Patent Application
20200281462
The present invention relates to an ophthalmic imaging apparatus, a control method for an ophthalmic imaging apparatus, and a computer-readable medium.
Currently, as ophthalmic apparatuses for observing and/or imaging an eye, for example, an anterior-eye imaging apparatus, a fundus camera, a confocal scanning laser ophthalmoscope (SLO) apparatus, an optical coherence tomography (OCT) apparatus using multi-wavelength light-wave interference, and so on are available. Among others, an OCT apparatus can obtain a high-resolution tomographic image of an object, and therefore, is becoming an ophthalmic apparatus specifically essential to the department of ophthalmology specializing in the retina.
In an ophthalmic OCT apparatus, in a case where the fundus of an eye to be inspected is imaged in a wide range, in a case where the fundus curves to a large degree, or in a case where a region to be imaged is long in the depth direction, the difference between the optical path length of reference light and the optical path length of irradiation light may be large. The optical path length of irradiation light means the optical path length of an optical path that is a combination of the optical path of the irradiation light and the optical path of reflected light thereof. If the optical path length difference is large, folding or chipping may occur in a tomographic image in the OCT apparatus, and it might not be possible to obtain a tomographic image that correctly represents the shape of the fundus.
Accordingly, a technique for flipping a tomographic image in which folding occurs, combining the flipped image with the original tomographic image, and detecting the retina by performing an image analysis is disclosed in Japanese Patent Application Laid-Open No. 2014-176566. This technique enables an appropriate shape analysis of the fundus in a wide range in the depth direction.
An ophthalmic imaging apparatus according to an aspect of the present invention includes a detector, a converter, and an arithmetic processing unit. The detector is arranged to detect, as an interference signal, interference light resulting from returning light and reference light, the returning light being light from an object to be inspected to which measurement light is radiated, the reference light corresponding to the measurement light. The converter is arranged to convert the detected interference signal that is an analog a to a digital signal. The arithmetic processing unit is configured to generate a tomographic image of the object to be inspected by using the converted interference signal. The arithmetic processing unit uses a plurality of components obtained from the converted interference signal to generate the tomographic image, the plurality of components including a component having a frequency higher than a Nyquist frequency of the converter and a component having a frequency lower than the Nyquist frequency of the converter.
Further features of the present invention will become apparent from the following description of exemplary embodiments with reference to the attached drawings.
The technique according to the related art is a technique in which a tomographic image in which folding occurs is flipped and connected to perform a shape analysis. Therefore, this technique does not remove the folding from the tomographic image. That is, it is not possible to obtain a tomographic image that is suitable to observation of the detailed structure of a region in which folding occurs.
Accordingly, an embodiment of the present invention has been made to obtain a tomographic image that is wider in the depth direction and in which folding is reduced.
For example, an ophthalmic imaging apparatus according to an aspect of the present embodiment uses a plurality of components (for example, information including the intensity and phase) obtained from an interference signal converted by a converter, the plurality of components including a component having a frequency higher than the Nyquist frequency of the converter and a component having a frequency lower than the Nyquist frequency of the converter, to generate a tomographic image of an object to be inspected. Accordingly, a tomographic image that is wider in the depth direction and in which folding is reduced can be obtained. The present embodiment will be described below.
The differential detector 120 converts the intensity of the interference light to an analog signal. The analog signal output from the differential detector 120 is separated. The analog signal that passes through a low-pass filter 121 includes only low-frequency components, and is converted to a digital signal by an analog-digital (AD) converter 131, which is an example of a converter. The analog signal that passes through a high-pass filter 122 includes only high-frequency components, and is converted to a digital signal by an AD converter 132, which is an example of another converter. The band of the low-pass filter 121 and that of the high-pass filter 122 can be electrically adjusted. The low-pass filter 121 and the high-pass filter 122 correspond to an example of a filter unit that attenuates a detected interference signal in accordance with predetermined frequency characteristics. These digital signals are processed by an arithmetic processing apparatus 114, which is an example of an arithmetic processing unit, by using a method described below. The arithmetic processing apparatus 114 performs a process including a Fourier transform on the digital signals to obtain information about the object to be inspected 107. A display unit 115 displays a tomographic image of the object to be inspected obtained by the arithmetic processing apparatus 114. The arithmetic processing apparatus may be connected to the ophthalmic imaging apparatus so as to enable communication. The arithmetic processing apparatus may be built into the ophthalmic imaging apparatus.
The above-described process is a process for obtaining a tomographic image at a certain point of the object to be inspected 107. Acquisition of information about a cross section of the object to be inspected 107 in the depth direction is called an A-scan. A scan for obtaining information about a cross section of the object to be inspected in a direction orthogonal to the A-scan, that is, for obtaining two-dimensional information, is called a B-scan. In the present embodiment, the B-scan is performed by the scanning mirror 108.
Acquisition of a tomographic image is described with reference to
At the position 201, the object to be inspected 107 is located at a depth position deeper than the depth position 240, and therefore, the optical path lengths differ, and a real image 202 is obtained. A Nyquist frequency 242 is a frequency half the sampling frequency of the AD converter 131 and is the maximum frequency of a signal that can be obtained (sampled) by the AD converter 131. When analog-digital conversion is performed by the AD converter 131, folding (aliasing) occurs around an axis which is the Nyquist frequency. A mirror image 203 is an image representing a signal having a frequency in which the signal is not distinguishable from a signal of the real image 202 when sampling is performed by the AD converter 131.
As can be seen from the above, in a case where the sampling frequency of the AD converter 131 is not sufficient for the range of the object to be inspected 107 in the depth direction, folding occurs in the tomographic image, and it is not possible to obtain a tomographic image that correctly represents the structure of the object to be inspected 107.
The position adjustment of the mirrors 110 and 111 described above is made as follows. A user adjusts the positions of the minors 110 and 111, which correspond to an example of an optical path length changing unit, by using an input unit not illustrated while watching the image illustrated in
In a case where the positions of the mirrors 110 and 111 are not appropriate, folding may occur at the upper end of the image. This is a case where the depth position 240 and the object to be inspected 107 overlap. In this case, frequency components of the signal are distributed across both sides of the DC component 241, and the negative range of the frequency is folded in the positive range of the frequency as a mirror image, which results in the occurrence of folding in the tomographic image. In an OCT apparatus according to the related art, the expression “folding in a tomographic image” sometimes refers to the above-described folding. On the other hand, the present invention assumes that the above-described folding at the upper end 243 of the image is avoided by adjusting the positions of the mirrors 110 and 111. As a result, folding due to the Nyquist frequency occurs at the lower end 244 of the image. A description is given below of a process that is additionally performed by the arithmetic processing apparatus 114 for suppressing the folding due to the Nyquist frequency and visually representing the structure of the object to be inspected 107 in a wide range in the depth direction.
For simplified description, the description given below assumes that no phase shift occurs in the filters. Further, the description given below assumes that signals obtained by the A-scan of the fundus are in phase across all frequencies. When such a signal is subjected to a Fourier transform, a frequency distribution formed of only real numbers is obtained. Therefore, only real numbers are used in the following description.
S
P
=P
L
S
L
+P
H
S
H
S
Q
=Q
L
S
L
+Q
H
S
H
When these equations are solved as simultaneous equations, the signal SL and the signal SH are calculated as follows.
Accordingly, the arithmetic processing apparatus 114 can obtain the signal SL and the signal SH by using the signals used to generate the tomographic image 312 and the tomographic image 322 and the transmittances of the filters. That is, the signal SL having a frequency lower than the Nyquist frequency and the signal SH having a frequency higher than the Nyquist frequency that overlap due to folding can be separated with the above-described method. The above-described calculation is performed for all frequencies other than the Nyquist frequency 242 to thereby obtain signals for the respective frequencies. At the Nyquist frequency 242, PL and PH match, and therefore, it is not possible to perform the above-described calculation. However, a signal obtained as a result of analog-digital conversion is used as is to thereby obtain the signal SL (which is a signal the same as the signal SH in this case).
The description given above assumes that no phase shift occurs in the filters and assumes that signals obtained by the A-scan of the fundus are in phase across all frequencies. However, in actuality, a phase shift occurs in the frequency filters. Further, in actuality, signals obtained by the A-scan of the fundus include signals having various phases, Calculation that takes into consideration these situations can be performed. Specifically, the signals SL and SH, the transmittances PL, PH, QL, and QH, and the luminances SP and SQ mentioned in the above description, which are all real numbers, need to be replaced with complex numbers. The transmittances PL, PH, QL, and QH can be obtained as complex numbers from the gain characteristics and phase characteristics of the filters. The signals SP and SQ can be obtained by performing a Fourier transform on signals obtained by the A-scan of the fundus. The signals SL and SH are obtained from the above-described calculation as complex numbers, and the intensities thereof need to be used as the luminances of the image. Note that a component described in the present embodiment is information that includes the phase as well as the intensity and is information that includes a complex number as well as a real number.
With the present embodiment, imaging in a short time is enabled for the following reason. It is not necessary to drive the mirrors 110 and 111 to change the optical path length of the reference light in order to obtain a plurality of images having different folding intensities, and only a single scan needs to be performed by the mirror 108 to obtain the tomographic image illustrated in
The values of the transmittances of the low-pass filter 121 and the high-pass filter 122 may be updated after an adjustment of the apparatus by taking into consideration, for example, environmental dependence. For example, the transmittances may be calculated from signals obtained from irradiation light reflected at a mirror (not illustrated) in the apparatus before and during an inspection. Alternatively, the values of the transmittances may be estimated on the basis of the luminance distribution of the obtained tomographic image illustrated in
Equations for calculating the signal SL and the signal SH may be equations other than the above-described equations. For example, in a case where the filters are close to ideal filters and the cutoff frequencies are close to the Nyquist frequency, substantially no folding portions are present in the tomographic image 312, and substantially only folding portions are present in the tomographic image 322. When the tomographic image 322 is vertically flipped and is simply connected with the tomographic image 312, a tomographic image close to a desired tomographic image can be obtained. This is equivalent to calculation in a case where PH and QL are set to 0 and PL and QHare set to 1 in the above-described equations.
The filters to be used need not be a combination of a low-pass filter and a high-pass filter. For example, a combination of only low-pass filters having different cutoff frequencies may be used. Even when only low-pass filters are used, if the values of the above-described PL and QL are different or the values of the PH and QH are different, similar calculation can be performed. Alternatively, taking into consideration frequencies higher than twice the Nyquist frequency, bandpass filters may be used.
Three or more filters may be used. Three or more filters having different frequency bands may be combined and similar calculation may be performed to obtain a tomographic image in a still deeper range. Further, signal intensities obtained by using a plurality of filters having the same bands may be averaged to increase the accuracy of luminance calculation.
The number of filters may be one. A signal that passes through the filter and a signal that does not pass through the filter may be used to perform similar calculation. In this case, the signal that does not pass through the filter is not attenuated substantially, and therefore, the values of PL and PH can be set to around 1. If the single filter is close to an ideal low-pass filter, QL can be set to 1, and QH can be set to 0. In this case, the above equation for SH corresponds to a simple subtraction of the signal that passes through the filter from the signal that does not pass through the filter. That is, with such a configuration, a tomographic image in a wide range in the depth direction can be obtained with a signal subtraction process.
Alternatively, one filter may be used and the band may be switched by time. When a plurality of tomographic images are successively obtained while the band is switched, a tomographic image in a deeper range can be generated from the plurality of tomographic images. In this case, in order to reduce the effect of motion of an eye to be inspected, a plurality of tomographic images are obtained in a short time. For this, the band of the filter is configured to be electrically changed as in the present embodiment. As the configuration for switching the band of the filter, for example, a configuration formed of a characteristic changing unit (for example, a variable resistor) that is provided in the filter and changes the frequency characteristics of the filter and a controller that controls the characteristic changing unit may be employed.
In the present embodiment, the values of the transmittances of the filters are determined by taking into consideration attenuation of signals caused by a factor other than the filters. That is, attenuation caused by a signal intensity decrease due to defocus and attenuation caused by the characteristics of filters in the AD converters are also regarded as attenuation caused by the low-pass filter 121 and the high-pass filter 122, and calculation is performed. However, these may be handled as separate parameters.
The sampling frequency of the AD converter 131 and that of the AD converter 132 may be different from each other. The arithmetic processing apparatus 114 may be configured to perform switching as to whether the above-described process is to be performed.
The final tomographic image (
The present embodiment assumes a case where a wavelength-sweep by the wavelength-swept light source 101 is stably performed at a constant rate for the wave number; however, a light source in which the constant rate is not maintained may be used. In this case, a configuration may be employed in which a k-clock for sampling at equal wave-number intervals in the light source or in the apparatus is generated and input to the AD converters. A generator of the k-clock is an example of a clock generator that generates a clock for the converters to sample analog signals. The generator of the k-clock may be configured as an interferometer in which an optical path through which part of light from the wavelength-swept light source 101 passes is branched into a first optical path and a second optical path having an optical path length difference relative to the first optical path. Even in a case where the rate at which a wavelength-sweep by the wavelength-swept light source 101 is performed is not constant, components having frequencies higher than the Nyquist frequency of the converters are regarded as components that oscillate in a time shorter than the time in which sampling of the interference signals (analog signals) by the converters is performed twice. Even in the case where the rate at which a wavelength-sweep by the wavelength-swept light source 101 is performed is not constant, components having frequencies lower than the Nyquist frequency of the converters are regarded as components that oscillate in a time longer than the time in which sampling of the interference signals (analog signals) by the converters is performed twice.
The present embodiment is applied to a swept-source (SS)-OCT; however, the present embodiment may be applied to the other OCTs. For example, in a spectral-domain (SD)-OCT, an optical low-pass filter may be disposed forward of a line sensor to fill a role equivalent to that of the low-pass filter 121 according to the present embodiment. A decrease in resolution based on the pixel size of the line sensor or the design of a diffraction grating, lens, etc. may be regarded as an effect equivalent to that of the low-pass filter, and the present invention may be applied.
As described above, according to the present embodiment, a signal having a frequency higher than the Nyquist frequency and a signal having a frequency lower than the Nyquist frequency can be separately calculated. Accordingly, a range of frequencies higher than the Nyquist frequency, that is, a deeper range of the fundus, can be visually represented, and a tomographic image that is wide in the depth direction can be obtained. Further, signals can be obtained in a time shorter than the time taken by the method for changing the optical path length, and an effect of small fixation movement of an eye to be inspected and an effect of fatigue can be suppressed.
Embodiment(s) of the present invention can also be realized by a computer of a system or apparatus that reads out and executes computer executable instructions (e.g., one or more programs) recorded on a storage medium (which may also be referred to more fully as a ‘non-transitory computer-readable storage medium’) to perform the functions of one or more of the above-described embodiment(s) and/or that includes one or more circuits (e.g., application specific integrated circuit (ASIC)) for performing the functions of one or more of the above-described embodiment(s), and by a method performed by the computer of the system or apparatus by, for example, reading out and executing the computer executable instructions from the storage medium to perform the functions of one or more of the above-described embodiment(s) and/or controlling the one or more circuits to perform the functions of one or more of the above-described embodiment(s). The computer may comprise one or more processors (e.g., central processing unit (CPU), micro processing unit (MPU)) and may include a network of separate computers or separate processors to read out and execute the computer executable instructions. The computer executable instructions may be provided to the computer, for example, from a network or the storage medium. The storage medium may include, for example, one or more of a hard disk, a random-access memory (RAM), a read only memory (ROM), a storage of distributed computing systems, an optical disk (such as a compact disc (CD), digital versatile disc (DVD), or Blu-ray Disc (BD™), a flash memory device, a memory card, and the like.
While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. The scope of the following claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2017-208428 | Oct 2017 | JP | national |
| 2018-147777 | Aug 2018 | JP | national |
This application is a Continuation of International Patent Application No. PCT/JP2018/039174, filed Oct. 22, 2018, which claims the benefits of Japanese Patent Application No. 2017-208428, filed Oct. 27, 2017, and Japanese Patent Application No. 2018-147777, filed Aug. 6, 2018, all of which are hereby incorporated by reference herein in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP2018/039174 | Oct 2018 | US |
| Child | 16857034 | US |
