Science topic

Optical Coherence Tomography - Science topic

Explore the latest questions and answers in Optical Coherence Tomography, and find Optical Coherence Tomography experts.
Questions related to Optical Coherence Tomography
  • asked a question related to Optical Coherence Tomography
Question
10 answers
1. The necessity of a polarization controller for single-mode fiber. Is a polarization controller necessary for single-mode fibers? What happens when you don't have a polarization controller?
2. Optical path matching problem. How to ensure that the two arms of the optical path difference match, any tips in the adjustment process? If the optical path difference exceeds the imaging distance, will interference fringes fail to appear?
Only these questions for the time being, if there are more welcome to point out.
Relevant answer
Answer
1. The polarization can change during propagation. This will degrade the visibility of the fringes. You need a polarization controller or a polarization-maintaining fiber.
2. By the imaging distance, do you mean the distance to the sample or the sample's dimension? In any case, if the OPD exceeds the coherence lenght of your source, the interference fringes disappear. In order to match the OPD, we typically sweep the reference arm a long distance and record the output intensity. Another method is to monitor the output spectrum using a spectrometer. The spectrum shows oscillations for non-zero OPD. When the OPD approaches zero, the spectrum oscillations tend to disappear.
  • asked a question related to Optical Coherence Tomography
Question
5 answers
Histologically, rat retinal blood vessels were reported to be running in the nerve fiber layer. Looking at OCT scans, I believe they dig deeper or at least making groove like path in inner plexiform layer (IPL, "blue line" of different lengths under the blood vessel "BV' and under the nerve fiber layer "NFL"). This is distorting the total retinal layer thickness measurements. I could find a comment in the literature. Any feedback is greatly appreciated?
Relevant answer
Answer
Thanks for sharing this scan. This is located in RNFL and GCL but IPL is posterior to it and differentiation of IPL became difficult because of shadowing. Perhaps this is a Retinal Vein that lie deeper than arteries
  • asked a question related to Optical Coherence Tomography
Question
13 answers
I want to analyze individual retinal layer thickness from Optical coherence tomography scans. Can it be done using ImageJ??
Relevant answer
Answer
How do you intend to calibrate the scale?
  • asked a question related to Optical Coherence Tomography
Question
2 answers
Looking for public database/source of en face OCT angiography (OCTA). Anybody know where can I find OCTA en face images? I have several b-scan sources but cannot find en face OCTA images. Thanks
Relevant answer
Answer
Hi, I hope this answer is not too late. You mat find datasets at:
and here:
  • asked a question related to Optical Coherence Tomography
Question
2 answers
Dry-wet mehod was used to obtain the porosity information of one UF/MF membrane or some other porous media, but it is not accurate in test the porosity of one biofilm or one cake layer one the membrane surface? Recently, Optical Coherence tomography (OCT) was used to obtain the porosity of the longterm biofilm, but it still work not very well in the initial process. I want find some accurate methods to obtain the porosity of one biofilm or cake layer. Any suggestion is welcomed. Thank you!
Relevant answer
Answer
Hi Yu
I think these papers can help you:
Hu M, Zhang TC, Stansbury J, Neal J, Garboczi EJ. Determination of porosity and thickness of biofilm attached on irregular-shaped media. Journal of Environmental Engineering. 2013 Jan 31;139(7):923-31.
Zhang TC, Bishop PL. Density, porosity, and pore structure of biofilms. Water Research. 1994 Nov 1;28(11):2267-77.
Garboczi EJ, Hu M, Zhang T. DETERMINATION OF MEAN DENSITIES, POROSITY AND THICKNESS OF BIOFILMS ATTACHED ON IRREGULAR-SHAPED MEDIA. 2011 Aug 10.
  • asked a question related to Optical Coherence Tomography
Question
1 answer
Does anybody know if there is an algorithm to extrapolate angio OCT sequences from a SDOCT?
Relevant answer
Answer
  • It is possible but the resultant image is small and has poor quality. As the scan speed is low in usual SD OCT(some 40 k ) while is some (80 k) in OCTA. But heidelberg provide users with hardware and sofware upgrade.
  • asked a question related to Optical Coherence Tomography
Question
11 answers
How to get good grades in the examination of RNFL peripapillary in OCT, because in most cases the note is yellow and it decreases the confidence in the result.
Relevant answer
Answer
Thank you for all remarks about my question!
Best Regards
Romar
  • asked a question related to Optical Coherence Tomography
Question
7 answers
In FD-OCT you analyse the interference signal of a broadband light source (going through a Michelson interferometer) with a spectrometer. You get a specific intensity for every chip of your detector. These chips are all in one row and each chip represents a specific wavelength or wavenumber. If you use the chip-row as your x-coordinate and add the measured amplitude of every chip to your diagramme, you get a function that oscillates with a specific frequency. By doing a FFT, I could read out the frequency from the FFT-plot. Everything worked well. But I'd like to understand what I am doing :)
Now my question: In the literature, people say that we use the FFT to get from the Frequency Domain to the Time Domain what confuses me because I was taught that I need a IFFT for this.
Could it be that FT-OCT is just called "Frequency Domain" because the x-axis of the signal is a wavenumber (=frequency of light)?
For my understanding, the signal from the detector is the classical signal you usually have in the time domain. However, I would like to call it "wavenumber domain" because the x-axis is given in 1/cm. When I apply FFT I get an x-axis which is in cm (or µm). I think it would make sense to call that "spatial domain". This µm-value is the Optical Path Difference between the two arms of my interferometer.
In the pictures below my post you can see the amplitude every pixel measures (spectrogram) and the FFT showing a Optical Path Difference of about 300µm.
Thank you very much. Best regards from Germany,
Niklas
Relevant answer
Answer
Donglin Wang's answer is correct - it's called "frequency domain" because the data is collected in k-space, or the "spatial frequency domain." If the data is not collected in the k-space (perhaps un-even k-space sampling by sampling wavelength instead), the data is sub-sampled or interpolated into k-space. Then, the FFT transforms the k-space data into image space.
  • asked a question related to Optical Coherence Tomography
Question
6 answers
In order to determine the effect of MS without ON on the RNFL thickness, how long does the impact of ON on the RNFL thickness last for to include patients with previous ON in the study?
Relevant answer
Answer
At least 3 months after acute bulbar or retrobulbar ON 
  • asked a question related to Optical Coherence Tomography
Question
1 answer
between spectral domain and optical coherence tomography
Relevant answer
Answer
It is very important opinion
  • asked a question related to Optical Coherence Tomography
Question
3 answers
I am currently performing a craniotomy/durotomy procedure on rats (Long Evans – Hooded Blue Spruce 450-600g) to expose brain tissue for optical coherence tomography. To open the skull, I am using an engraving bit (Dremel size 105 & 106) to cut a 4mm x 4mm area of the skull. Once dura mater is exposed I am able to lift the skull off cleanly without damaging the brain tissue. The dura mater is then carefully lifted with a tweezers and cut with a micro scissors. During roughly half of the procedures the underlying meningeal layers are removed cleanly along with the dura. However, if the pia does not come off cleanly (typically when it contains large perforating cortical vessels), any attempts to remove it results in significant bleeding. This obviously inhibits the imaging process by blocking the light aimed to penetrate tissue. Occasionally the skull bleeds significantly as well. Thus, if I am unable to avoid excessive bleeding, does anyone know of any non-pharmacological methods for addressing this problem (other than Gelfoam Sponge or Actcel hemostatic gauze)? Or any pharmacological methods that would not compromise hemodynamics? In addition, the exposed brain tissue becomes inflamed, varying the focal length of the lens and further distorting the images. What is the best way to control for this inflammation? I have been applying aCSF (stored in an ice bath) to control for this. I have also tried opening another area of the skull (left lateral the sagittal suture, posterior to bregma and anterior to lamda) to allow for a release of some pressure. So far, this doesn’t appear to be beneficial. Lastly, what is the ideal dosage of Isoflurane to administer a rat under hypoxia and/or hypercapnia? After collecting baseline data (at normoxia), the rats are exposed to a hypoxic/hypercapnic gas mixture (10% O2, 5% CO2, 85% N) in order to examine capillary dilation and perfusion of the cerebrovascular reserve. Some of the rats remain stable under this condition, others expire within a few short minutes. Does anyone have an explanation for the variability? The rats are all the same breed, age, and are close to the same weight.
My goal is to examine hemodynamic activity in real-time. Therefore, any solutions to these issues should not further compromise normal blood flow, or vascular restriction and dilation.
Relevant answer
Answer
Hi Jacob,
1. This is a protocol nearly for the same goal but applied for mice. Maybe this can help you.
2. In this article, you can find anesthesia protocol and surgical procedure in the Surgical Procedure section.
3. On the other hand, how did you remove the pia mater? What kind of tweezer do you use?
Kind regards
  • asked a question related to Optical Coherence Tomography
Question
5 answers
I have been reading about lateration behaviour in fish and it has been suggested that one or the other eye is used in detection of predators. This should be possible as nerves from the right retina terminate in the left optic tectum, and the left retina to the right optic tectum. There are studies with fish exposed to high CO2 that in a two choice situation turns to one or the other side (laterization)(e.g. Miklosi et al. 1997, Physiol. Behav. 63, 127-135) . What do think about these ideas? First of all do all nerves from for instance the right eye terminate in the left side of the brain? Best Wishes Håkan Olsén
Relevant answer
Answer
Hi Håkan
You could go to my ResearchGate page and download this paper, and maybe the one just above it.
Visuomotor perimetry in fish: A new approach to the functional analysis of altered visual pathways  by
D.P.M. Northmore · L.C. Skeen · J.M. Pindzola
There is research on lateralization in larval zebrafish that might be of interest to you:
Best regards  - David
  • asked a question related to Optical Coherence Tomography
Question
8 answers
Our recent study revealed AMD-related CNV recurrence during anti-VEGF therapy after 14 days of follow-up. It will be very interesting if we try to reduce interval between injections to two weeks. I didn't find evidence that it is unsafe.
Dear colleagues, I want to hear you opinion...
Relevant answer
Answer
There are a few patients in whom this is necessary (see Dr. Rosen's comment). No adverse events were observed. Sill, these are individual cases and it is hard to generalise.
  • asked a question related to Optical Coherence Tomography
Question
3 answers
On the one hand, Thomas et al reported a wide series of patients with foveal hypoplasia, and they found good correlation between the grade of foveal hypoplasia and the range of visual acuity that those subjects could achieve. On the other hand, some single reports have been published of patients with high grade of foveal hypoplasia and good visual acuity, calling into the question the importance of a foveal pit in achieving good visual acuity. 
What do you think about this matter?
Relevant answer
First of all the groups taken into account are different genetically : 
Sixty-nine patients with foveal hypoplasia (albinism, n = 34; PAX6 mutations, n = 10; isolated cases, n = 14; achromatopsia, n = 11) and 65 control subjects were examined.
This induces a first bias in the evaluation of different types of hypoplasia: in all the aforementioned pathologies the genetic and thus the biomolecular pathway targeted by the specific alteration is different and the evaluation by OCT doesn't allow a morfofunctional differentiation of the subgroups at all. If you consider achromatopsia for example which involves a complete or incomplete development of the cones you will observe a worse visual acuity due to the anatomic absence of the photoreceptors or a part of them.
  • asked a question related to Optical Coherence Tomography
Question
4 answers
i want to know more about technique of coherent beam combining.for example active and passive soret of combination fiber laser beam.
Relevant answer
Answer
Hi Ali,
There are 3.5 basic methods, but with numerous application schemes for each.
Coherent: promises the ultimate output beam, spatial quality and phase manipulation / steering and so... But; cost in complexity: phase ctrl / polarization management, ambient noise, actuators in loop with sensors.
Spectral incoherent: using dispersive element to spatially align beams from different spectrum. Can have high spatial qality, at the expense of lowering the spectral density. less sensitive, but needs costy dispersive element. Polarization might need good control.
Spatial incoherent: family of methods to overlap the individual beams, at the expense of spatial brightness compromise. Fiber fusion  / free space collimators are common. Low sensitivities to almost all parameters. Ultimate spatial quality is dictated by brightness law.
Polarization cimbiuning is also a sub-group, enables to overlap two orthogonal polarizations via one polarization selective element (e.g. PBS).
Surf the web for info. 1990s and further on are full with scientific / industrial progress...
  • asked a question related to Optical Coherence Tomography
Question
12 answers
Is anyone aware of any reliable and precise methods for estimates of choroidal thickness using SS-OCT given that both currently used manual (point measurements) as well as automatic algorithms(built in) have their own inherent errors in measurements?
Relevant answer
Answer
Generally the technology behind SS-OCT should yield a more accurate choroidal thickness measurement.
Both high definition SD-OCT (manual segmentation of a single scan) and SS-OCT (automated segmentation) are comparable with a minor negligible difference (the SD-OCT is being thicker by 10 - 15 microns) and interchangeable. The difference being more significant the thicker the choroid (and also decreasing accuracy of SD-OCT).
  • asked a question related to Optical Coherence Tomography
Question
3 answers
    I am doing some research about the extraction of the eccrine sweat gland using oct fingerprint,but the image in our lab is too bad to recognize the gland. I want to know where can I find some OCT fingertip images? 
Relevant answer
Answer
Optical coherence tomography should be well suited for that purpose. Although I saw some of papers dealing with fingertips (e.g. the vascular structure) I highly doubt, there is some kind of accessible database.
May I ask which kind of OCT device you are using? (wavelength, FOV and type)
Another question is: Which resolution do you need? (let's say pixel per gland)
Do you need B-scans or C-scans?
  • asked a question related to Optical Coherence Tomography
Question
4 answers
Optical Coherence Tomography data is available from spectrum analyzer that needs to be assembled in the form of image. 
Relevant answer
Answer
ImageJ can be used and is Java scriptable. But you have to be more specific. What sort of data do you have and what is is that you want to measure. What is the scientific question that you need answered?
-Peter
  • asked a question related to Optical Coherence Tomography
Question
1 answer
Is it possible to see a differentiation between conj and tenons on AS-OCT or do they have similar signal? I'm looking specifically at trabeculectomy bleb morphology. (Topcon 3000 AS OCT). Thanks 
Relevant answer
Answer
no  we canoot differentiate it from tenons layers buct we can differentiate it from sclera
  • asked a question related to Optical Coherence Tomography
Question
4 answers
What algorithm do OCT machines use to generate automatic measurements of thicknesses and volumes?
Relevant answer
Answer
While there is not any easy way to access the actual software that are used in OCT devices, There are a few publicly available OCT segmentation software. Here are the link to 2 of them:
  • asked a question related to Optical Coherence Tomography
Question
10 answers
Optical coherence tomography (OCT) is a new, noninvasive, noncontact, transpupillary imaging technology which can image retinal structures in vivo with a resolution of 10 to 17 microns. Cross-sectional images of the retina are produced using the optical backscattering of light in a fashion analogous to B- scan ultrasonography. The anatomic layers within the retina can be differentiated and retinal thickness can be measured.
I could not find any OCT image database. Kindly help me about such image database.
Thank you.
Relevant answer
Answer
You can find some OCT data from Dr. Hossein Rabbani's website. Here is the link:
I hope it is helpful.
  • asked a question related to Optical Coherence Tomography
Question
5 answers
We are doing research project on OCT images. The objective is to observe the progress of open angle Glaucoma OAG.
Is there any international database that you know available through the NET?
Many thanks for help
Relevant answer
Answer
It is always best to come up with a normative database for a particular population (as structural parameters have been reported to vary with age, race, ethnicity etc). We studied the normal distribution of OCT parameters in our population and used it subsequently for glaucoma diagnosis!!! 
  • asked a question related to Optical Coherence Tomography
Question
2 answers
At the Mouse Clinic in Strasbourg, we are using a Bioptigen R2200 SD-OCT to image mouse retinas.
Layer thicknesses are measured locally with calipers using the Bioptigen InVivoVue software, or eventually in manual mode with their Diver software, but it is a bit frustrating to have nice volumetric data of 400 x 400 or more, and just get a handful of values. The automatic segmentation feature of Diver does not seem (to me) fully mature yet (many steps to review the segmented data and eventually correct the mis-segmented part). The Iowa reference algorithm seem to work nicely on human retina, so I have tried their mouse implementation, called MouseSeg. It requires an awful lot of RAM (they say 40 Gb, but it uses more than 50 on my machine), and once segmented ... I just can't see the results, as the generated XML file is in a format not accepted by their OCTExplorer viewer (at least in the recent versions I can have access to). I have contacted them in June, and they are supposed to work on this issue ...
So I have two questions:
- is any one using MouseSeg on mouse data, and if so, what process are you using to visualize the segmentation results?
- is anyone using successfully any other segmentation software on mouse data ?
Best regards,
Michel
Relevant answer
Answer
Thanks Shanu, Diver is doing a rather good job, but it is not perfect. In your IOVS paper, the RNFL looks like what we are getting here. The blood vessels are not continuous, and they should be. We are also getting variations between the blue to green hues, and in our data they do not always correspond to changes in thickness, but to mis-segmentation. Overall, the average values you're getting are representative, but the segmentation data cannot be used for "flattening" without a lot of tweaking - and the tweaking has to be done using various apps outside of Diver. On human data; OCT Explorer from Iowa was looking more promising and smoother for correction / projections of flattened data - but the Mouse implementation does not seem to match up.
Michel
  • asked a question related to Optical Coherence Tomography
Question
2 answers
To do the spectral Analysis, why should we do the fourier transform instead of doing inverse fourier transform to get the wavelength dependent depth profile?
Relevant answer
Answer
Do you mean performing spectroscopic optical coherence tomography (SOCT) analysis? If yes than it is all about getting the right scale of your signal.
Lets consider that you do Fourier transform to go from the spectral-domain interferogram to the time-domain OCT A-scan. Then, to get wavelength dependent depth profile you need to multiplying scan by some window function and make inverse Fourier transform. If you do the normal Fourier transform (instead of inverse) you will get mirror image of your spectrum (wavelength scale will be fliped).
If you want to get the wavelength dependent depth profile directly from the spectral-domain interferogram, then it doesn't really matter what transform you will use because interferogram is a real function and its Fourier transform is antisymetrical. So, Fourier transform and inverse Fourier transform will have the exactly same magnitudes. Actually, it is up to you if you want to use Fourier transform or inverse Fourier transform. You just need to remember what is the scale of your results. 
I hope this helps. If not please give a little more details about the exact problem you are trying to solve.
  • asked a question related to Optical Coherence Tomography
Question
2 answers
I have a problem with OCT setup I am trying to make.
The setup is based on Michelson interferometer with mirrors in the sample and the reference arms. Light source is 639 nm high-power (up to 150 mW) diode laser running in CW mode with coherence length ~150 microns. After the laser I have non-polarizing cubic beamsplitter (50%/50%). The mirror in reference arm is modulated by PZT (2.3 kHz, ~375 nm amplitude). After the reference and sample beams combined I have high-speed photodetector connected to the oscilloscope with 50 Ohms termination load.
The photodetector is DET36A/M (http://www.thorlabs.us/thorcat/13000/DET36A-Manual.pdf) provided by Thorlabs (rise time ~14 ns, 350-1100 nm wavelength range, 13 mm^2 active area). I am supposed to get only one reference mirror position when I have constructive interference. In the case of constructive interference on the oscilloscope I see periodic 6-8 Intensity oscillations (mirror goes forward and backward) with Doppler frequency. The problem is in reality I have repetitive constructive interferences ~ every 8.8 mm (reference mirror position is 0 mm, 8.8 mm, 17.6 mm, etc.). And the oscillation amplitude on the oscilloscope is almost the same.
Any suggestions/comments what can be wrong in the setup?
Relevant answer
Answer
I have found out the reason to this. The coherence function of the diode laser is quite complex. It has quite long envelope (~several tens of cm) with periodic sharp (~150 microns) peaks. The manufacturer defines the width of the peak as the coherence length, not the width of the envelope. And this is what I actually see in the experiment.
  • asked a question related to Optical Coherence Tomography
Question
3 answers
I see full-field optical coherence tomography (OCT) as a technique to do OCT with a light bulb and a 2-D detector. In my opinion, this is attractive for two main reasons: 1) you lose the complexity associated with scanning mirrors and 2) Halogen lamps and CCD/CMOS sensors are generally cheaper than conventional FDOCT sources like SLDs/femto laser and line detectors. It's been many years now, and I still only see microscopy setups using full-field OCT. Why is that? Is the challenge mainly in setting up the Kohler illumination when a real eye is involved?
Relevant answer
Answer
Why would axial resolution be any better than any other OCT setup that uses a broadband source? SLD with bandwidths beyond 60 nm are available from Superlum, and femto laser have huge bandwidths. You didn't mention cost which makes it a different story but if that's what you meant, then you definitely get more bandwidth for your buck using a light bulb to do FFOCT.
Assuming you can design the illumination optics so that it would be appropriate for human eyes (pretty sure this is one of the biggest obstacles), I don't see why it would be any less comforting to the patient than standard flood illuminated fundus camera currently on the market.
  • asked a question related to Optical Coherence Tomography
Question
8 answers
I got the amplitudes and phases of each Cartesian components of non-paraxial vector field E(Ex, Ey ,Ez) and H(Hx, Hy, Hz), and note them as Ampex, Ampey, Ampez, Amphx, Amphy, Amphz, Phaex, Phaey, Phaez, Phahx, Phahy and Phahz. The amplitudes of E and H are |E|=sqrt(Ampex.^2+ Ampey.^2+ Ampez.^2) and |H|=sqrt(Amphx.^2+ Amphy.^2+ Amphz.^2), then, how to calculate the phases of E and H?
Relevant answer
Answer
I'm not sure what do you mean. In case of monochromatic electromagnetic wave its ordinary description is E(r,t) ~ E0 * exp (i*\omega t - k.r), where E, E0, k, and r are all 3D vectors. The expression within parentheses is called a phase. It is continuously changing with time (linearly when the observation position, r, is fixed), but it has exactly the same value for all 3 components of E at any moment of time. The same goes for H component.