivity of the near infrared spectroscopy signal to absorption changes in the brain is strongly affected by the CSF layer even in the presence of the arachnoid trabeculae. The increased partial optical path length results from an increased lateral spreading of the NIR light within the gray matter of the cortex.

We have developed a three-wavelength (759,797, and 833 nm) time-resolved spectroscopy (TRS) system as a tissue oxygenation monitor employing a time-correlated single photon counting method. This system achieved a high data acquisition rate and system miniaturization maintaining a high sensitivity and time resolution. Our system succeeded in accurately measuring concentrations of oxy-(HbO₂) and deoxyhemoglobin (Hb) by means of TRS data observation through studies using a phantom model and living tissue.

Diffuse optical imaging and tomography is of some interest in the diagnosis of testicular pathologies. For a clinical evaluation of 3D optical tomography a special laser scanning device as well as dedicated tomography algorithms have been developed. With the device we are able to obtain continuous- wave tomographic scans from an object under investigation using different laser wavelengths. Tomographic image reconstruction is based on the solution of the linearized inverse problem of optical absorption imaging for a three- dimensional volume. Priority is given to a spatial resolution adapted volume discretization and an efficient matrix solution algorithm based on singular value decomposition.

We have demonstrated the feasibility of tagging the photons trajectories with a focused ultrasonic field, to reveal optical contrast in biological tissues. 3D images have been obtained on real ex-vivo structures of animal as well as human tissues, through a thickness ranging from 2 cm to 4.5 cm. We are developing the coupling of this acousto-optical imaging with a traditional echograph.

In diffuse optical diffuse tomography (DOT) one attempts to reconstruct cross-sectional images of various body parts given data from near-infrared transmission measurements. The cross- sectional images, display the spatial distribution of optical properties, such as the absorption coefficient (mu) _(alpha ), the scattering coefficient (mu) _s, or a combination thereof. Most of the currently employed imaging algorithms are model- based iterative image reconstruction (MOBIIR) schemes that employ information about the gradient of a suitably defined objective function with respect to the optical properties. In this approach the image reconstruction problem is considered as a nonlinear optimization problem, where the unknowns are the values of optical properties throughout the medium to be reconstructed. It is well known that gradient-based schemes are inefficient in areas where the gradient is close to zero. These schemes often get caught in local minima close to the starting point of the search and have problems finding the global minimum. To overcome this problem, we propose to employ optimization algorithms that make use of evolution strategies. These schemes are in general much better suited to find global minima and may be a better choice for the image reconstruction problem in diffuse optical tomography.

NIR spectroscopy is a method principally capable of measuring tissue perfusion and oxygen saturation in subsurface and deeper tissue layers. In urology such perfusion related parameters are of some importance for the differentiation and evaluation of certain types of testicular pathologies. Among them are in the first place the differentiation between inflammations and torsion of the testis, both with similar symptoms, and the assessment of tissue viability in cases of torsion and necrosis, which is sometimes not sufficiently covered by sonography. Although NIR spectroscopy is the method of choice to measure blood oxygen saturation in tissue non- invasively the strong light scattering complicates spectroscopy quantification methods. Spatially-resolved spectroscopy (SRS) is one method to quantify absolute oxygen saturation and relative blood volume. To evaluate this method for in-vivo measurements we have developed a laser scanning device and evaluated quantification algorithms by help of numerical and experimental investigations. As first results suggest the method can in principle quantify absorption differences in tissue, oxygen saturation measurements, however, not work on the testis under the simplified assumptions made for other parts of the body.

The properties of the field re-emitted by an absorbing- fluorescent object embedded in a highly scattering media excited by a dual-interfering sources system are investigated. The geometry considered is transmission geometry with a breast like phantom. This study is based on simulations performed in the frequency domain with a finite difference method to solve the diffusion equation. The fluorescent detected field possessed the features of the interfering excitation pattern. This specific pattern allows to accurately localization of the inhomogeneity.

Phased array techniques, based on interference of diffuse photon density waves, have been demonstrated to be a sensitive method of detecting and localizing objects embedded in heavily scattering media. They have also been used as a sensitive method of detecting functional changes in the brain. Previous experiments using anti-phase sources have shown that the phase response to the scan of an object through the medium is (pi) radians regardless of object size. This provides information about the position of an object but none about the size. In this paper we demonstrate that controlling the relative phase between the sources enables different phase gradients to be set within the medium. The consequence of this is that the phase response is dependent on the size of an object while still maintaining the localization information. Furthermore, it is demonstrated that the phase response can be tuned to be most sensitive to the object size under investigation. The effect of source separation on the system response is also investigated. It is demonstrated that there is an optimum source separation to produce the maximum signal to noise ratio. Sources positioned close together provide too much destructive interference whereas sources well separated do not provide enough interference.

Research into the near-infrared biomedical optical imaging has produced a multitude of inverse imaging algorithms. Recent experience has shown that when these algorithms are tested with experimental data, they falter due to a mismatch between observed and simulated measurements. When considering measurements for imaging, one must consider both measurement and model error. If data is recorded properly, then measurement error tends to be normally distributed with a mean of zero. Model error can be biased and spatially correlated due to inaccuracies in the diffusion approximation, inaccurate parameter estimates, numerical error, and other factors. This contribution discusses trends in the measurement and model error observed from measurements on a single-pixel, frequency domain photon migration system developed for biomedical optical imaging. In order to reduce the model error bias, an empirical approach was applied to find experimental variables that significantly affect it. This approach reduced the mean of the model error on a test data set and produced a slight smoothing effect on its distribution. Image reconstruction attempts show that the modified data set produces an improved image over the image reconstructed from the raw data set. To our knowledge, this is the first time that model and measurement error information have been incorporated into a three dimensional image reconstruction algorithm.

Time-Correlated Single-Photon-Counting (TCSPC) systems have been often adopted to prepare more compact and less expensive instruments for medical imaging. Then it is important to investigate in detail the performances of each imaging system in measuring the optical properties of turbid media. Our experimental apparatus is composed by an Hamamatsu PLP02 pulsed diode laser at 824 nm with 1 Mhz repetition rate and a pulse duration of 40 ps. The signal has been collected from the investigated sample by means of fiber bundles and has been analyzed by an Edinburgh Instrument TCSPCS equipped with an 8 channels Hamamatsu multichannel plate R411OU-F008 and an SPC300 acquisition module. Solutions of distilled water and commercial Intralipid 10% at different concentrations have been investigated. To obtain optical parameters, the experimental data have been fitted with an analytic solution to the diffusion equation. Also the convolution effect of the measured Temporal Point Spread Functions (TPSF) by the Impulse Response Function (IRF) of the system has been investigated. A linear trend has been obtained for the reduced scattering coefficient (mu) '_s with concentration in solution of the scattering agent (Intralipid 10%) and an agreement within few percent has been reported with Mie theory predictions. The results here obtained confirm that all the details in fitting and convolution procedures become particularly important when slight difference in absorption and scattering coefficients have to be examined.

The absolute quantified measurement of haemoglobin skin blood saturation from collected reflectance spectra of the skin is complicated by the fact that the blood content of tissues can vary both in the spatial distribution and in the amount. These measurements require an understanding of which vascular bed is primarily responsible for the detected signal. Knowing the spatial detector depth sensitivity makes it possible to find the best range of different probe geometries for the measurements of signal from the required zones and group of vessels inside the skin. To facilitate this we have developed a Monte Carlo simulation to estimate the sampling volume offered by small source detector spacing (in the current report 250 micrometer, 400 micrometer and 800 micrometer) in the fiber-optic probes, and confocal microscope probe (the lens parameters are: diameter - 5 mm; focal length - 10 mm; the pinhole diameter is 10 micrometer). The optical properties of the modeled medium were taken to be the optical properties of the Caucasian type of skin tissues in visible range of the spectrum.

Spectral characteristic biophotoreceptor in the human eye were determined using reflected optical wave from the retina and in this measurement have been used microspectrometric techniques. This measurements is possible only in vitro. We present new non-invasive correlation method ocular low coherence interferometry, which can be used in ophthalmology. Number of observed fringes in the Michelson interferometer is connected with spectral width biophotorecipient across Wiener-Khinchine transformation in case using of the partial coherent light sources. Such method allow determinate spectral sensitivity rod if sounded light is low intensity and application optical filter permit value red and violet border human eye. Change width interference fringes it is possible simultaneity measurement space sensitivity of the human retina which is very important in the ophthalmic diagnostic.

The use of optical techniques for diagnostic purpose relies on the capability to measure the optical properties of healthy and pathological tissues and in appreciating their relative differences. In fact, a degree of contrast must exist between absorption and scattering coefficients for effective detection of a tissue alteration using optical imaging. In this contest, it is important to study the accuracy limits of different optical measurement techniques in order to establish their performances in recovering the optical parameters. The accuracy limit of laser techniques in the determination of optical methods have been recovered partly from literature and, as far as concerns time-resolved techniques, from experimental work carried on in our laboratory using a conventional time-resolved system. The results from this analysis allow us to better identify the role of different experimental techniques, which are generally proposed in optical imaging for diagnostic purpose.

Polarization sensitive technique is reported for visualization of eye scattering and birefringent inhomogeneities using digital subtracting of eye images captured by a CCD camera for two orthogonal polarizations of light forming images in the CCD camera. For fast capturing of images the polarization plane of the backscattered light is periodically switched at 90 degrees by an electro-optical PLZT phase plate. This plate is placed close to the CCD camera together with a sheet polarizer inserted between camera and the phase plate. Polarization plane is switched applying the voltage 1200 V to the phase plate at a rate of 7.5 Hz. The technique improves visualization due to diminishing of the impact of eye movements and due to accumulation of the imaging digital data.