This PDF file contains the front matter associated with SPIE Proceedings Volume 10591, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

As techniques of retinal imaging have evolved, anatomic features that were only assessable in the laboratory have become available in the clinic for patient care. The retinal capillaries were initially described on microscope sections in the pathology laboratory. As optical methods have advanced these features have become part of the routine clinical landscape inspected daily by physicians. This paper briefly traces the evolution of these techniques and shows how they fit into the modern diagnostic armamentarium of ophthalmic retinal care.

Subretinal hyper-reflective material (SHRM) seen on optical coherence tomography (OCT) is thought to be a collection of fibrous tissues and vascular networks that are identified in age-related macular degeneration (ARMD). We have carried out a retrospective analysis of 91 OCT scans of neovascular ARMD subtypes including classic and occult choroidal neovascularization (CNV) and retinal angiomatous proliferation (RAP). All three subtypes received ranibizumab, an anti-vascular endothelial growth factor (Anti-VEGF) intravitreal injections on an as-needed basis following the loading doses. Volumes of SHRM were calculated using caliper measurements of maximal height and length of SHRM seen on OCT. The ellipsoid formula derived from tumour models was used to calculate the volume. It was found that occult CNV and RAP have larger SHRM volumes than those of classic CNV. SHRM volumes reduced overall following loading doses of Anti-VEGF injections at 4 months in all three subtypes. However, a rebound increase in volume was noticed in both occult CNV and RAP cohort at 12 months despite the initial, steeper reductions in the subtypes. These findings were consistent with the data seen in volume measurement using Topcon’s automated segmentation algorithm in a smaller cohort of patients. We propose that SHRM should be used as a potential biomarker to quantify both disease progression and prognosis of neovascular ARMD alongside other conventional methods.

Ultra high axial resolution (UHR) was demonstrated early in the development of optical coherence tomography (OCT), but has not yet reached clinical practice. We present the combination of supercontinuum light source and line field (LF-) OCT as a technical and economical route to get UHR-OCT into clinic and other OCT application areas. We directly compare images of a human donor cornea taken with low and high resolution current generation clinical OCT systems with UHR-LF-OCT. These images highlight the massive information increase of UHR-OCT. Application to pharmaceutical pellets, and the functionality and imaging performance of different imaging spectrograph choices for LF- OCT are also demonstrated.

We assessed the spectral sensitivity of the pupillary light reflex in mice using a high power super continuum white light (SCWL) source in a dual wavelength configuration. This novel approach was compared to data collected from a more traditional setup using a Xenon arc lamp fitted with monochromatic interference filters. Irradiance response curves were constructed using both systems, with the added benefit of a two-wavelength, equivocal power, output using the SCWL. The variables applied to the light source were intensity, wavelength and stimulus duration through which the physiological output measured was the minimum pupil size attained under such conditions. We show that by implementing the SCWL as our novel stimulus we were able to dramatically increase the physiological usefulness of our pupillometry system.

Retinal OCT image segmentation is a precursor to subsequent medical diagnosis by a clinician or machine learning algorithm. In the last decade, many algorithms have been proposed to detect retinal layer boundaries and simplify the image representation. Inspired by the recent success of superpixel methods for pre-processing natural images, we present a novel framework for segmentation of retinal layers in OCT volume data. In our framework, the region of interest (e.g. the fovea) is located using an adaptive-curve method. The cell layer boundaries are then robustly detected firstly using 1D superpixels, applied to A-scans, and then fitting active contours in B-scan images. Thereafter the 3D cell layer surfaces are efficiently segmented from the volume data. The framework was tested on healthy eye data and we show that it is capable of segmenting up to 12 layers. The experimental results imply the effectiveness of proposed method and indicate its robustness to low image resolution and intrinsic speckle noise.

This work is dedicated to development of the OCT system capable to visualize blood vessel network for everyday clinical use. Following problems were solved during the development: compensation of specific natural tissue displacements, induced by contact scanning mode and physiological motion of patients (e.g. respiratory and cardiac motions) and on-line visualization of vessel net to provide the feedback for system operator.

A detailed assessment of embryo development would assist biologists with selecting the most suitable embryos for transfer leading to higher pregnancy rates. Currently, only low resolution microscopy is employed to perform this assessment. Although this method delivers some information on the embryo surface morphology, no specific details are shown related to its inner structure. Using a Master-Slave Swept-Source Optical Coherence Tomography (SS-OCT), images of bovine embryos from day 7 after fertilization were collected from different depths. The dynamic changes inside the embryos were examined, in detail and in real-time from several depths. To prove our ability to characterize the morphology, a single embryo was imaged over 26 hours. The embryo was deprived of its life support environment, leading to its death. Over this period, clear morphological changes were observed.

In achieving minimally invasive accessibility to deeply located regions the size of the imaging probes is important. We demonstrated full-field optical coherence tomography (FF-OCM) using an ultrathin forward-imaging short multimode fiber (SMMF) probe of 50 μm core diameter, 125 μm diameter, and 7.4 mm length for optical communications. The axial resolution was measured to be 2.14 μm and the lateral resolution was also evaluated to be below 4.38 μm using a test pattern (TP). The spatial mode and polarization characteristics of SMMF were evaluated. Inserting SMMF to in vivo rat brain, 3D images were measured and 2D information of nerve fibers was obtained. The feasibility of an SMMF as an ultrathin forward-imaging probe in FF-OCM has been demonstrated.

Multimode optical fibres are attractive for biomedical and industrial applications such as endoscopes because of the small cross section and imaging resolution they can provide in comparison to widely-used fibre bundles. However, the image is randomly scrambled by propagation through a multimode fibre. Even though the scrambling is unpredictable, it is deterministic, and therefore the scrambling can be reversed. To unscramble the image, we treat the multimode fibre as a linear, disordered scattering medium. To calibrate, we scan a focused beam of coherent light over thousands of different beam positions at the distal end and record complex fields at the proximal end of the fibre. This way, the inputoutput response of the system is determined, which then allows computational reconstruction of reflection-mode images. However, there remains the problem of illuminating the tissue via the fibre while avoiding back reflections from the proximal face. To avoid this drawback, we provide here the first preliminary confirmation that an image can be transferred through a 2x2 fibre coupler, with the sample at its distal port interrogated in reflection. Light is injected into one port for illumination and then collected from a second port for imaging.

Supercontinuum (SC) sources based on optical fibers exhibiting flat and normal group velocity dispersion (GVD) in the entire wavelength region of interest, so-called all-normal dispersion (ANDi) fibers, have become a well-accepted alternative to the more conventional approach of pumping fibers in the anomalous GVD region. The conservation of a single ultrashort pulse and the suppression of noise-amplifying nonlinear effects during SC generation make ANDi fibers particularly attractive for ultrafast and noise-sensitive applications. Here we discuss current possibilities and future prospects, but also the limitations of ANDi fiber SC sources, focusing particularly on coherence and noise performance for various pump pulse regimes.

Commercially available silica-fiber-based and ultra-broadband supercontinuum (SC) sources are typically generated by pumping close to the zero-dispersion wavelength (ZDW) of a photonic crystal fiber (PCF), using high-power picosecond or nanosecond laser pulses. Despite the extremely broad bandwidths, such sources are characterized by large intensity fluctuations, limiting their performance for applications in imaging such as optical coherence tomography (OCT). An approach to eliminate the influence of noise sensitive effects is to use a so-called all-normal dispersion (ANDi) fiber, in which the dispersion is normal for all the wavelengths of interest. Pumping these types of fibers with short enough femtosecond pulses allows to suppress stimulated Raman scattering (SRS), which is known to be as noisy process as modulation instability (MI), and coherent SC is generated through self-phase modulation (SPM) and optical wave breaking (OWB). In this study, we show the importance of the pump laser and fiber parameters in the design of low-noise ANDi based SC sources, for application in OCT. We numerically investigate the pulse-to-pulse fluctuations of the SC, calculating the relative intensity noise (RIN) as a function of the pump pulse duration and fiber length. Furthermore, we experimentally demonstrate the role of the fiber length on the RIN of the ANDi SC, validating the results calculated numerically. In the end, we compare the RIN of a commercial SC source based on MI and the ANDi SC source developed here, which shows better noise performance when it is carefully designed.

Ultrahigh resolution optical coherence tomography (UHR-OCT) is enabled by using a broad band source. Simultaneously, this makes the OCT image more sensitive to dispersion mismatch in the interferometer. In spectral domain OCT, dispersion left uncompensated in the interferometer and detector non-linearities lead together to an unknown chirp of the detected interferogram. One method to compensate for the chirp is to perform a pixel-wavenumber calibration versus phase that requires numerical extraction of the phase. Typically a Hilbert transform algorithm is employed to extract the optical phase versus wavenumber for calibration and dispersion compensation. In this work we demonstrate UHR-OCT at 1300 nm using a Super continuum source and highlight the resolution constraints in using the Hilbert transform algorithm when extracting the optical phase for calibration and dispersion compensation. We demonstrate that the constraints cannot be explained purely by the numerical errors in the data processing module utilizing the Hilbert transform but must be dictated by broadening mechanisms originating from the experimentally obtained interferograms.

We describe a swept source based phase sensitive optical coherence microscopy (OCM) system for photothermal imaging of gold nanorods (GNR). The phase sensitive OCM system employed in the study has a displacement sensitivity of 0.17 nm to vibrations at single frequencies below 250 Hz. We demonstrate the generation of phase maps and confocal phase images. By displaying the difference between successive confocal phase images, we perform the confocal photothermal imaging of accumulated GNRs behind a glass coverslip and behind the scattering media separately. Compared with two-photon luminescence (TPL) detection techniques reported in literature, the technique in this study has the advantage of a simplified experimental setup and provides a more efficient method for imaging the aggregation of GNR. However, the repeatability performance of this technique suffers due to jitter noise from the swept laser source.

Photonic time stretch (PTS) has enabled real time spectral domain optical coherence tomography (OCT). However, this method generates a torrent of massive data at GHz stream rate, which requires capturing as per Nyquist principle. If the OCT interferogram signal is sparse in Fourier domain, which is always true for samples with limited number of layers, it can be captured at lower (sub-Nyquist) acquisition rate as per compressive sensing method. In this work we report a data compressed PTS-OCT system based on photonic compressive sensing with 66% compression with low acquisition rate of 50MHz and measurement speed of 1.51MHz per depth profile. A new method has also been proposed to improve the system with all-optical random pattern generation, which completely avoids electronic bottleneck in traditional binary pseudorandom binary sequence (PRBS) generators.

This paper illustrates specific features and use of optical coherence tomography (OCT) in the raster-scanning and in comparison in the full field version of this imaging technique. Cases for nondestructive testing are discussed alongside other application schemes. In particular monitoring time-dependent processes and probing of birefringent specimens are considered here. In the context of polymer testing birefringence mapping may often provide information about internal strain and stress states. Recent results obtained with conventional raster-scanning OCT systems, with (dual and single-shot) full field OCT configurations, and with polarization-sensitive versions of (full field) OCT are presented here.

Spectroscopic spectral domain interferometry and spectroscopic optical coherence tomography combine depth information with spectrally-resolved localised absorption data. These additional data can improve diagnostics by giving access to functional information of the investigated sample. One possible application is measuring oxygenation levels at the retina for earlier detection of several eye diseases. Here measurements with different hollow glass tube phantoms are shown to measure the impact of a superficial absorbing layer on the precision of reconstructed attenuation spectra of a deeper layer. Measurements show that a superficial absorber has no impact on the reconstructed absorption spectrum of the deeper absorber. Even when diluting the concentration of the deeper absorber so far that an incorrect absorption maximum is obtained, still no influence of the superficially placed absorber is identified.

One of the most common fabrication techniques for dental ceramics is sintering, a process of heating of the ceramic to ensure densification. This occurs by viscous flow when the firing temperature is reached. Acceptable restorations require the alloy and ceramic to be chemically, thermally, mechanically, and aesthetically compatible. Thermal and mechanical compatibility include a fusing temperature of ceramic that does not cause distortion of the metal substructure. Decalibration of ovens used for firing of the ceramic layers for metal ceramic dental prostheses leads to stress and cracks in the veneering material, and ultimately to the failure of the restoration. 25 metal ceramic prostheses were made for this study. They were divided in five groups, each sintered at a different temperature: a group at the temperature prescribed by the producer, two groups at lower and two groups at higher temperatures set in the ceramic oven. An established noninvasive biomedical imaging method, swept source (SS) optical coherence tomography (OCT) was employed, in order to evaluate the modifications induced when using temperatures different from those prescribed for firing the samples. A quantitative assessment of the probes is performed by en-face OCT images, taken at constant depths inside the samples. The differences in granulation, thus in reflectivity allow for extracting rules-of-thumb to evaluate fast, by using only the prostheses currently produced the current calibration of the ceramic oven. OCT imaging can allow quick identification of the oven decalibration, to avoid producing dental prostheses with defects.

There are several important non carious lesions affecting the tooth structure, lesions which may be classified into four clinical forms of dental wear: abfraction, erosion, attrition and abrasion, and different types of root resorption. Search for new, non-invasive and fast methods able to detect and describe such injuries is of utmost importance. Optical coherence tomography (OCT) proved itself as an appropriate investigation method for several medical fields including ophthalmology, dermatology, cardiology etc. Our study reveals OCT preliminary investigations as a promising tool for detecting and evaluating of the mentioned lesions.

Prevention and allopathic medicine gained attention, since it is possible for dentists to prevent demineralization, using plants and natural substances with well-known efficacy. The purpose of this study was to present new methods for teeth remineralization. It was made a selection of 10 extracted teeth, maintained in physiological serum, with no color fading, decay or demineralization. It was induced demineralization, with ortho-phosphoric acid (concentration 45 %), for one minute. The probes were visually and with optical coherence tomography (OCT) inspected. The natural product and the bonding with additional nanoparticles of argent were created and applied on the demineralization zone of the both groups of teeth. Each tooth in the first group had one plain surface demineralized. The second group of teeth had a cavity prepared on one of each tooth’s side. The pastes were applied on the demineralized surfaces and in the demineralized cavities for two minutes. After time expired, the pasta applied on the first group of teeth was washed away; the bonding above the second group of teeth was light cured. The probes were again visually and with OCT inspected. It was observed an improvement in remineralizing the white marks on plain surfaces and in the created cavities, the OCT being able to detect different levels of remineralization. The efficacy of natural pasta depends on the time it is applied and the concentration of the different main substances. Also, the type of surface, plain or occlusal facets, may influence the substances’ penetration ability. The non-invasive specific feature of these products, low costs and safety are strong positive aspects of this method of remineralization. However, the natural process of remineralization is a long-lasting one; perfecting the main substances in order to accelerate the process, in addition to several in vivo studies would be necessary to be fulfilled.

The study purpose is to highlight dental hard tissues changes of a tooth with dental wear as a consequence of excessive occlusal forces, using OCT. Methods: a central incisor extracted for periodontal reason was cleaned and it was embedded in a black acrylic resin block. The block was sectioned along the longitudinal axis of the tooth and prepared for OCT analysis. Results: The OCT signal showed differences between the labial and palatal dental hard tissue structures, even in areas without excessive occlusal solicitations. Conclusion: The OCT signal highlights changes of dental hard tissues structures according to excessive occlusal solicitations areas.

INTRODUCTION: The use of Optical Coherence Tomography (OCT) as a non-invasive and high precision quantitative information providing tool has been well established by researches within the last decade. The marginal discrepancy values can be scrutinized in optical biopsy made in three dimensional (3D) micro millimetre scale and reveal detailed qualitative and quantitative information of soft and hard tissues. OCT-based high resolution 3D images can provide a significant impact on finding recurrent caries, restorative failure, analysing the precision of crown preparation, and prosthetic elements marginal adaptation error with the gingiva and dental hard tissues. During the CAD/CAM process of prosthodontic restorations, the circumvent of any error is important for the practitioner and the technician to reduce waste of time and material. Additionally, OCT images help to achieve a new or semi-skilled practitioner to analyse their crown preparation works and help to develop their skills faster than in a conventional way. The aim of this study is to highlight the advantages of OCT in high precision prosthodontic restorations.

MATERIALS AND METHODS: 25 preparations of frontal and lateral teeth were performed for 7 different patients. The impressions of the prosthetic fields were obtained both using a conventional optoelectronic system (Apolo Di, Syrona) and a Spectral Domain using OCT (Dental prototype, working at 860 nm). For the conventional impression technique the preparation margins were been prelevated by gingival impregnated cords. No specific treatments were performed by the OCT impression technique.

RESULTS: The scanning performed by conventional optoelectronic system proved to be quick and accurate in terms of impression technology. The results were represented by 3D virtual models obtained after the scanning procedure was completed. In order to obtain a good optical impression a gingival retraction cord was inserted between the prepared tooth and the gingival tissue for a better elevation of the tooth cervical margin preparation. Spectral OCT was enforced in order to observe the quality but also the advantages coming from this technology. No special preparation was performed for this operation.

CONCLUSION: Considering these aspects, OCT could be used as a valuable tool for dental impression technology, being non-invasive but also non-destructive on the marginal gingival tissue, in comparison with conventional optoelectronic technology where the gingival retraction cord is still mandatory.

Swept Source optical coherence tomography (SS-OCT) has become a well established imaging modality for both medical and industrial diagnostic applications. A cross-sectional SS-OCT image is obtained by applying an inverse discrete Fourier transform (DFT) to axial interferogram measured in the frequency domain (k-space). Fourier inversion of the obtained interferogram typically produces a potentially overlapping conjugate mirror image, whose overlap could be avoided by restricting the object to have its highest reflectivity at its surface. However, this restriction may not be fulfilled when imaging a very thin object that is placed on a highly reflective surface, or imaging an object containing a contrast agent with high reflectivity. In this paper, we show that oversampling of the SS-OCT signal in k-space would overcome the need for such restriction on the object. Our result is demonstrated using SS-OCT images of Axolotl salamander eggs.

Optical coherence tomography (OCT) images scattering tissues with 5 to 15 μm resolution. This is usually not sufficient for a distinction of cellular and subcellular structures. Increasing axial and lateral resolution and compensation of artifacts caused by dispersion and aberrations is required to achieve cellular and subcellular resolution. This includes defocus which limit the usable depth of field at high lateral resolution. OCT gives access the phase of the scattered light and hence correction of dispersion and aberrations is possible by numerical algorithms. Here we present a unified dispersion/aberration correction which is based on a polynomial parameterization of the phase error and an optimization of the image quality using Shannon’s entropy. For validation, a supercontinuum light sources and a costume-made spectrometer with 400 nm bandwidth were combined with a high NA microscope objective in a setup for tissue and small animal imaging. Using this setup and computation corrections, volumetric imaging at 1.5 μm resolution is possible. Cellular and near cellular resolution is demonstrated in porcine cornea and the drosophila larva, when computational correction of dispersion and aberrations is used. Due to the excellent correction of the used microscope objective, defocus was the main contribution to the aberrations. In addition, higher aberrations caused by the sample itself were successfully corrected. Dispersion and aberrations are closely related artifacts in microscopic OCT imaging. Hence they can be corrected in the same way by optimization of the image quality. This way microscopic resolution is easily achieved in OCT imaging of static biological tissues.

Optical coherence tomography (OCT) images are capable of detecting vascular plaque by using the full set of 26 Haralick textural features and a standard K-means clustering algorithm. However, the use of the full set of 26 textural features is computationally expensive and may not be feasible for real time implementation. In this work, we identified a reduced set of 3 textural feature which characterizes vascular plaque and used a generalized Fuzzy C-means clustering algorithm. Our work involves three steps: 1) the reduction of a full set 26 textural feature to a reduced set of 3 textural features by using genetic algorithm (GA) optimization method 2) the implementation of an unsupervised generalized clustering algorithm (Fuzzy C-means) on the reduced feature space, and 3) the validation of our results using histology and actual photographic images of vascular plaque. Our results show an excellent match with histology and actual photographic images of vascular tissue. Therefore, our results could provide an efficient pre-clinical tool for the detection of vascular plaque in real time OCT imaging.

In this work we investigate three techniques for estimation of the non-linear phase present due to defocus in optical coherence tomography, and apply them with the angular spectrum method. The techniques are: Least squares fitting the of unwrapped phase of the angular spectrum, iterative optimization, and sub-aperture correlations. The estimated phase of a single en-face image is used to extrapolate the non-linear phase at all depths, which in the end can be used to correct the entire 3-D tomogram, and any other tomogram from the same system.

Optical coherence tomography systems suffer from noise that could reduce ability to interpret reconstructed images correctly. We describe a method to increase the signal-to-noise ratio of swept-source optical coherence tomography (SSOCT) using oversampling in k-space. Due to this oversampling, information redundancy would be introduced in the measured interferogram that could be used to reduce white noise in the reconstructed A-scan. We applied our novel scaled nonuniform discrete Fourier transform to oversampled SS-OCT interferograms to reconstruct images of a salamander egg. The peak-signal-to-noise (PSNR) between the reconstructed images using interferograms sampled at 250MS/s andz50MS/s demonstrate that this oversampling increased the signal-to-noise ratio by 25.22 dB.

Optical coherence tomography (OCT) is a non-destructive optical technique, which uses a light source with a wide band width that focuses on a point in the sample to determine the distance (strictly, the optical path difference, OPD) between this point and a reference surface. The point can be superficial or at an interior interface of the sample (transparent or semitransparent), allowing topographies and / or tomographies in different materials. The Michelson interferometer is the traditional experimental scheme for this technique, in which a beam of light is divided into two arms, one the reference and the other the sample. The overlap of reflected light in the sample and in the reference generates an interference signal that gives us information about the OPD between arms. In this work, we work on the experimental configuration in which the reference signal and the reflected signal in the sample travel on the same arm, improving the quality of the interference signal. Among the most important aspects of this improvement we can mention that the noise and errors produced by the relative reference–sample movement and by the dispersion of the refractive index are considerably reduced. It is thus possible to obtain 3D images of surfaces with a spatial resolution in the order of microns. Results obtained on the topography of metallic surfaces, glass and inks printed on paper are presented.

Typically, swept source optical coherence tomography (SS-OCT) imaging instruments are capable of a longer axial range than their camera based (CB) counterpart. However, there are still various applications that would take advantage for an extended axial range. In this paper, we propose an interferometer configuration that can be used to extend the axial range of the OCT instruments equipped with conventional swept-source lasers up to a few cm. In this configuration, the two arms of the interferometer are equipped with adjustable optical path length rings. The use of semiconductor optical amplifiers in the two rings allows for compensating optical losses hence, multiple paths depth reflectivity profiles (Ascans) can be combined axially. In this way, extremely long overall axial ranges are possible. The use of the recirculation loops produces an effect equivalent to that of extending the coherence length of the swept source laser. Using this approach, the achievable axial imaging range in SS-OCT can reach values well beyond the limit imposed by the coherence length of the laser, to exceed in principle many centimeters. In the present work, we demonstrate axial ranges exceeding 4 cm using a commercial swept source laser and reaching 6 cm using an “in-house” swept source laser. When used in a conventional set-up alone, both these lasers can provide less than a few mm axial range.

A general theoretical framework is described to obtain the advantages and the drawbacks of two novel Fourier Domain Optical Coherence Tomography (OCT) methods denoted as Master/Slave Interferometry (MSI) and its extension denoted as Complex Master/Slave Interferometry (CMSI). Instead of linearizing the digital data representing the channeled spectrum before a Fourier transform can be applied to it (as in OCT standard methods), channeled spectrum is decomposed on the basis of local oscillations. This replaces the need for linearization, generally time consuming, before any calculation of the depth profile in the range of interest. In this model two functions, g and h, are introduced. The function g describes the modulation chirp of the channeled spectrum signal due to nonlinearities in the decoding process from wavenumber to time. The function h describes the dispersion in the interferometer. The utilization of these two functions brings two major improvements to previous implementations of the MSI method. The paper details the steps to obtain the functions g and h, and represents the CMSI in a matrix formulation that enables to implement easily this method in LabVIEW by using parallel programming with multi-cores.

Medical imaging using Optical Coherence Tomography (OCT) provides clinicians with 3D, high resolution reconstructions of microscopic structures, in depth. It has been initially developed for ophthalmology, in order to scan the retinas of patients to diagnose illness. The quality of the images depends upon their axial and lateral resolutions and the properties of the light being used. Research using a polygon mirror (PM) as a spectral filter in Swept Source OCT (SS-OCT) has resulted in a variety of different experimental arrangements. Although the application of PM-based SS-OCT sources has been successfully demonstrated, the combination of their components’ fundamental properties and the overall impact they have on imaging performance is rarely reported. A more detailed examination of these properties would lead to a full description of their operation and to the best methods to employ if system performance is to be maximised. This work presents our current findings of on-going research into the optimisation of PM-based SS-OCT systems. A swept source spectral filter, consisting of a collimator, a transmission grating, a two-lens telescope and an off-axis PM with an end reflector mirror has been evaluated experimentally and compared with theoretical predictions. The system’s performance has been compared for two different fibre collimators. Although the beam width on the grating is different for each of the two collimators, the spot size at the PM facet is made the same by selecting appropriate focal lengths. An improvement in the signal roll-off at the interferometer output of ~1.0 dB/mm was obtained when using a 3.4 mm collimator compared to a 1.5 mm collimator.

Backward supercontinuum generation seeded by random distributed feedback fiber laser is proposed and demonstrated. Fully-distributed random lasing is firstly used as an effective pump laser to generate supercontinuum in the hybrid configuration with TrueWave fiber and dispersion compensated fiber. Supercontinuum in both propagating directions are generated through the collective nonlinear effects. It is found backward supercontinuum possesses much lower dynamical amplitude fluctuation in time domain. Random distributed lasing pumped supercontinuum generation not only enrich both the research scope of supercontinuum and random fiber laser, but also provide a practical way for development of stable broadband sources for OCT applications.