Numerical Simulated Concept and Mechanical Proof of Concept

for a Transmission OCT (tOCT)

Andreas Drauschke

, Katharina Dettelbacher

, Michaela Kienberger

, Sarah Nagl

and Christian Milz

Department Life Science Engineering, UAS Technikum Wien, H

ochst

adtplatz 6, Vienna, Austria

Faculty of Physics, TU Wien, Wiedner Hauptstraße 8–10, Vienna, Austria

Keywords:

Medical Imaging, Biomedical Optics, OCT.

Abstract:

Optical coherence tomography is a very powerful tool in imaging methods, but its practical use is limited due

to the shallow scan depth. A concept draft of a transmission OCT is presented, which eliminates the greatest

limitation of the OCT – the small penetration depth. The functional principle, based on the basic structure of

a Mach–Zehnder interferometer, is developed in a numerical simulation in OpticStudio. The numerical setup

includes beam expander and beam compressor systems to maximize the contrast generated in the interference

pattern and the design of an A–scan to realize a depth scan. As the result, the numerical simulation of the

complete setup is presented and the concept is then demonstrated in a simpliﬁed mechanical setup without an

A–scan.

1 INTRODUCTION

Imaging processes play a central role in medical diag-

nostics and in assisted surgical interventions (Bush-

berg et al., 2012; Webster et al., 2000; Drexler and

Fujimoto, 2015). Among other things, this is due to

the fact that modern medical imaging processes have

made enormous progress in recent years and not only

have the quality of existing diagnostic processes and

assistance systems been optimized, but a large num-

ber of new areas of applications has been established

as well (Samei and Peck, 2019). Some of the most

frequently used imaging procedures are listed in table

1. The methods are compared with one another with

regard to three important characteristics: the depth of

penetration in tissue, the achievable lateral and depth

resolution and the distinction between invasive and

non-invasive techniques. In addition, some typical ap-

plications of the respective processes are listed.

Optical coherence tomography is mostly used as

a non–invasive optical imaging technique with which

near–surface tissue structures can be imaged in three

dimensions or as two–dimensional cross–sections in

high resolution (Fercher et al., 2003a; Brezinski,

2006; Drexler and Fujimoto, 2015; Fujimoto and

Swanson, 2016). OCT is most commonly used in

ophthalmology, for example to make diagnostic scans

of the retina.

The OCT is based on the realization of a Michel-

son interferometer. The backscattered light is used in

the sample arm. The penetration depth is limited to

a maximum of 3 mm (Drexler and Fujimoto, 2015),

due to the physical principle of using backscattered

light, which considerably limits the use of the OCT.

An interesting aspect of OCT is that the lateral and ax-

ial resolution can be inﬂuenced independently of one

another. The generated images usually have a high

resolution and are only exceeded in terms of lateral

resolution by microscopy techniques, which lack ax-

ial penetration depth.

In order to overcome the essential limitation of

OCT, the evaluation of the forward–scattered light ap-

pears to be effective. Since no light is scattered an-

tiparallel to the illumination in such a setup in the

sample arm, the basic structure of a Michelson in-

terferometer can no longer be used. It makes sense

to use a Mach–Zehnder interferometer (Demtr

oder,

2009) as an alternative interferometric setup for the

forward–scattered light, in which the interference pat-

tern is not generated by interference with reﬂected or

backscattered light, but in the transmitted or forward–

scattered light. Since this corresponds to an OCT

setup in transmission, it is referred to as a transmis-

sion OCT (tOCT).

The assignment of an interference pattern to a

speciﬁc point in the analyzed tissue is more com-

Drauschke, A., Dettelbacher, K., Kienberger, M., Nagl, S. and Milz, C.

Numerical Simulated Concept and Mechanical Proof of Concept for a Transmission OCT (tOCT).

DOI: 10.5220/0010198500230031

In Proceedings of the 9th International Conference on Photonics, Optics and Laser Technology (PHOTOPTICS 2021), pages 23-31

ISBN: 978-989-758-492-3

Table 1: Comparison of different imaging techniques in terms of resolution, scan depth and invasive or non–invasive interven-

tion and applications. Computed tomography (CT), positron emission tomography (PET), single-photon emission computed

tomography (SPECT), magnetic resonance imaging (MRI), ultrasound (US), optical coherence tomography (OCT) and vari-

ous microscopy techniques (Bushberg et al., 2012; Webster et al., 2000; Vo-Dinh, 2003; Drexler and Fujimoto, 2015; Hoppert,

2003; Fujimoto and Swanson, 2016; Guggenheim et al., 2017; Rathod et al., 2015; Sch

utzenberger et al., 2019).

imaging technique invasive/non–

invasive

resolution

[axial/lateral]

scanning

depth of

penetration

examples of

application

CT invasive/ionizing

radiation

0.5 mm/0.6 mm whole body Imaging of the brain,

heart, lung, abdomen,

tumors, complicated

bone fractures,

angiography

PET invasive/ionizing

radiation

5 mm/5 mm whole body Oncology,

neuroimaging,

cardiology

SPECT invasive/ionizing

radiation

10 mm/10 mm whole body Myocardial

scintigraphy, skeletal

scintigraphy, brain

perfusion scintigraphy,

oncology

MRI non–invasive 0.9 mm/0.9 mm whole body Neuroimaging,

cardiovascular

imaging,

musculoskeletal

imaging, angiography,

liver and

gastrointestinal

imaging

MRI invasive/contrast

agent

0.9 mm/0.9 mm whole body Detection of

inﬂammations or

tumors or the

higher-contrast

imaging of vessels

US non–invasive 10 µm/300 µm 20 cm Imaging of muscles,

tendons, and most

internal organs

OCT non–invasive 1–15 µm/5 µm 2–3 mm Ophthalmology,

dermatology (early

cancer diagnosis)

OCT invasive/in

combination with

endoscopic

methods

1–15 µm/5 µm 2–3 mm Imaging of blood

vessels, heart,

intestines, bladder,

cancer diagnosis (early

stage), follow-up

treatment and

diagnosis of a heart

attack

various microscopic

techniques

non–invasive –/50 nm

(superresolution)

200 nm (classical)

0 mm Imaging on surfaces or

transparent tissue,

ophthalmology,

imaging in cell cultures

various microscopic

techniques

invasive/in

combination with

endoscopic

methods

–/50 nm

(superresolution)

200 nm (classical)

0 mm Imaging on surfaces of

blood vessels, heart,

bladder

PHOTOPTICS 2021 - 9th International Conference on Photonics, Optics and Laser Technology

plicated in forward–scattered light, since unscattered

light would be directly superimposed on scattered

light. Therefore, the scattered light offset with regard

to the illumination path will have to be analyzed.

In the presented work the design of a tOCT is real-

ized as a numerical concept and its fundamental capa-

bility for generating an interference pattern is demon-

strated. The basic suitability for the realization of an

A–scan is demonstrated in the numerical simulation.

The suitability of a Mach–Zehnder interferometer

is then shown in a simpliﬁed setup that does not in-

clude an A–scan. For this purpose, an interference

pattern is generated as a result of the superposition of

the reference beam with forward-scattered light in the

sample arm.

The proofs of concepts, both the numerical simu-

lation and the practical implementation in the demon-

strator, prove the basic suitability of the setup for the

implementation of a transmission OCT.

2 MATERIALS & METHODS

2.1 Physical Principles

The conceptual transmission OCT is intended to an-

alyze the depth information of a tissue by means of

the evaluation of forward–scattered light in contrast to

classic OCT, in which backscattered light is analyzed.

The greater scan depth of the tOCT is explained by the

larger forward scatter in relation to the back scatter of

the light. Due to the complexity of biological tissue,

an analytical description of the overall physical sit-

uation is not possible. The theory of Mie scattering

(Mie, 1908; Born and Wolf, 2005; Bhandari et al.,

2011) can be used to estimate the ratio of the forward

to backscattering. For this it is assumed that the light

is scattered by approximately spherical particles, as

shown in ﬁgure 1.

This is a very rough approximation because, on

the one hand, the particles in biological tissue are usu-

ally not spherical and, on the other hand, the distance

between the scattering particles is small enough that

there is a direct interaction between the scattering par-

Direction of Propagation

Figure 1: Basic geometry for describing the scattering of

light on spherical particles (Mie scattering) according to

Born & Wolf (Born and Wolf, 2005).

-90°

-30°

-20°

-10°

0°

10°

20°

30°

90°

180°

Figure 2: Scattering by a spherical water droplet with drop

radius of a = 260 nm (red line), a = 2600 nm (blue line)

with n = 1.33 + i10

−8

illuminated by visible light of wave-

length λ = 550 nm according to ﬁgure 4.9 from Bohren &

Huffman (Bohren and Huffman, 1983). Linear scale; solid

line: perpendicular polarization direction, dashed line: par-

allel polarization direction.

ticles. However, the estimate shows at least the rough

tendency of the expected scatter distribution.

An analysis of the scattered light distribution takes

place in the graphic evaluation of the scatter ampli-

tude as a function of the scatter angle. The asso-

ciated graphic representation is referred to as a po-

lar diagram as shown in ﬁgure 2. The result of the

Mie theory is that the forward scattering tends to in-

crease with increasing particle size, while at the same

time the scattering amplitude of the backward scatter-

ing tends to decrease (Bohren and Huffman, 1983).

In backscattering, a local maximum of the amplitude

can be found at a scattering angle of 180 deg under

certain physical conditions. With sufﬁciently large

particles (far from the Rayleigh domain) one always

ﬁnds a local maximum in forward scattering, the am-

plitude of which is signiﬁcantly greater than that of

backward scattering. For this reason, light in forward

scattering can generally achieve a higher scan depth

than in backward scattering with the same absorption.

Since these effects are wavelength–dependent, it will

be necessary to determine the optimal wavelength for

use in the tOCT.

It is assumed here that light, which is scattered

only once, emerges from the sample at a certain an-

gle. As a further approximation, the tissue is initially

assumed to be a plane–parallel plate as shown in ﬁg-

ure 3, so that the geometric relationships for analyzing

the scattered light are further simpliﬁed.

Numerical Simulated Concept and Mechanical Proof of Concept for a Transmission OCT (tOCT)

pinhole (black)

Sample (gray area)

Figure 3: Sketch for determining the correlation between

penetration depth L

and transmission angle ϕ

Both approximations of the plane–parallel arrange-

ment of the boundary surfaces most likely lead to

inaccuracies in the signal at large exit angles, while

the approximation of the single scattering most likely

leads to inaccuracies in the interference pattern for

small scattering angles. In practice, both errors will

become noticeable as a decreasing contrast of the in-

terference pattern.

By applying the two approximations, a direct cor-

relation between the exit angle of a beam – offsetted

to the illuminating beam – and the depth–localization

of the scattering location in the tissue can be derived.

An A–scan is therefore converted into an exit angle

spectrum. The physical conditions are shown in ﬁg-

ure 3. Simple geometrical considerations and the ap-

plication of the law of refraction at the point of beam

exit lead to

) = sin

−

(D − L

)

+ x

. (1)

An amplitude scan (A–scan) describes the basic form

of a depth–scan (Fercher et al., 2003b). This term

most commonly appears in the ﬁeld of ultrasonogra-

phy but, can be applied to other imaging modalities

such as OCT and tOCT in the same manner. For the

A–scan the magnitude of the signals, which in reﬂec-

tive setups results from echoes returning from a differ-

ent depth within the tissue, are displayed as a function

Light Source

Beamsplitter

Flat Mirror in

Reference Arm

Flat Mirror in

Sample Arm

Detector

Sample

Pinhole

Figure 4: General setup of the Mach–Zehnder interferom-

eter without A–scan for the realization of a transmission

OCT according to Demtr

oder (Demtr

oder, 2009).

of distance (Dance et al., 2014).

In the case of tOCT the more general explanation

of the A-scan, as a scan along one axis, is needed.

However, the analysis of this A-scan differs from that

of reﬂective setups. While imaging tools such as ul-

trasonography detect the depth information along one

axis at a single detector element, the scan is trans-

formed to a detection along several detector elements

for tOCT. This is due to the exiting angle being a

function of penetration depth thus transferring depth-

information to an axis perpendicular to the incident

light.

The basis of the conceptual transmission OCT is

a Mach–Zehnder interferometer, as shown in ﬁgure 4.

The unmodiﬁed reference arm of the classical Mach–

Zehnder interferometer is adopted, while a pinhole is

added in the sample arm next to the test sample. With

the help of this pinhole, those rays scattered in the

test sample can be selected. Since the scattered rays

do not run parallel to the illuminating beam, it is to be

expected in this setup that the structure only leads to

an interference pattern on the receiver for small exit

angles. Therefore, for the proof of concept, the pin-

hole was only shifted by approximately 1 mm so that

the illuminating beam was just faded out, but the scat-

tered beam was still propagated almost parallel to the

illuminating beam. Due to the small angle, it is to be

expected that a slightly distorted interference pattern

will be generated.

Since only nearly parallel rays generate an in-

terference pattern after passing through the second

beam splitter cube, the generated angle spectrum must

be transformed into a bundle of parallel rays. The

simplest method of parallelizing rays emitted from a

point is to place a concave mirror or converging lens

at the distance of the focal length to the exit point. To

avoid aberrations, a parabolic mirror or an aspherical

lens must be used. We decided to use an aspherical

lens for the construction in the full setup as shown

in ﬁgure 5. The A–Scan is thereby converted into

a bundle of parallel light beams, with different dis-

tances between the light beams and the unscattered

beam representing different depths of scattering in the

tissue.

2.2 Numerical Design

The concept of the transmission OCT was examined

in a numerical simulation. For this purpose, in ad-

dition to a simulation of the scattering properties of

the samples, an aspherical lens was integrated into the

structure that implements the A–scan. Additionally, a

beam expander was set into the reference arm so that

all the beams generated of the A–scan overlap with

PHOTOPTICS 2021 - 9th International Conference on Photonics, Optics and Laser Technology

100 mm

d e

h i

Figure 5: Numerical simulation of the Mach–Zehnder interferometer including A–Scan using OpticStudio (ZEMAX LLC,

2020). Some parameters of the components are listed in table 2: (a) coherent light source, (b) & (k) beam splitter cube,

(c) & (m) redirection mirror, (d) & (e) lenses for simulation of a beam expander, (f) & (g) lenses for simulation of a beam

compressor, (h) test sample, (i) pinhole, (j) collimator lens, and (n) possible detector position.

beams in the reference arm and thus generate a corre-

sponding interference pattern. In order to determine

the location of the scattering in the sample more pre-

cisely and to achieve a higher intensity of the beams

in the sample arm, a beam compressor was integrated

into the sample arm.

The scatter in the sample can in principle be simu-

lated in different ways. In the analyses, Lambert scat-

tering on the one hand and Gauss scattering on the

other hand were examined. It turned out that the sim-

ulation with a Gauss scattering produced more realis-

tic results, so that in the end a simulation with a Gauss

scattering with σ = 0.2 was carried out.

Two scattering angles were simulated: 3.52 deg

and 6.15 deg. A radius of 1.5 mm was assumed for

the reference sphere. Both scattering angles generate

an interference pattern on the detector. The interfer-

ence pattern were determined using a Jones calculus.

The two angles were then collimated with an as-

pherical lens. This special aspherical lens was de-

signed for this in order to minimize aberrations. In

practice, the lens will have to be adapted to the condi-

tions of manufacture. However, the minimization of

the aberrations will also be essential for a lens in a real

setup in order to keep the reduction of the contrast in

the interference pattern as low as possible.

Since only two relatively small angles were simu-

lated, no total internal reﬂection occurred at the in-

terface between the tissue and the surrounding air.

Total internal reﬂection is to be expected for larger

scattering angles and limits the scanning depth of the

method. The total internal reﬂection could be reduced

or even completely prevented by adding a transparent

contact gel with an adapted refractive index and a ﬁeld

lens at the level of the pinhole at the interface between

tissue and surrounding air.

2.3 Proof of Concept in Mechanical

Setup

In order to provide the proof of concept by means of a

generated interference pattern, the simpliﬁed mechan-

ical setup from ﬁgure 4 was implemented. In addition,

only a very small offset of the pinhole by 1 mm could

be implemented. If the offset was greater, the inten-

sity of the scattered beam dropped signiﬁcantly, so

that the interference pattern could no longer be gen-

erated. This effect was reinforced by the increasing

deviation from the parallel beam.

The implemented setup is shown in ﬁgure 6. A

helium–neon laser at a wavelength of 632.8 nm was

used for illumination. Non–polarizing beam splitter

plates with split ratios of 50:50 were used for beam

splitting. Due to the sharp drop in intensity in the

A B C D E

Figure 6: Mechanical setup for the proof of concept. (A)

helium–neon laser at a wavelength of 632.8 nm, (B) beam-

splitter, (C) sample, (D) pinhole, (E) ﬂat mirror, (F) ﬂat mir-

ror, (G) beamsplitter.

Numerical Simulated Concept and Mechanical Proof of Concept for a Transmission OCT (tOCT)

Table 2: Parameters of the components of the Mach–Zehnder interferometer design. The numbering corresponds to that

shown in ﬁgure 5.

Component Position

(y, z)

[mm]

Radius

[mm]

Conic

[mm]

Thickness

[mm]

Radius

[mm]

Conic

[mm]

Material

Beamsplitter (b) (0,20) ∞ – 10 ∞ – N–BK7

Flat mirror (c) (50,30) ∞ 0 0 – – Silver

Beamexpander lens (d) (50,40) 5 0 1 ∞ 0 N–BK7

Beamexpander lens (e) (50,190) 97.75 0 – ∞ 0 N–BK7

Beamcompressor lens (f) (0,44) 97.75 0 5 ∞ 0 N–BK7

Beamcompressor lens (g) (0,237) 1 0 – ∞ 0 N–BK7

Sample (h) (0,250) ∞ – 1 ∞ – scattering

surfaces

Pinhole (i) (0.9,260) – – 0 ∞ – N–BK7

Collimator lens (j) (0.9,300) −36.549 −7.721 15 −18.79 -0.607 N–BK7

Beamsplitter (k) (50,325) ∞ – 10 ∞ – N–BK7

ﬂat mirror (m) (10,340) ∞ 0 0 – – Silver

sample arm, a split ratio of 90:10 would have been

more suitable, with the 90% being localized in the

sample arm. The new split ratio will additionally im-

prove the contrast in the interference pattern.

A thin preparation of artiﬁcially grown cells was

used as sample. The used sample consisted of skeletal

muscle–like constructs. The tissue was grown from

myoblasts that were embedded in a ﬁbrin hydrogel

and mechanically stimulated in a bioreactor (Heher

et al., 2015). Due to the small thickness of the prepa-

ration, multiple scattering with signiﬁcantly different

optical paths could be practically excluded. With the

selected setup no A–Scan could be realized. However,

the structure made it easier to generate an interference

pattern.

3 RESULTS

3.1 Numerical Simulation

The numerical simulation was carried out in accor-

dance with the illustration from ﬁgure 5 in OpticStu-

dio (ZEMAX LLC, 2020). The most important op-

tical properties of the used components are listed in

table 2. N–BK7 was used for all transmissive optical

components. In practice, different materials may be

used.

Beam expander and beam compressor were de-

signed manually. In the ﬁnal setup, these will be re-

placed by commercial beam expander systems. In the

simulation the selection of the different scan depths

takes place by using two scatter points located at a dis-

tance of one millimeter from one another. In a practi-

cally implemented setup, the interference patterns of

the scattering points of different depths (A–Scan) are

possibly recorded sequentially by a movable pinhole

in the sample arm in front of the beam splitter cube (k)

in ﬁgure 5. This is not necessary in the simulation be-

cause the generated interference patterns are located

sufﬁciently distant from one another.

The corresponding simulated interference patterns

are shown in ﬁgure 7. Two sharply separated ring pat-

terns of the two scattering points, situated at different

depths, can clearly be seen.

The simulation basically demonstrates that the

theoretical setup is suitable for reconstructing an A–

Scan from the recorded interference patterns. It

should be noted that the interference patterns will

overlap in practice. A separation can be achieved via

a pinhole in the sample arm in front of the beam split-

ting cube (k). In order to accelerate the A–scan, a

line of pinhole diaphragms is recommended, which

generates separate interference patterns from differ-

Y coordinate value

3.72

2.58

-1.44

-2.56 0 2.58

X coordinate value

150.8

135.7

120.6

105.6

90.5

75.4

60.3

45.2

30.2

15.1

Coherent

Irradiance

Figure 7: Simulation of the generated interference patterns

for two different depths 1 mm apart in the sample.

PHOTOPTICS 2021 - 9th International Conference on Photonics, Optics and Laser Technology

ent sample depths, but which are relatively far apart.

By moving the pinhole line over the distance between

two holes, the entire scan is obtained in the time T /m,

where T is the duration for a scan with shifting one

pinhole and m is the number of used pinholes.

Due to the different optical paths, the two interfer-

ence patterns are different, as expected. The center of

the interference pattern is shifted, so that the center

of the ring–shaped interference pattern in the upper

image of the ﬁgure 7 lies in the center of the sensor,

while the center of the ring–shaped interference pat-

tern is shifted upwards for the second depth, shown

in the lower image of ﬁgure 7. Using a larger sensor

allows the center points of the interference patterns to

remain searchable, which is necessary for the inter-

pretation of the A–scan.

Basically, the simulation proves that the scattering

from both tissue depths leads to an angular spectrum

of the exiting rays which can be converted into paral-

lel rays by means of the lens (j). The distance between

the rays corresponds to the scan depth, starting with

the point of entry of the light ray into the tissue up to

a maximum scanning depth, which is limited on the

one hand by the total internal reﬂection occurring at

the exit surface and on the other hand by the NA of

the lens (j) used. Both effects could be minimized in

practice by using a contact gel between the tissue (h)

and the pinhole (i) and a ﬁeld lens that is positioned

in the pinhole.

3.2 Experimental Proof of Concept

In order to provide the proof of concept in a simpli-

ﬁed experimental version, all components for the re-

alization of an A–scan were removed from the setup.

The setup as shown in ﬁgures 4 and 6 was realized.

Therefore, a simpliﬁed design as shown in ﬁgure 4

was used. The implemented setup is shown in ﬁgure

In order to minimize the absorption of the light

beam in the sample, a thin preparation should be used.

More precisely, tissue with low absorption should be

chosen. Artiﬁcially grown skeletal muscle cells in a

thin–layer preparation were therefore used as a sam-

ple. These show a signiﬁcant forward scatter.

Since no lens was used in the setup to parallelize

the scattered rays, only small scattering angles could

be detected, which are scattered almost parallel to the

illumination light beam. At the same time, it had to

be guaranteed that the illuminating beam would be ab-

sorbed as completely as possible. Since the waist of

the illumination beam is approximately 0.5 mm, the

pinhole behind the sample was shifted by 1 mm com-

pared to the illumination beam. This guarantees that,

Figure 8: Measurement of the generated interference pat-

terns for nonscattered light (A) and light scattered at a mus-

cle cell sample (B).

on the one hand, only scattered rays can pass through

the aperture, but the illuminating beam is absorbed

and, on the other hand, the transmitted light has only

a small angle with respect to the illuminating beam.

This deviation in parallelism should be reﬂected in a

distorted interference pattern.

With the simpliﬁed setup, the interference pattern

shown in ﬁgure 8 was recorded. Figure 8 (A) shows

the undistorted interference pattern generated at the

output of a Mach–Zehnder interferometer. The setup

did not contain a test sample or a pinhole, so that the

interference pattern corresponds to that of an undis-

turbed Mach–Zehnder interferometer.

Figure 8 (B) shows the generated interference pat-

tern after adding the test sample and the offset pin-

hole. The distortion in the interference pattern can be

seen as a deformed envelope. However, the interfer-

ence fringes can also be clearly detected in the sim-

pliﬁed setup. The location of the scattering cannot be

exactly determined due to the small thickness of the

preparation and the lack of a lens for separating the

various rays of the A–scan.

Since this setup is a pure proof of concept, no fur-

ther measurement of the interference pattern was car-

ried out. The setup proved to be very sensitive to vi-

brations. In addition, the relatively low intensity of

the illumination laser used proved to be limiting.

4 DISCUSSION

The relatively shallow scan depth is the greatest limit-

ing factor in optical coherence tomography. OCT can

be used to create both laterally and in depth high reso-

lution, three–dimensional scans of tissue, without the

use of ionizing radiation. The achievable scan depth is

limited by the usable light source intensity, the wave-

length and the type of tissue to be scanned.

In order to reduce this limiting factor, but at the

same time provide the advantageous properties of the

OCT, in particular the independence from the scanned

Numerical Simulated Concept and Mechanical Proof of Concept for a Transmission OCT (tOCT)

tissue and the high, achievable resolution, a transmis-

sion version of the OCT was designed.

The Michelson interferometer usually used in

classical OCT, with which backscattered light is an-

alyzed interferometrically, is replaced by a Mach–

Zender interferometer in which forward scattered

light is to be analyzed.

In order to generate interference patterns with a

sufﬁciently high contrast, some approximations must

be fulﬁlled. On the one hand, the once scattered rays

of the sample beam must be dominant over the multi-

ple scattered rays. The setup in use also assumes that

the examined tissue can be approximated as a plane–

parallel plate.

The approximation that only single scattering can

be measured limits the achievable result, since for

multiple scattered light the formula 1 is no longer

valid. Multiple scattering would result in the wrong

depth of the scattering point, so that a clear assign-

ment of the scattering point is no longer possible. This

would manifest itself in a decrease in the contrast in

the generated interference image. With a sufﬁciently

large scattering angle, however, as can be seen in ﬁg-

ure 2, it can be assumed that the amplitude of the mul-

tiple scattered light drops sufﬁciently quickly so that

the reduction in the achievable contrast still allows the

result to be clearly interpreted.

In a numerical simulation it was shown that the

proposed setup is suitable for generating interference

from rays that are scattered at different depths of the

tissue. In addition, the proof of concept was provided

that the proposed design for carrying out an A–scan

works in principle. However, the tissue cannot be

scanned in full depth, since, above a certain depth,

scattered rays are totally reﬂected on the exit surface.

In addition, the NA of the A–Scan–lens limits the

measurable range of the representable scattered rays.

In a simpliﬁed experimental setup, the proof of

concept was achieved and demonstrated that interfer-

ence patterns can be generated with scattered rays

in transmission in a Mach–Zehnder interferometer.

Therefore this method is in principle able to function

as the core of a tOCT.

5 CONCLUSION & FURTHER

CHALLENGES

Possibilities for optimization were identiﬁed in both

the numerical and the experimental setup.

The scan depth is limited by the total reﬂection

occurring and by the NA of the A–scan lens. The for-

mer can be reduced by using a contact gel, similar to

ultrasound, and the latter appears to be reducible by

using a ﬁeld lens in the position of the pinhole. In

addition, a numerical analysis should be carried out

with more realistic dissemination models for tissue

obtained from laboratory experiments. The Gaussian

model used in the current simulation may not reﬂect

reality adequately.

The inﬂuence of different split ratios of the beam

splitter should be examined and optimized numeri-

cally and there is still no tolerance analysis of the

components used.

In the next step, a tOCT including a complete

A–scan will be implemented experimentally. In this

setup it is to be examined how the A–scan can be

implemented in practical application. The possible

choice of a second pinhole in front of the beam splitter

(k) from ﬁgure 5 and the scanning speed will depend

on this.

Furthermore, it has to be analyzed in experimen-

tal investigations whether the back–calculation is also

possible from the generated interference patterns for

the illuminated structure. This analysis will be based

on measurements of scattered light on tissue samples.

The maximum depth of penetration will depend

on the wavelength used, the intensity of the light used

and the tissue being screened. These examinations are

to be carried out in the future.

In addition, the usability of different light sources,

especially with regard to the desired wavelength

used, should be investigated. Multi–wavelength scans

could signiﬁcantly improve the resolution of the scans

generated.

REFERENCES

Bhandari, A., Hamre, B., Frette, O., Stamnes, K., and

Stamnes, J. J. (2011). Modeling optical properties of

human skin using mie theory for particles with dif-

ferent size distributions and refractive indices. Op-

tics Express, 19(15):14549–14567. https://doi.org/10.

1364/OE.19.014549.

Bohren, Craig F. and Huffman, Donald R. (1983). Absorp-

tion and Scattering of Light by Small Particles. John

Wiley & Sons, Inc, Clearance Center, 222 Rosewood

Drive, Danvers, MA 01923.

Born, M. and Wolf, E. (2005). Principles in Optics. Perg-

amon Press LTD., Headington Hill Hall, Oxford 4 &

5 Fitzroy Square, London W.1., Great Britain, 7

edi-

tion.

Brezinski, Mark E. (2006). Optical Coherence Tomogra-

phy – Principles and Applications. Elsevier Inc., 84

Theobald’s Road, London WC1X 8RR, UK.

Bushberg, Jerold T., Seibert, J. A., Leidholdt, Edwin M.,

and Boone, John M., editors (2012). The Essential

Physics of Medical Imaging. Lippincott Williams &

Wilkins, a Wolters Kluwer business, Two Commerce

PHOTOPTICS 2021 - 9th International Conference on Photonics, Optics and Laser Technology

Square, 2001 Market Street, Philadelphia, PA 19103

USA, 3

edition.

Dance, David R., Christoﬁdes, S., Maidment, A. D. A.,

Mclean, I. D., and Ng, K. H., editors (2014). Diagnos-

tic Radiology Physics: A Handbook for Teachers and

Students. International Atomic Energy Agency, Vi-

enna international Centre. Po box 100. 1400 Vienna,

Austria.

Demtr

oder, W. (2009). Experimentalphysik – Band 2: Elek-

trizit

at und Optik. Springer–Verlag Berlin Heidelberg,

edition.

Drexler, W. and Fujimoto, James G., editors (2015). Opti-

cal Coherence Tomography – Technology and Appli-

cations. Springer Cham Heidelberg New York Dor-

drecht London, 2

edition.

Fercher, A. F., Drexler, W., Hitzenberger, Christoph K., and

Lasser, T. (2003a). Optical coherence tomography –

principles and applications. Rep Prog Phys, 66:239–

303. https://doi.org/10.1088%2F0034-4885%2F66%

2F2%2F204.

Fercher, A. F., Drexler, W., Hitzenberger, Christoph K.,

and Lasser, T. (2003b). Optical coherence tomogra-

phy – principles and applications. Rep Prog Phys,

66(2):239–303. https://doi.org/10.1088/0034-4885/

66/2/204.

Fujimoto, J. and Swanson, E. (2016). The development,

commercialization, and impact of optical coherence

tomography. IOVS, Special Issue, 57(9). https://www.

readcube.com/articles/10.1167/iovs.16-19963.

Guggenheim, Emily J., Lynch, I., and Rappoport, Joshua

Z. (2017). Imaging in focus: Reﬂected light imag-

ing: Techniques and applications. Int J Biochem Cell

Biol, 83:65–70. https://doi.org/10.1016/j.biocel.2016.

12.008.

Heher, P., Maleiner, B., Pr

uller, J., Teuschl, A. H.,

Kollmitzer, J., Monforte, X., Wolbank, S., Redl, H.,

unzler, D., and Fuchs, C. (2015). A novel biore-

actor for the generation of highly aligned 3d skele-

tal muscle-like constructs through orientation of ﬁb-

rin via application of static strain. Acta Biomater,

24:251–26. https://doi.org/10.1016/j.actbio.2015.06.

033.

Hoppert, M. (2003). Microscopic Techniques in Biotech-

nology. WILEY–VCH Verlag GmbH & Co. KGaA,

Weinheim, Germany.

Mie, G. (1908). Beitr

age zur Optik tr

uber Medien,

speziell kolloidaler Metall

osungen. Annalen der

Physik, 4(25):377–445. https://doi.org/10.1002/andp.

19083300302.

Rathod, K. S., Hamshere, S. M., Jones, D. A., and Mathur,

A. (2015). Intravascular ultrasound versus optical co-

herence tomography for coronary artery imaging –

apples and oranges? Interv Cardiol, 10(1). https:

//doi.org/10.15420/icr.2015.10.1.8.

Samei, E. and Peck, Donald J., editors (2019). Hendee’s

Physics of Medical Imaging. JohnWiley & Sons, Inc.,

9600 Garsington Road, Oxford, OX4 2DQ, UK, 5

edition.

Sch

utzenberger, K., Pﬁster, M., Messner, A., Fr

ohlich,

V., Garh

ofer, G., Hohenadl, C., Schmetterer, L.,

and Werkmeister, R. M. (2019). Comparison of

optical coherence tomography and high frequency

ultrasound imaging in mice for the assessment of

skin morphology and intradermal volumes. Sci-

entiﬁc Reports, 9(13643). https://doi.org/10.1038/

s41598-019-50104-4.

Vo-Dinh, T., editor (2003). Biomedical Photonics Hand-

book. CRC Press LLC, 2000 N.W. Corporate Blvd.,

Boca Raton, Florida 33431, USA, 1

edition.

Webster, John G., Ritenour, E., Tabakov, S., and Ng, K.,

editors (2000). Webb’s Physics of Medical Imaging.

CRC Press Taylor & Francis Group, 6000 Broken

Sound Parkway NW, Suite 300, USA, 2

edition.

ZEMAX LLC (2020). Opticstudio

. https://www.zemax.

com/products/opticstudio.

Numerical Simulated Concept and Mechanical Proof of Concept for a Transmission OCT (tOCT)