FREE ELECTRONIC LIBRARY - Dissertations, online materials

Pages:     | 1 |   ...   | 3 | 4 || 6 | 7 |   ...   | 13 |

«Novel Biophotonic Imaging Techniques for Assessing Women’s Reproductive Health by Tyler Kaine Drake Department of Biomedical Engineering Duke ...»

-- [ Page 5 ] --

Broadband light from a superluminescent diode with a central wavelength, λ0, of 842.9 nm and a output power of 15.51 mW is coupled into a 1x8 tree coupler to create 6 parallel channels (2 remain unused) as shown in Figure 3.1.

The SLD has a bandwidth, Δλ, of 50.4 nm which results in a coherence length of 6.2 µ;m, as calculated from Eqn.


–  –  –

The six utilized channels are directed into six single- mode, 50/50 fiber couplers, which act to create distinct sample and reference arms for each channel.

The sample arm fibers are fusion spliced to pigtailed fibers that are positioned within into a silicon v- groove chip using UV- cure epoxy.

Light output by the fibers is imaged onto the sample as shown in Figure 3.1.

Lens L1 simultaneously collimates the beams exiting the fibers, and lens L2 focuses these beams to create a linear array of illumination points on the sample.

For the multiplexed reference arm, the six output fibers of the 50/50 fiber couplers are each fusion spliced to individual pigtailed fiber collimators mounted in a machined bracket that allows individual adjustment of each channel.

This is necessary 33 for path length matching from channel to channel, as well as path length matching for the entire group.

After each channel is individually path length matched, light from each reference arm is reflected by a mirror, located within a 6- mirror array, which can be translated to provide depth ranging.

The reflected light from each sample and reference arm is returned through the same fibers that delivered the light, and recombined by the 50/50 couplers before being directed to a high- resolution, six- channel spectrometer for detection.

The spectrometer has a resolution of 0.05 nm/pixel with a 180 nm wavelength range and allows simultaneous detection of all channels.

In the spectrometer, the combined fields are dispersed by 1200 lines/mm gratings onto linear CCDs.

The data is downloaded in real time via a USB 2.0 interface at a rate of 0.27 kHz, per six spectra.71

–  –  –

Labview software is used to acquire and analyze each spectrum.

First, background spectra are captured and subtracted from the signal, I, in Eqn.

(2.1) to remove noninterferometric and common path signal components.

Spectra are captured with the reference arm blocked, yielding IS, the sample arm intensity, with the sample 34 arm blocked, yielding IR, the reference arm intensity, and both arms blocked, yielding ID, the background intensity.

The background- subtracted spectra can then be calculated from the interferogram as captured at the spectrometer, IRAW,


–  –  –

Next, the spectra are converted from wavelength to wavenumber, where wavenumber, k, is given by k = 2π /λ.

This conversion is non- linear, making it necessary to resample the wavenumber spectra using a spline interpolation to restore linearity.

The interference spectra, as a function of wavenumber, are then digitally corrected for chromatic dispersion effects of the system by the following



The k- space spectrum is Hilbert transformed to obtain the analytical signal, and the phase of the signal at each wavenumber is determined.


In an ideal case, the phase should be linear with increasing wavenumber, so the signal is fit to a first order (linear) function.


The difference between the first- order fit and the actual signal as a

–  –  –


This difference is then used to correct dispersion by multiplying it at each wavenumber of the actual signal.

After dispersion compensation, the corrected spectra are Fourier transformed to reveal the spatial cross- correlation terms between the sample reflectors and reference fields, which contain information on the OPL difference between the two.

The process is 35 executed for all channels in real time, and the data are viewed as six separate A- scans.

Combining these A- scans creates a B- scan, or a two- dimensional cross- sectional view of the sample.

A critical element in the sample arm imaging optics, is the silicon v- groove chip.

The v- groove chip allows for extremely accurate spacing between the delivery fibers in the imaging array, and also protects the cleaved and polished ends of the fibers.

As shown in Figure 3.2, the chip is 3.8 mm wide, with eight triangular grooves milled into it, spaced 250 µ;m, center to center.

–  –  –

During the fiber mounting process, each fiber is stripped of its outer jacketing, cleaved, and laid into a single groove of the chip using a custom micrometer positioning stage under a stereo microscope.

UV curing epoxy (Norland Optical Adhesive 81) is then applied to the fiber to secure it in place.

Once all six fibers are aligned in position, a Pyrex lid is epoxied in place to protect the fiber ends, creating a linear six- fiber array.

The micrometer stage setup allows 3- way position adjustment (x, y, z), as well as tilt and 36 rotation.

This results in accurate and repeatable manipulation of the fibers in the chip.

The fiber faces are then hand polished with XW film polish (Krell Technologies).

It is also necessary to include path length adjustments in the reference arm that can be independently adjusted for each channel, to compensate for relative OPL differences between the channels, and still move the entire group at once.

Small differences in OPL from channel to channel (less than 1- 2 mm) arise during the construction of the system due to differences in fiber length within the 50/50 couplers, fusing splicing, and FC/APC coupling of the 6 parallel interferometers.

Therefore, a collimator mount, shown in Figure 3.3, was custom machined to allow manual compensation for the difference in OPL between channels.

–  –  –

Each fiber collimator is slid into place in the mount, and secured in position with a set screw.

This allows adjustment of each reference OPL to match the length of its corresponding sample arm.

Once each channel is aligned and secured, the entire group can then be moved in unison using a translation stage.

Figure 3.4, below, shows an A-  scan for each channel, displayed side- by- side as a B- scan, for a mirror sample.


–  –  –

The OPL’s were matched to within approximately 150 µ;m of each other by manual alignment.

With digital correction, the channels were matched within about 5 µ;m.

The digital correction changes the absolute path length of the modified channels, but relative distances between objects are preserved.

Figure 3.5 shows a digitally aligned B- scan of a coverglass.

–  –  –

38 The performance of the benchtop mLCI device was then evaluated via calculations of key characteristics, including optical signal- to- noise ratio, cross- talk, resolution, and falloff.

Also, initial gel thickness measurements were made using a calibration socket as well as a gel- tissue phantom in order to verify the ability of the mLCI instrument to measure gel distributions.

Optical signal- to- noise ratio (OSNR) was characterized for each channel of the mLCI system.

OSNR is essentially the ratio of the peak signal to the standard deviation of the shot noise floor, and is given


–  –  –

where ES is the maximum intensity of the Fourier transformed signal, EN is the standard deviation of the shot noise, and ND is the attenuation value of a neutral density filter.

In order to utilize Eqn.

(3.3), a spectrum from each channel is recorded with no sample in place.

The standard deviation of the noise floor is then found by using a mirror as the sample to create a single point reflector, and using a 0.5 ND filter placed in front of the sample in order to prevent saturation of the detector.

The maximum intensity value of the peak resulting from the mirror in the A- scan is then recorded, and the OSNR is calculated with Eqn.


The results of OSNR from Eqn.

(3.3) are shown in Table 3.1.

The OSNR values vary from channel to channel which can be attributed to the differences in losses at each fiber junction, differences in 50/50 coupler losses, and non-  39 uniformity of the 1x8 coupler.

However, even with the lowest OSNR of 87 dB (channel 5), the signal is clearly separated from the noise floor.

–  –  –

Lateral and axial resolution are completely decoupled in LCI, as is the case in OCT.

Theoretically, the axial resolution of a LCI system is governed by its coherence length as given by Eqn.

(3.1), or in this setup 6.2 µ;m.

In order to find the actual achieved resolution, a mirror is used as the sample, and a dispersion corrected A- scan is captured for each channel.

The full width at half maximum (FWHM) of the signal peak is acquired form the mirror sample is measured for each channel, and the results are shown in Table 3.1.

Measured resolution actually exceeds the expected value based on the coherence length for channels 2, 4, and 6.

This may be due in part to uncertainty in the determination of each peak FWHM, as the process interpolates sparse data points in order to find the resolution value at exactly 50% of the maximum signal.

Falloff, or roll- off, is another characteristic of Fourier- domain LCI that determines imaging depth.72, 73 Falloff, or depth dependent signal attenuation, occurs because as depth increases, fringe frequency increases as incident on the detector.

Since 40 the CCD array has a fixed element size, fewer points are sampled on each oscillation, leading to degradation of the signal as it is reconstructed digitally.

The depth at which the signal falls to 50% of its maximum SNR can be used to characterize falloff, and this depth is theoretically given by,

–  –  –

where Z- 3dB is the depth of which SNR is reduced by half, and δrk is the spectral resolution of the spectrometer.72, 73 The depth at which the signal falls to 50% of its maximum SNR can be used to characterize falloff.72, 73 To test the falloff of the LCI system, a mirror is placed in the sample arm and the reference arm is translated until the signal peak due to the mirror falls to 50% of its maximum value, as seen at zero OPL difference.

This distance value is then recorded and the results are also presented in Table 3.1.

The actual performance of the system is somewhat below the theoretical value from Eqn.

(3.4), where ෯r is the spectral resolution of the spectrometer.74 Eqn.

(3.4) predicts the

- 6 dB signal loss at 849.2 µ;m in depth, but Table 3.1 reveals that all channels fall short of this depth.

However, this is most likely due to the fact that the spectrometer was not completely optimized for the LCI application, as it was a commercially purchased instrument.75 Spectrometer resolution and wavelength of the light source both govern the achievable imaging depth in Fourier- domain LCI.

The maximum imaging range, Zmax,

–  –  –

where λ02 is the center wavelength of the light source, and (Δλ/N) is the detector wavelength resolution.

Each of the six channels’ spectrometers is identically configured with a resolution of 0.05 nm/pixel, and the source had a central wavelength of 842.9 nm, yielding a maximum axial range of 1.59 mm from Eqn.


This depth is sufficient to measure typical gel thickness in vivo, as distributions typically range around 100–500 µ;m.11, 16 Finally, cross- talk between adjacent channels is characterized for the mLCI instrument by disconnecting the fiber connectors from the light source for all but one channel, and measuring the power scattered by a tissue phantom onto adjacent channels.49 The cross- talk between signal channel, A, and adjacent channel, B, can be described


–  –  –

The cross- talk between channels was seen to be more than

- 26 dB for all pairs of channels in the benchtop mLCI device.71 In order to properly calibrate the pathlength across channels within the system, a mirror is used as a sample and each channel is offset 500 µ;m from its zero OPL difference by adjusting the reference arm with a micrometer driven translation stage.

The x- axis of each channel’s A- scan is then properly scaled so that the mirror peak 42 appeared at 500 µ;m depth.

Linearity of the system is then verified by taking measurements in each well of a custom calibration socket, where the wells vary in depth from 132 µ;m to 484 µ;m, as shown below in Figure 3.6.

–  –  –

Figure 3.6:

Calibration socket used to verify the linearity of the mLCI system.

Part (a) shows a basic sketch of the calibration socket.

Part (b) shows a diagram of the measurement procedure and geometry.

Part (c) shows the mLCI thickness measurements versus those of digital calipers.

Figure originally shown in Drake et al.71 The grooves of the socket are first imaged when filled with air and then when filled with a placebo gel (KY Jelly;

Johnson & Johnson).

For each imaging experiment, a transparent polycarbonate tube is placed on top of the sample as shown in Figure 3.6(b).

Two measurements per groove at 0.5 cm intervals are taken for the six channels, and the average OPL is compared to results from a digital indicator, shown in Figure 3.6(c).

The y- axis error bars are given by the axial resolution values presented in Figure 3.6(c), and the x- axis error bars are fixed at ± 5 µ;m given by the uncertainty of the measurement by digital calipers.

The slope of best fit for the gel filled socket measurements is 1.0028 with a correlation coefficient of 0.9902.

These measurements are in good agreement with 43 the refractive index, n, of the placebo gel.

Pages:     | 1 |   ...   | 3 | 4 || 6 | 7 |   ...   | 13 |

Similar works:

«THE LOGIC OF TEMPTATION THE LOGIC OF TEMPTATION PAUL M. H U G H E S * Introduction Temptation is normally thought to be morally dubious, in part because it involves desiring what we think is immoral, imprudent, illegal, unaesthetic, or in some other way wrong or bad. Temptation is also believed to be morally questionable in part because tempting is often a form of manipulation, which we regard as a violation of a person's integrity. ~As the O.~fordEnglish Dictionary makes clear, temptation has...»

«Accents, Punctuation or Cantillation Marks? A Study of the Linguistic Basis of the ṭəʿ Matthew Phillip Monger Masteroppgave i SEM4090 Semittisk Språkvitenskap 60 studiepoeng Program: Asiatiske og afrikanske studier Studieretning: Semittisk språkvitenskap med hebraisk Instituttet for kulturstudier og orientalske språk UNIVERSITETET I OSLO 1. juni 2012 II Accents, Punctuation or Cantillation Marks? A Study of the Linguistic Basis of the ṭəʿ Matthew Phillip Monger ‫( יִ ְר ַאת...»

«Vereniging voor de Verenigde Naties Model United Nations Flanders, 4th Edition SIMULATION EXERCISE November 2008 CASE 2008-2009 Negotiations in the UN Security Council on the EYES EYES aftermath of the “Georgian-Russian War”© ONL Y ONL Y Case developed by Dr. David Criekemans 1, Louis-Alfons Nobels 2 & Karen Van Laethem 3 Problem and setting Between August 7th and August 12th 2008, hostilities broke out between the Republic of Georgia and the Russian Federation. On August 26th, Russia...»

«Fun Factory Solaris Melee Challenge League (based on Catalyst convention SMC version 2.67) June 2015 Original rules by Ben Rome (2005) Solaris VII. The Game World. Where novice and veteran Mechwarriors come to try their hand at fame and fortune – or go down in flames, forgotten and penniless. This tournament attempts to immerse participants in the glitz, glamour, danger, and risk of arena combat and the ever-fickle attention of the audience. The brutality of 'Mech combat takes second seat to...»

«Instructions for using poisoned hen eggs for control of stoats (Mustela erminea) E B Spurr and S J Hough Manaaki Whenua Landcare Research PO Box 69 Lincoln New Zealand Published by Department of Conservation Head Office, PO Box 10-420 Wellington, New Zealand This report was commissioned by Science & Research Division ISSN 1171-9834 © 1997 Department of Conservation, P.O. Box 10-420, Wellington, New Zealand Reference to material in this report should be cited thus: Spurr, E.B. and Hough, S.J.,...»

«AS-11 MXF Program Contribution AMWA Specification AMWA Application Specification AS-11 MXF Program Contribution October 8, 2013 (version 1.1) Executive Summary AS-11 specifies a vendor-neutral subset of the MXF file format to use for delivery of finished programming from program producers and program distributors to broadcast stations. AS-11 includes the functionality of AS-03 and extends it to include high definition AVC-Intra and includes explicit support for standard definition D-10 video...»

«Operating mode of high pressure straws with high spatial resolution K.I. Davkov, V.V. Myalkovskiy, V.D. Peshekhonov, V.D. Cholakov The article presents results of studying the operating mode of thin-walled drift tubes (straws) at flushing it with a high-pressure gas mixture, which allowed obtaining extremely high spatial resolution for straw detectors. The results of studying the radiation ageing of straws operating in this mode are also described.1. Introduction Thin-walled drift tubes (TDT or...»

«Diss. ETH No. 17790 Astrogeodetic Geoid and Isostatic Considerations in the North Aegean Sea, Greece A dissertation submitted to the ETH ZURICH for the degree of Doctor of Sciences presented by ANNA EVE SOMIESKI Dipl. Ing., TU Dresden born April 08, 1977 citizen of Germany accepted on the recommendation of Prof. Dr. H.-G. Kahle, examiner Prof. Dr. I.N. Tziavos, co-examiner Dr. B. Bürki, co-examiner Abstract The present work is concerned with the enhancement, evaluation and deployment of the...»

«T H E 7 0 0 T H A N N I V E R S A R Y C A M PA I G N “The past is not dead, it is living in us, and will be alive in the future which we are now helping to make.” WILLIAM MORRIS | 1852, Theology Letter from The Rector FRANCES CAIRNCROSS IT IS WITH GREAT EXCITEMENT THAT WE EMBARK ON THIS CAMPAIGN – Exeter Excelling – which will guarantee tutorial excellence, transform the collegiate campus at the centre of Oxford, and secure financial resources for students and Fellows of the College....»

«Moorland Village Plan The Moorland Community Association In conjunction with Adopted by Council 18 August 2010 1 TABLE OF CONTENTS 1. Introduction 1.1. Purpose of this Plan 2. Background to the Village Plan 2.1. Consultation and Engagement 3. Characteristics of Moorland 3.1. Context 3.2. History and Heritage Significance of Moorland 3.3. Infrastructure, Facilities and Services 3.4. Environmental Features 3.5. Social Considerations 3.6. Planning Considerations 3.7. Tourism and the Local Economy...»

«SECTION: COMMUNICATION AND PUBLIC RELATIONS LDMD I COMMUNICATIVE MESSAGE AS NUCLEAR THINKING OF AN ASPIRATIONAL DESIRE Ştefan VLĂDUŢESCU, Associate Professor, PhD, University of Craiova Abstract: The paper is positioned in the Fundamentals of Communication Science zone. The approach is message-meanings centered and assumes that, at meta-analytic level, the message is the nucleus of communication. The first axiomatic finding is that, on semantic dimension, the message is an edifice of...»

«HDMI Demystified HDMI 1.3 ● Eye Pattern ● Cliff Effect ● Cable ● Speed Rating Xiaozheng Lu, Senior Vice President, Product Development, AudioQuest The release of the new HDMI 1.3 specification in June 2006 created both excitement and confusion in the consumer electronics industry. The discussion below is provided to help clarify this new technology and provide you with a better understanding of what you need to know when buying or selling HDMI products. What is HDMI? High-Definition...»

<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.