Apparatus and methods for enhancing optical coherence tomography imaging using volumetric filtering techniques

The present invention relates to an apparatus and method for improving the function of OCT (Optical Coherence Tomography) image formation, and more particularly, to three-dimensional filtering of measurement data that can be used to generate these OCT images. Devices and methods that can improve the contrast of OCT images using techniques that utilize (eg, volumetric filtering). (Cross-reference to related applications) This application claims the priority of U.S. Patent Application No. 60 / 840,213 filed on August 25, 2006, the entire disclosure of which is incorporated herein by reference. , Incorporated herein.

For example, the possibility of using "OCT (Optical Coherence Tomography)" as a diagnostic procedure and a diagnostic technique capable of providing a high-resolution cross-sectional image of a tissue microstructure with a depth of 2 mm has been fully recognized. There is. However, in some clinical applications, the usefulness of conventional OCT techniques in diagnosis is limited by the confounding effect of speckle noise. This speckle noise can be a large amplitude noise on the size scale of image formation resolution, but this speckle noise can be used to provide a depth tomographic image of the tissue under evaluation. It may be generated by coherence ranging technique). Specific clinically relevant structures are unique enough to clearly identify the structure to surrounding tissue due to this speckle noise, even though it is larger in size than the image formation resolution of about 10 mm. It may not be possible to secure the light scattering contrast.

An already proposed method for reducing the effects of speckle noise is physical compounding. It can be classified as either method) or digital processing. For example, physical composite methods generally work by combining multiple speckle-uncorrelated measurements at the same location in the tissue being analyzed. Implementations of such methods may require changes to the imaging system, which can complicate catheter design and probe design with minimal invasion. Examples of these physical composite methods can include angle composite methods, frequency composite methods, and polarization composite methods (eg, polarization diversity detection). In contrast, digital processing methods that use procedures or filters aimed at selectively removing speckle noise while retaining certain features related to tissue structure have traditionally been exclusively used. It has been applied to two-dimensional images. Such digital processing methods include adaptive filtering, normalization, and noise reduction by the wavelet method. However, unlike the physical composite method described above, this digital processing method is likely to be limited to the information content contained in the image having the original speckle noise. Therefore, it is important that this digital processing method has a performance that exceeds the considerably high ability of an experienced OCT developer who visually filters noise and recognizes the underlying tissue structure. is there.

However, such restrictions will generally not apply if the digital processing method is extended to work in three dimensions for volumetric OCT datasets. Some improvements in OCT imaging rate have enabled clinical practice of volumetric imaging using the OCT method and OCT system. Therefore, there is currently potential clinical motivation to utilize such OCT methods and OCT systems as tools for a wide range of disease screenings. Since these 3D datasets cannot be directly visualized, it is usually possible to perform a diagnosis based on one or more tomographic images from the dataset. Preferably, these tomographic images include measurement information obtained from both in-fault planes (eg, in-plane measurements) and adjacent locations outside the tomographic planes (eg, out-of-plane measurements). It is possible.

In fact, there may be a need to overcome at least some of the shortcomings associated with conventional devices and methods as described above. For example, this need can be achieved by performing volumetric filtering on the dataset prior to tomographic imaging. In such a typical process, the information content of the resulting image can be increased by incorporating the out-of-plane measurement, so that it is possible to realize a significant improvement in function.

<p><patcit num="1"><text>International Patent Application PCT / US 2004/029148</text></patcit><patcit num="2"><text>U.S. Patent Application No. 11 / 266,779</text></patcit><patcit num="3"><text>U.S. Patent Application No. 10 / 501,276</text></patcit></p>

<p> To process and / or overcome at least some of the problems and / or drawbacks described above, the present invention provides three-dimensional filtering of measurement data (eg, for example) that can be used to generate these images. It is an object of the present invention to provide an exemplary embodiment of an apparatus and method capable of improving the contrast of an OCT image using a technique utilizing (volumetric filtering).</p>

<p> According to an exemplary embodiment of the invention, it is possible to filter the dataset in three dimensions so that an image with improved functionality can be generated. In a first exemplary embodiment of the invention, an asymmetric volumetric median filter is applied to a 3D OCT dataset prior to generating an image of a particular tomographic surface of the tissue. Is possible. In this exemplary embodiment, the filtering kernel may have out-of-plane dimensions greater than in-plane dimensions with respect to the tomographic plane of the image. In the second exemplary embodiment of the invention, the dithered beam (dithered) Provides an OCT image formation system that can generate scan patterns for beam) (dithering beams) and enable high-fidelity volumetric image formation to be performed in the presence of large sample movements. It is possible to do. By appropriately filtering the data set acquired by this dithering beam, it is possible to generate a two-dimensional image with improved functions.</p><p> Accordingly, exemplary embodiments of the devices and methods according to the invention can be provided at least for the reasons described above. For example, it is possible to provide at least one first fiber component and at least one second fiber component (each of these first fiber component and second fiber component has light transmission characteristics. ). The first fiber component can be configured to transmit at least one electromagnetic radiation and transfer at least one electromagnetic radiation to at least one sample. The second fiber component can be configured to transmit at least one electromagnetic radiation received from the sample, and at least a part of the first fiber component can be accommodated therein. ..</p><p> According to another exemplary embodiment of the present invention, the first and second fiber components may be fibers, respectively. Filtering the first and second fibers using at least one of the first and second filtering components, each of the individual transmitted and received electromagnetic radiation with a particular wavelength. It is possible to prevent at least part of it from being transferred within it. Furthermore, the received electromagnetic radiation may be Raman radiation associated with the sample.</p><p> Furthermore, according to certain exemplary embodiments of the invention, it is possible to provide the devices and methods according to the invention, in which case the particular radiation, including at least one first electromagnetic radiation, is at least. One sample is directed and at least one second electromagnetic radiation is directed to the reference. It is possible to apply a first electromagnetic radiation with a particular cross-sectional width to at least a portion of the sample to produce at least one third electromagnetic radiation. The first electromagnetic radiation can be provided within at least a portion of the sample along a particular axis for a distance corresponding to a given multiplier between a multiplier of 0.5 and a multiplier of 100 for a particular cross-sectional width. It is possible that there is. It is possible to detect interference between the third electromagnetic radiation associated with the first electromagnetic radiation and at least one fourth electromagnetic radiation associated with the second electromagnetic radiation. Furthermore, it is possible to provide an asymmetric cross-section area for the first electromagnetic radiation.</p><p> According to another exemplary embodiment of the invention, the upper limit of the multiplier may be 50, 60, 70, 80, and / or 90 for a particular cross-sectional width. The first electromagnetic radiation can translate within at least a portion of the sample along a further axis that is different from a particular axis. It is possible to generate at least one image associated with at least a portion of the sample as a function of interference. The first electromagnetic radiation is translatable within at least a portion of the sample in sinusoidal, triangular, serrated, and / or spiral patterns.</p><p> In yet another exemplary embodiment of the invention, the first electromagnetic radiation can have a particular cross-sectional width along a particular axis, this particular cross-sectional width other than the particular axis. It may exceed the further cross-sectional width of the first electromagnetic radiation along any other axis. Also, the first electromagnetic radiation can have a specific cross-sectional width along a specific axis, and this specific cross-sectional width is a further addition of the first electromagnetic radiation along a specific axis and another axis. At least twice the cross-sectional width. Furthermore, the first electromagnetic radiation can be translated within at least a portion of the sample along a further axis, which may be approximately perpendicular to a particular axis. It is possible to modulate the amplitude profile and / or phase profile of the first electromagnetic radiation. It is possible to provide at least one of a spatial light modulation component, a galvanometer component, an acoustic-photo modulation component, a waveguide mode scramble component, and / or an asymmetric waveguide component. The asymmetric waveguide component can be configured to propagate at least three orthogonal modes of first electromagnetic radiation.</p><p> The above objectives, features, and advantages of the present invention, as well as other objectives, features, and advantages, provide the following detailed description of the embodiments of the present invention in the context of the claims of the appended claims. It will be clear by reference.</p><p> On the other hand, further objectives, features, and advantages of the present invention are made clearer by reference to the following detailed description in the context of the accompanying drawings showing some exemplary embodiments of the present invention. Will be.</p>

<figref num="1A">FIG. 6 is a schematic block diagram of an exemplary embodiment of an OFDI (Optical Frequency Domain Imaging) system according to the invention that can be used to acquire volumetric data sets.</figref><figref num="1B">It is a typical irradiation figure by an example example of this invention which is a pattern which made the raster scan.</figref><figref num="1C">FIG. 6 is a typical irradiation diagram according to another exemplary embodiment of the present invention, which is a scan pattern of a dithered beam.</figref><figref num="2A">This is a typical image without filtering.</figref><figref num="2B">Improved functionality such as applying a volumetric median filtering procedure with various in-plane and out-of-plane dimensions of kernel size according to another exemplary embodiment of the invention. This is a typical set of images.</figref><figref num="2C">A second typical with improved functionality such as applying the volumetric median filtering procedure according to an exemplary embodiment of the invention, which has a kernel size similar to that of FIG. 2A. It is an image.</figref><figref num="3A">This is a typical image in which the function was not improved by applying the dithering beam.</figref><figref num="3B">It is an exemplary image whose function is improved by applying a dithering beam having a first amplitude according to an exemplary embodiment of the present invention.</figref><figref num="3C">It is another typical image whose function is improved by applying a dithering beam having a first amplitude according to another exemplary embodiment of the present invention.</figref><figref num="4A">FIG. 5 is a diagram of a first typical dithering beam scan pattern according to an exemplary embodiment of the present invention.</figref><figref num="4B">FIG. 5 is a diagram of a second typical dithering beam scan pattern according to another exemplary embodiment of the present invention.</figref><figref num="4C">FIG. 5 is a diagram of a third typical dithering beam scan pattern according to a further exemplary embodiment of the present invention.</figref><figref num="5A">It is a figure of the symmetric ellipse image formation beam by an exemplary example of this invention.</figref><figref num="5B">FIG. 6 is a diagram of an asymmetric elliptical image forming beam according to a specific exemplary embodiment of the present invention.</figref><figref num="6A">FIG. 5 is a schematic front view of a typical asymmetric waveguide component that can be used to transmit an image forming beam from an image forming system according to an exemplary embodiment of the present invention to a sample.</figref><figref num="6B">FIG. 6 is a side view of a typical endoscopic optical image forming probe according to an exemplary embodiment of the invention in which the fibers shown in FIG. 6A can be used.</figref><figref num="6C">It is a top view of the schematic view of the probe of FIG. 6B.</figref><figref num="7">It is an operation diagram which shows the typical method for reducing speckle by using a rectangular core fiber in combination with a rotating mirror by an exemplary embodiment of the present invention.</figref><figref num="8">It is an operation diagram showing another typical method for reducing speckle using a rectangular core fiber in combination with a linear spatial light modulator according to another exemplary embodiment of the present invention.</figref><figref num="9">FIG. 5 is an operation diagram showing a typical method for modulating the optical phase profile / optical amplitude profile of a rectangular mode fiber according to an exemplary embodiment of the present invention.</figref><figref num="10">An endoscopic image forming component according to a specific exemplary embodiment of the present invention, typically for using a piezoelectric actuator to generate a dithering beam to vibrate the fiber end prior to focusing by a lens. It is a figure of a typical endoscopic image forming component.</figref>

Throughout the drawings, unless otherwise noted, the same reference numbers and letters are used to describe similar features, elements, components, or parts of the illustrated embodiments. Further, the present invention will be described in detail below with reference to the drawings, but this description will be given in the context of exemplary embodiments. Interpretation that it is possible to make changes and modifications to the illustrated examples without departing from the true scope and spirit of the invention as defined by the appended claims. I want to be.

FIG. 1A shows a diagram of a second generation OCT (Optical Coherence Tomography) image forming system based on the "OFDI (Optical Frequency Domain Imaging)" technique according to an exemplary embodiment of the present invention. The system in Figure 1A utilizes a wavelength sweep narrowband laser source (wavelength sweep laser) 100 and is capable of recording interference fringes as a function of wavelength using a single element photo receiver. is there. Although the exemplary system shown in FIG. 1A is described herein as being capable of utilizing OFDI technology, other exemplary embodiments of the methods and devices according to the invention are described. , Similarly, other OCT image forming systems including, but not limited to, these time domain OCT technology and spatial domain OCT technology are also available.

As shown in FIG. 1A, the light or other electromagnetic radiation supplied by the wavelength sweep narrowband laser source 100 can be split in the splitter 105 into a reference path 106a and a sample path 106b. The sample path 106a can be directed to the sample 140 via an optical circulator 120, a two-dimensional galvanometer mirror (two-dimensional galvanometer scanner) 130, and a focusing lens 135. The reference light is directed through the reference path 106b. This reference path may be intended to match the optical path length of sample path 106a. Certain typical configurations are known to provide such functionality. These typical configurations include the non-reflective path and configuration shown in FIG. 1A such that a circulator can be used to direct the reference light to the variable delay line 115. The returned reference light and sample light interfere with each other in the coupler 145. The output beam from the combiner 145 can be directed to the first polarization beam splitter (PBS) 150a and the second polarization beam splitter (PBS) 150b. The outputs of these first polarization beam splitters and second polarization beam splitters can be directed to the first balanced receiver 155a and the second balanced receiver 155b, respectively.

Using conventional processing techniques, it is possible to convert the measured interference fringes into an A-line that represents the depth resolution reflectance in the sample. An exemplary image can be obtained by scanning the image forming beam 136 in two dimensions using the two-dimensional galvanometer mirror 130. For example, it is possible to generate any beam scan pattern in the xy plane on the surface of sample 140. As shown in FIG. 1B, volumetric image formation, for example, by scanning rapidly along the x-dimension and repeating for various displacements (160a, 160b, 160c, and 160d) in the y-dimension. It is possible to carry out the technology. Alternatively, or in addition to this, as shown in FIG. 1C, the beam oscillates rapidly in the y-dimension while scanning at a lower speed in the x-dimension (165). It is also possible to perform dithering. With such a typical scanning technique, it is possible to smoothly record an image in three dimensions, and it is possible to realize the application of a volumetric filtering method.

Figure 2A was obtained from the dataset using the scan pattern shown in Figure 1B without applying the volumetric filtering method, as shown in the large speckle noise in the baseline image. A typical single section image 190 (in xz plane) is shown. Figures 2B and 2C are typical taken from human skin using the scan pattern of Figure 1B at various sizes of volumetric median filters operating in the in-plane (xz) and out-of-plane (y) dimensions. The OFDI image is shown.

For example, a subset of the image shown in Figure 2A, including the border between the epidermis and the dermis, is shown in Figure 2B unfiltered (leftmost image 191), and out-of-plane filtering. Images when the kernel size of the in-plane filter is increased without doing so are shown in images 192 to 194 in the upper row. Illustrative results for these images demonstrate the capabilities and limitations of traditional 2D (two-dimensional) median filtering algorithms. In these images, considerable blurring is formed due to the reduction of speckle. The lower columns 195-197 of the image shown in Figure 2B provide the effect of increasing the kernel size of the out-of-plane filter (without in-plane filtering). In this case, structural visibility is clear without blurring the features, despite using a filter size equal to that used for the results of increasing the kernel size of the in-plane filter. It is possible to observe the improvement. These exemplary results are preferred image enhancements obtained by the use of volumetric filters such that the size is significantly asymmetric with minimal in-plane filtering on the tomographic plane of the image and large-scale out-of-plane filtering. Provides the results of. FIG. 2C shows a typical cross-sectional image 198 generated based on a combination of in-plane and out-of-plane filtering estimated to achieve optimal image enhancement.

Figure 3A shows typical OFDI images 50, 51 generated without the correction of the OFDI image by scanning the dithering beam (eg, the dithering amplitude between peaks is 0 μm). Shown. FIGS. 3B and 3C show typical images 52, 53 and 54, 55, respectively, whose function has been improved by applying dithering beams having different amplitudes according to an exemplary embodiment of the present invention. For example, a scan using a dithering beam scan can be performed using a 2D (two-dimensional) galvanometer with a y-axis mirror driven by a sinusoidal waveform whose amplitude changes at 500 Hz. For example, in the case of a system A-line rate of 10 kHz, a single dithering period can contain 20 unique A-lines. A typical filtering method can be executed by organizing the acquired data set as a single image and applying a 2D (two-dimensional) median filter. To determine the preferred dithering amplitude, it is possible to obtain biometric values of human skin, for example, at a dithering amplitude between peaks of 0 μm to 70 μm, for example by a step of 17.5 μm. is there. An exemplary median filtering method is performed over a single dithering period to generate, for example, an in-plane filter size of 5 μm (x dimension) x 7.5 μm (z) and an out-of-plane filter size that varies from 0 to 70 μm. Is possible. Figures 3B and 3C show the resulting images 52, 53 and 54, 55 at dithering amplitudes between peaks at 35 μm and 70 μm, respectively.

4A-4C show three exemplary dithering beam scan patterns according to the exemplary embodiments of the present invention. For example, FIG. 4A shows a typical sinusoidal scan pattern 200 that includes fast zero average modulation in the y dimension and a slow constant velocity scan operation in the x dimension. FIG. 4B shows a typical spiral scan pattern 205 that can be generated by scanning the x and y dimensions with a 90 degree phase difference at the same or similar frequencies. It also includes a fixed speed slow scan operation in x-dimensional. FIG. 4C shows a typical orthogonal scan pattern 210, in which the y dimension is driven by a fast sawtooth pattern and the x dimension is driven by a slow fixed velocity scan. There is. The degree of displacement in the y dimension may be, for example, about 0.5 to 100 times the cross-sectional width of the focused beam.

FIG. 5A shows a typical symmetric image formation beam profile that can achieve sample image formation with unequal in-plane and out-of-plane resolution scales. FIG. 5A shows a circular Gaussian beam focal point, in which case the beam profile at focal point 300 is symmetrical in x-dimensional (scan direction) and y-dimensional (out-of-plane dimension). FIG. 5B shows a typical asymmetric image formation beam profile that can achieve sample image formation. In FIG. 5B, it is possible to use a relatively large asymmetric beam profile 305 in the y dimension. The beam scan shown in FIG. 5B can be performed by using a combination of spherical focusing optics and cylindrical focusing optics, or by using a non-circular waveguide, as shown in FIGS. 6A-6C. , Can be generated.

For example, FIG. 6A shows a typical asymmetric waveguide component that can be used to transmit an image forming beam from an image forming system to a sample. For example, the core 400 of this waveguide component, which may optionally be a glass optical fiber or a photonic band-gap fiber, is one in relation to another dimension. It is possible to have a relatively large spread in dimension. The cladding 405 of the waveguide component may be circular or asymmetric in shape, as shown in FIG. 6A. As a result, by imaging this image-forming beam on the sample using spherical focusing optics, it is possible to obtain a similarly asymmetric beam profile on the sample as a result.

FIG. 6B shows a typical endoscopic optical imaging probe according to an exemplary embodiment of the invention in which the fiber shown in FIG. 6A can be used, thereby sampling an asymmetric imaging beam profile. Can be obtained as a result above. For example, an optical fiber with an asymmetric core 410 can be rotated and placed within the external drive shaft 425 to increase torque transmission. At the ends of the optical fiber, light can be magnified through portion 411, such as air or amorphous glass, focused by the lens 415, and directed laterally by the prism 420 or mirror. is there. The lens 415 can generate focused spots at a distance of approximately Δr from the fiber probe. The angular direction of the fiber allows the focused beam to be larger in the z dimension than in the x dimension, as shown in FIG. 6B. By rotating the fiber or drive shaft 425, the asymmetric imaging beam can be translated, and thereby the imaging of hollow organs can be performed smoothly. FIG. 6C shows a front view of the image forming probe, which shows a relatively tight focus and a relatively small spot size in the x dimension compared to the y dimension.

FIG. 7 shows an operation diagram of a typical method for reducing speckle using a rectangular core fiber in combination with a rotating mirror according to an exemplary embodiment of the present invention. FIG. 7 also shows a configuration according to an exemplary embodiment of the invention configured to couple an image forming beam from the system to the proximal end of a rectangular core fiber (eg, used for image forming). ing. For example, according to the configuration of FIG. 7, it is possible to modulate the phase and amplitude profiles of the image forming beam at the distal end of the fiber. In this exemplary configuration, the asymmetric core 502 of the fiber can be configured to support multiple optical modes in the dimension of the core with relatively large spreads. By exciting each mode or different combinations of these modes, it is possible to obtain multiple measurements of reflectance, each with uncorrelated speckle noise. By combining these measured values, it is possible to realize image formation with reduced speckle. As shown in FIG. 7, it is possible to direct the Gaussian symmetric input beam 510 from the image forming system to the galvanometer mirror 515. The mirror 515 is capable of guiding light through the lens 505 and focusing the light on the core at various lateral positions within the rectangular spread of the core. By tilting the galvanometer mirror 515, it is possible to modulate the excitation phase profile / excitation amplitude profile of the core at the proximal end and, therefore, at the distal end. A probe design similar to that shown in Figure 6B can be used for endoscopic imaging. For example, the returned light can be recollected through the same or similar optics.

FIG. 8 shows an operation diagram of a typical method for modulating the optical phase profile / amplitude profile of a rectangular core fiber according to an exemplary embodiment of the present invention, which is shown in FIG. It may be similar to what you are doing. FIG. 8 also shows another exemplary embodiment of the configuration in which this exemplary method can be performed. However, as shown in FIG. 8, the linear spatial light modulator 615 can replace the galvanometer mirror 515. For example, the input beam 610 from the system can be passed through a linear spatial light modulator 615 that has the ability to quickly change the phase and / or amplitude profile of the beam. This light can be focused through the lens 605 onto the core 602 of the rectangular core fiber 600. Instead of the linear spatial light modulator 615, it would be possible to change the beam profile using an acoustic-optical modulator or an electrical-optical modulator.

FIG. 9 shows an operation diagram of a typical method for modulating the optical phase profile / optical amplitude profile of a rectangular mode fiber according to an exemplary embodiment of the present invention. For example, the light in the core 702 passes through a portion of the fiber placed between the rigid stationary support 705 and the actuator 710. By activating the actuator 710, it is possible to generate a downward or upward movement that changes the compressive stress in the fiber, and it is possible to change the mode profile. The actuator 710 may optionally be a piezoelectric stack actuator, and the orientation of the core 702 with respect to the actuator may be that shown in FIG. 9 or may be rotatable. ..

FIG. 10 shows an endoscopic image forming component according to an exemplary specific embodiment of the present invention, which is a piezoelectric actuator type small diameter endoscopic image forming probe for vibrating the fiber end before focusing by a lens. Shows a diagram of a typical endoscopic image formation component for generating a dithering beam scan (similar to the pattern shown in Figure 4A). As shown in FIG. 10, the optical fiber 800 is capable of directing the image forming light to the focusing lens 820. The piezoelectric actuator 805 can be driven by a sinusoidal signal so that the tip of the optical fiber vibrates. The light from the optical fiber can be magnified within the air gap 815, focused by the lens 820, and directed laterally by the prism 825. As a result of vibrating the tip of the optical fiber, the focused spots of the image forming beam 830 can vibrate in the illustrated dithering pattern 835. The rotation of the entire catheter housed within the housing 810 allows, for example, to scan the beam internally within the hollow cylindrical organ 840.

The contents described above merely exemplify the principle of the present invention. In view of the disclosures herein, various modifications and variations to the embodiments described above will be apparent to those skilled in the art. In fact, the devices, systems, and methods according to the exemplary embodiments of the present invention can be used with the image forming system and, for example, the international patent application PCT / US2004 filed on September 8, 2004. 029148, as described in US Patent Application No. 11 / 266,779 filed on November 2, 2005, and US Patent Application No. 10 / 501,276 filed on July 9, 2004. Can be used with. All disclosures of these patent application specifications are incorporated herein by reference. Thus, one of ordinary skill in the art, although not explicitly illustrated and described herein, is a number of systems, devices, and methods for realizing the principles of the invention, and the spirit and spirit of the invention. It should be understood that it is possible to come up with a number of systems, devices, and methods that belong to the scope. Furthermore, all prior art knowledge is expressly included herein, even if it is not explicitly incorporated herein by being cited herein as described above. .. All the documents cited above are incorporated herein by reference in their entirety.