WO2011020482A1

WO2011020482A1 - Method and apparatus for optical coherence tomography

Info

Publication number: WO2011020482A1
Application number: PCT/EP2009/006016
Authority: WO
Inventors: Tomasz Bajraszewski; Piotr Targowski; Pawel Dalasinski; Maciej Szkulmowski; Pawel Wojdas
Original assignee: Optopol Technology S.A.
Priority date: 2009-08-19
Filing date: 2009-08-19
Publication date: 2011-02-24

Abstract

The invention relates to optical coherence tomography (OCT), more specifically to spectral optical coherence tomography (SOCT). The invention deals with problem of artefacts occurring in certain scan positions due to sharp increases of reflection by the object under investigation. It solves this problem by using an extra intensity sensor located at an output for non-dispersed light of the dispersive element used for spectral analysis, e. g. in the zero order of a diffraction grating, and using the signal of this sensor for the modification of the interferometric data.

Description

Method and apparatus for optical coherence tomography The invention relates to optical coherence tomography

(OCT) , more specifically to spectral optical coherence tomography (SOCT) .

OCT is a technique to explore three-dimensional structures of partially reflecting or back-scattering matter with electromagnetic waves in the optical range (wavelengths between 600 nm and 1,700 nm) . The fundamentals of OCT were presented by Huang et al. in Science, Vol. 254 (1991), p. 1178 - 1181. Huang et al. placed the object to be investi- gated in the object arm of an interferometer. A piezoelectric transducer was located in the reference arm in order to modulate the optical path difference (OPD) of the reference beam. Numerical evaluation of the signal generated by the interferometer (interferometric signal) and the modu- lated OPD facilitated the calculation of the spatial distribution of reflecting or back-scattering structures along the axis of the object beam. Vertical and horizontal scanning of the object beam would allow for three-dimensional imaging of the internal structure of the object.

Application of this time domain OCT to white light inter- ferometry was straightforward (DE 43 09 056 Al) . In white light interferometry, the modulation of the OPD in the reference beam is replaced by the use of light at different wavelengths. The interferometric signal is spectrally analyzed to provide a spectral distribution of the light in- tensity. Numerical evaluation of this spectrum allows for the calculation of the spatial distribution of reflecting or back-scattering structures along the axis of the object beam (cf. DE 43 09 056 Al, US 2008/0013093 Al, Lin An et al., OPTICS LETTERS (Vol. 32) 2007, 3423-3425).

There are several effects which are detrimental for the accuracy of spectral optical coherence tomography measurements. As far as noise is concerned, it is generally desir- able to keep the levels of intensity noise and beat noise below the level for shot noise (shot noise limited detection) . In order to obtain shot noise limited detection with objects with low reflexion or back-scattering characteristics, the intensity of the reference arm is normally ad- justed to a level close to the saturation of the detector (e. g. CCD camera) . To this end, the source power may be monitored by splitting off a portion of the light and measuring the intensity of this portion (US 2007/0291277 Al) . However, such arrangements fail in cases where certain ef- fects cause the intensity of the object beam to increase strongly in certain scan positions. For example, when the human eye is examined, the convex curvature of the cornea and its good reflexion characteristics may cause the intensity of the object beam to increase very sharply when the object beam impinges centrally on the pole, the tangent plane of which is perpendicular to the beam. While this effect may be exploited for the measurement of the corneal optical thickness in the absence of corrections of refraction (Kaluzny et al., Optica Applicata, Vol. XXXII, No. 4, 2002, 581 - 589, 585), it may easily cause the intensity of the interference signal to exceed the level of saturation of the detector. As a consequence, artefacts occur during the numerical evaluation. It has also been shown that coherent noise parasitic elements in the interferometric signal may be suppressed effectively by keeping exposure time and optical power within certain limits (Szkulmowska et al., J. Phys . D: Appl. Phys . 38 (2005) 2606 - 2611) . Again, this measure is of no help if a disturbance of the interference signal occurs in certain scan positions only. The present invention seeks to mitigate this problem of artefacts occurring in certain scan positions. The solution is represented by the subject-matter of claim 1. The further claims describe improvements thereof. An interferometer within the meaning of the present invention is defined as a device which can split an incoming beam of light into at least two beams of light, the reference beam and the object beam, and recombine both beams after reflection and/or back-scattering by different media to form an exiting beam of light. If the light is sufficiently coherent, the exiting beam of light contains information about the structures of the media. If one of these structures is known, e. g. if such medium is a mirror, information about the three-dimensional structure of the other me- dium may be extracted from the exiting beam.

Means for spectral analysis within the meaning of the invention is defined as an assembly which can measure the intensities of light separately for a plurality of, however at least two, preferable more than 200, spectral ranges. Such ranges shall normally as narrow as possible and consist ideally of a single wavelength only. Any dispersive element can be employed for said means for spectral analysis, as long as it has an output for non- dispersed light. For instance, if a prism is used, the residual reflectance from the first surface may serve as out- put for non-dispersed light. However, due to the superior dispersion efficiency of gratings, said dispersive element is preferably a diffraction grating where the zero order diffraction position may serve as output for non-dispersed light.

Means for spectral analysis always comprise at least one sensor element, usually a plurality of sensor elements, e. g. a CCD camera. A sensor element in this context is a device which generates a signal as a function of the inten- sity of incoming light. Spectral analysis according to the invention is defined as operating such means for spectral analysis .

Interferometric data according to the invention are defined as data representing the light intensities obtained by spectral analysis of the beam of light exiting from an interferometer .

Means for numerical evaluation within the meaning of the invention comprises the combination of a computer system and software performing the numerical steps. The numerical evaluation involves computation of the interferometric data to obtain SOCT images. An intensity sensor device according to the invention measures the intensity of incoming light and generates an intensity signal as a function thereof. This intensity signal may represent the instant intensity or the light quantity reaching the sensor device during a certain period of time, so that the light intensity is integrated over time. It follows that the intensity signal can be dependent on the instant intensity or on the integrated intensity or on a combination thereof.

According to the invention, the intensity sensor device is arranged to receive light from the output for non-dispersed light of the dispersive element, e. g. in the zero order of the diffraction grating. As a consequence it will receive a portion of the non-diffracted and thus not spectrally decomposed light. This light is perfectly eligible for the generation of a signal representing the overall intensity of the light. This intensity signal is utilized for modify- ing the interferometric data so that artefacts caused by too strong light intensities can be suppressed. Placing the intensity sensor device in the zero order allows operating the device with as few optical components as possible. In particular, an additional coupler or other extra splitting device is not needed. Moreover, the overall signal strength is not attenuated by the intensity measurement, as the non- dispersed light, e. g. the zero order light of a diffraction grating, is lost anyway for spectral analysis. Modifying interferometric data according to the invention can be performed physically by controlling the quantity of light reaching one ore more of the sensor elements of the means for spectral analysis. Additionally or alternatively, such modification can be performed by modifying the elec- tronic data generated by the means for spectral analysis, especially within the numerical evaluation of the digitized data. It is specifically possible to perform such physical, electronic and/or numerical modification automatically so that the operator does not have to deal with artefact problems causes by saturation. Means for modifying the inter- ferometric data can, accordingly, be physical devices or numerical routines, specifically software routines.

Controlling the quantity of light reaching the sensor element (s) can be performed either by adjusting the intensity of the impinging light, or by adjusting the exposure time, or by a combination of these two measures. Adjusting the light intensity may be made, for instance, with the help of light source current control, a motorized gradient neutral density filter or a liquid crystal tunable filter. As such devices always have a certain response time to be taken into account, adjusting the exposure time is advantageous.

Adjusting the exposure time can easily be carried out by interrupting the exposure during an exposure cycle. An exposure cycle within the meaning of the invention is the period of time in which the sensor elements of the means for spectral analysis are exposed to the incoming light for measurement in one scan position. If a CCD camera is used within the means for spectral analysis, an exposure cycle is the period of time during which charge carriers within the CCD pixels are accumulated for the generation of a sig- nal proportional to the quantity of the impinging light.

Usually, such exposure cycles alternate with pixel resets.

Advantageously, a focusing lens is placed between said output for non-dispersed light and said intensity sensor de- vice and arranged for focusing said non-dispersed light emerging from said output on said intensity sensor device. This lens helps to better exploit the non-dispersed light and to make the intensity sensor device more sensitive. Advantageously, the interferometer of the apparatus according to the invention comprises splitting means arranged for splitting incoming light into a first portion to be di- rected into the object arm and a second portion to be directed into a reference arm, whereas said splitting means is further arranged for recombining light returning from the object arm with light returning from the reference arm and for directing the recombined light into a detection arm, whereas said splitting means is adapted to direct between 51 and 99 per cent, advantageously between 60 and 80 per cent, more advantageously 70 per cent, of the light returning from the object arm into the detection arm. The use of such an asymmetric splitting means helps to optimize the exploitation of the light intensity. In practical applications, the light in the object arm is attenuated to a much larger degree than the light in the reference arm. The aforementioned improvements causes a larger portion of the light returning from the object to contribute to the inter- ferometric signal. This improves the quality of the measurement .

More advantageously, said splitting means is a fiber optical coupler. This facilitates simple and robust setups.

It is also advantageous if the apparatus according to the invention comprises exposure controlling means arranged for adjusting the quantity of light impinging on at least one sensor element of the means for spectral analysis. In this case, it is even more advantageous if said exposure controlling means is arranged for interrupting the exposure for the sensor element during an exposure cycle, especially if it is arranged for interrupting the exposure in depend- ence of said intensity signal. Such improvements allow for an effective modification of the interferometric data and thus suppression of artefacts. An exemplary embodiment of the invention is explained in greater detail below with reference to the drawings, in which:

Fig. 1 is a schematic block diagram showing an apparatus according to the invention with which the method according to the invention can be performed;

Fig. 2 shows the time relationship of two triggering

signals used for carrying out the invention.

A beam of light is generated by a light source 1 and then coupled into an optical fiber system 2. The light beam propagates through the fiber system 2 to reach a coupler 3 acting as a beam splitter having a splitting ratio of

30/70. 70 percent of the light is directed into an object arm 4, while 30 percent is directed into a reference arm 5.

The object arm 4 comprises lenses and a galvoscanner set in order to focus the beam of light on and scan it across the object 6 in xy directions, which object is a human eye in this example. The reference arm 5 comprises lenses and a mirror to adjust the desired optical path delay. Polarisation controllers (not shown) may be placed along the fiber system 2 for both the beam in the object arm 4 and the beam in the reference arm 5. One or more neutral density filters (not shown) and dispersion compensators (not shown) may be placed in the reference arm 5, e. g. between two lenses, to adjust the intensity in order to reach an optimal signal to noise ratio.

In the Michelson interferometer setup as shown in this ex- ample, the beams of light in both arms 4, 5 return to coupler 3, where they are recombined. Due to the coupling ratio of coupler 3, 30 % of the recombined light passes the port of coupler 3 connected to the light source 1. An optical insulator (not shown) may be placed between the coupler 3 and the light source 1 to prevent returning light from reaching and damaging the light source 1. The other light portion of 70 % is leaves coupler 3 through the remaining port to enter the detection arm 7. In detection arm 7, the light exits the fiber system 2 and is collimated by a collimating lens 8. The collimated light beam impinges on a grating 9. The grating 9 decomposes the light into its spectral components which are focused by a first focusing lens 10 to impinge on the sensor elements of a line sensor array 11. The signals received by the sensor elements are fed to a computing system 12 for numerical evaluation and display on a screen 13 as known in the art of SOCT. According to the invention, the portion of light not diffracted by grating 9 (zero order) is focused by a second focusing lens 14 on an intensity sensor 15. This intensity sensor 15 generates an intensity signal which is dependent on, e.g. proportional to, the overall intensity of the light passing the grating 9 and reaching the line sensor array 11. This intensity signal is fed to an exposure control unit, which is integrated in the computing system 12 in this example. The exposure control unit generates first trigger signals

16 to trigger the commencements of exposure cycles of the sensor elements of said line sensor array 9. The exposure control unit further generates second trigger signals 17 triggering the end of the exposure cycles. As long as said intensity signal generated by the intensity sensor 15 remains below a predetermined value, the second trigger signal 17 is generated after constant periods of exposure time after the first trigger signal 16 so there is no interruption of the exposure cycles. If however, said intensity signal exceeds said predetermined value, the period of exposure time is shortened, i. e. the second trigger signal

17 is generated earlier. Fig. 2 illustrates this way of op- eration. The first trigger signal 16 is generated in a repetitive manner after constant periods of time Tc. As long as the light intensity and thus said intensity signal does not exceed a predetermined value, the second trigger signal 17 is generated after a constant period of exposure time Tl. Otherwise, it is generated after a shortened period of exposure time T2.

The conditions under which the second trigger signal 17 is generated may be selected as it is practicable for the par- ticular application. For instance, the higher the degree in which the intensity signal exceeds said predetermined value, the earlier the second trigger signal 17 will be generated. Likewise, the intensity signal may be integrated over time so that the second trigger signal 17 is generated as soon as the integral exceeds a predetermined value. It is also possible not to shorten the exposure signal but account for the intensity signal or its integral during the numerical evaluation.

Claims

Claims 1. Apparatus for spectral optical coherence tomography

comprising a light source (1) , an interferometer with an object arm (4), means for spectral analysis comprising a dispersive element having an output for non- dispersed light and arranged for generating interfer- ometric data, means (12) for numerical evaluation of the interferometric data, characterized in that said apparatus further comprises an intensity sensor device (15) arranged to receive light from output for non- dispersed light of said dispersive element of said means for spectral analysis, said intensity sensor device (15) being arranged for generating an intensity signal as a function of the light intensity, and means for modifying said interferometric data according to said intensity signal.

2. Apparatus according to claim 1, characterized in that said dispersive element is a diffraction grating (9) and the output for non-dispersed light is the zero order position.

3. Apparatus according to claim 1 or 2, characterized in that a focusing lens (14) is placed between said output for non-dispersed light of said dispersive element and said intensity sensor device (15) and arranged for fo- cusing the light emerging from said output on said intensity sensor device (15) .

4. Apparatus according to any of the preceding claims,

characterized in that said interferometer comprises splitting means (3) arranged for splitting incoming light into a first portion to be directed into the object arm (4) and a second portion to be directed into a reference arm (5), whereas said splitting means (3) is further arranged for recombining light returning from the object arm (4) with light returning from the reference arm (5) and for directing the recombined light into a detection arm (7), whereas said splitting means (3) is adapted to direct between 51 and 99 per cent, advantageously between 60 and 80 per cent, more advantageously 70 per cent, of the light returning from the object arm (4) into the detection arm (7).

5. Apparatus according to claim 4, characterized in that said splitting means is a fiber optical coupler.

6. Apparatus according to any of the preceding claims,

characterized in that it further comprises exposure controlling means arranged for adjusting the quantity of light impinging on at least one sensor element of the means for spectral analysis.

7. Apparatus according to claim 6, characterized in that said exposure controlling means is arranged for inter- rupting the exposure for the sensor element (s) during an exposure cycle.

8. Apparatus according to claim 7, characterized in that said exposure controlling means is arranged for inter- rupting the exposure in dependence of said intensity signal .