JP5197217B2 - Biological information imaging apparatus and image construction method - Google Patents

Description

  The present invention relates to a biological information imaging apparatus that receives and images ultrasonic waves generated by light absorption, and an image construction method thereof.

  Conventionally, imaging apparatuses using X-rays (mammography), ultrasound, and MRI (nuclear magnetic resonance method) are often used for diagnosis of breast cancer and the like. In recent years, a photoacoustic wave imaging device that propagates pulsed light into a living body, detects photoacoustic waves generated by absorption of the propagating light on the surface of the living body, and images the initially generated pressure distribution or light absorption coefficient value distribution in the living body Has attracted attention (Patent Document 1).

  In general, the advantages of photoacoustic wave imaging (photoacoustic wave tomography) include the following.

  First, it is possible to perform functional imaging such as oxygen metabolism and brain activity, instead of imaging internal forms such as X-rays, ultrasound, and MRI. For example, oxygenated hemoglobin and deoxygenated hemoglobin have different light absorption spectra. Therefore, by measuring the absorption spectrum, oxygen saturation in the blood can be obtained, and oxygen metabolism measurement can be performed. . If oxygen metabolism can be imaged, there are advantages such as the ability to examine cancerous tumors and brain active sites.

  Secondly, since light is not exposed compared to radiation such as gamma rays or X-rays, it can be diagnosed repeatedly and continuously without being invasive.

  Thirdly, it is possible to reduce the size and price of medical devices as compared with MRI and PET (positron emission tomography).

  Fourth, compared with a light diffusion imaging device that detects light diffusion and performs in vivo imaging, the photoacoustic wave is less scattered in the living body, so that the resolution can be increased relatively. There is.

In order to acquire an image of a light absorber in a living body, it is necessary to reconstruct the image from the received photoacoustic wave signal. As a method of this image reconstruction, there are various methods such as a time domain method and a Fourier domain method. A technique has been proposed (Non-Patent Document 1).
Jonathan I. Sperl et al., “Photoacoustic Image Reconstruction-A Quantitative Analysis”, SPIE-OSA Vol. 6631, 663103, 2007 US Pat. No. 5,713,356

  In photoacoustic wave imaging, the number of photons reaching the deep part of the living body is greatly reduced due to light diffusion in the living body. Therefore, when the light absorber exists in the deep part of the living body, it becomes difficult to observe the photoacoustic wave signal with the sound wave detector disposed on the surface of the living body. Therefore, even if the sound wave detector is installed in all directions of 360 degrees with respect to the human body, the part where the signal can be acquired is limited, and the uncertainty of the acquired information is increased. .

In addition, depending on the measurement target part, the sound wave detector may not be installed in all directions of 360 degrees. When there is such a restriction, for example, a sound wave detector is arranged on one side of the human body and a signal is measured. However, the greater the distance between the light absorber and the sound wave detector, the smaller the solid angle formed by the sound wave detector and the light absorber, which also increases the uncertainty of the information to be acquired. become.

  Thus, when the uncertainty of the information to be acquired increases, a virtual image (artifact) that does not actually exist is generated in the image reconstruction process, and there is a problem that the diagnostic accuracy is lowered.

  Therefore, an object of the present invention is to provide a biological information imaging apparatus and an image construction method that can suppress the occurrence of artifacts as much as possible even when the uncertainty of information to be acquired increases.

  In order to achieve the above object, the present invention adopts the following configuration.

  A first aspect of the present invention includes a light source that irradiates light to a living body, an ultrasonic detector that receives an ultrasonic wave generated by light absorption and outputs a signal, and a time domain from a signal output from the ultrasonic detector A time domain processing unit for reconstructing a first image by a method, a Fourier domain processing unit for reconstructing a second image by a Fourier domain method from a signal output from the ultrasonic detector, and the first image And an image calculation unit that generates a diagnostic image from the second image.

  A second aspect of the present invention is an image construction method in a biological information imaging apparatus that irradiates a living body with light, receives ultrasonic waves generated by light absorption, and images them, and obtains the ultrasonic waves. Reconstructing a first image from a signal obtained by a time domain method, reconstructing a second image from a signal obtained by receiving the ultrasonic wave by a Fourier domain method, and the first image And a step of generating a diagnostic image from the second image.

  According to the present invention, the occurrence of artifacts can be suppressed as much as possible even when the uncertainty of information to be acquired increases, such as when sound waves cannot be detected from all directions of 360 degrees with respect to the human body.

  Next, embodiments of the present invention will be described with reference to the drawings. First, an image reconstruction method used in the embodiment of the present invention will be described. Hereinafter, in order to facilitate the description, the outline will be described in the case of two dimensions (the case of three dimensions can be considered similarly).

  The image reconstruction method can be roughly classified into a time domain method and a Fourier domain method.

The time domain method is a method of drawing sound wave source candidates in real space by utilizing the property that a sound wave is a spherical wave. Typical techniques include DnS (Delay and Sum) method, CBP (Circular Back Projection) method, and HTA (Hough Transform Algorithm) method.

In the image reconstruction by the DnS method, the fact that the trajectory of the time series data forms a quadratic curve when a spherical wave is observed on one side is used. For example, as shown in FIG. 4A, the point sound source signal draws a quadratic curve as indicated by reference numeral 101 in the space-time in which the horizontal axis is the coordinate of the side where the signal is observed and the vertical axis is the time. Become.

  Therefore, if the data on a curve similar to the quadratic curve is added to the vertex of the quadratic curve in the space-time, a sound wave source candidate is drawn.

  Finally, if the time space is converted into the real space using the sound speed (z = ct, c: sound speed), the sound wave source 102 is drawn as shown in FIG. 4B.

  Examples of image reconstruction using the DnS method are shown in FIGS. 4C and 4D. In FIG. 4C, there are two circular sound wave sources. When they are reconstructed, a reconstructed image as shown in FIG. 4D is obtained.

  Thus, the DnS method tends to generate an artifact 103 perpendicular to the measurement side in the vicinity of the measurement side due to the characteristic that the signals on the quadratic curve are added.

  Further, in image reconstruction by the CBP method, which is one of the time domain methods, as shown in FIG. 5A, a radius ct (c: sound speed, t : Elapsed time after light irradiation). As a result, a portion having a high possibility as a sound wave source is drawn.

  FIG. 5B shows an example of image reconstruction using the CBP method. As the sound wave source, the same data of FIG. 4C as in the case of the DnS method is used. In the case of using this CBP method, due to the characteristic of the method of drawing an arc, as shown in FIG. 5B, there is a tendency that an artifact 103 in the direction perpendicular to the side where the signal is measured is generated as in the DnS method. .

  Furthermore, image reconstruction by the HTA method, which is one of the time domain methods, is essentially the same as the CBP method except that the space in which the sound wave source is first drawn is a time space. Therefore, as in the CBP method, there is a tendency that artifacts in the direction perpendicular to the side where the signal is measured are generated.

  On the other hand, in the image reconstruction by the Fourier domain method (FTA method), the measured spatio-temporal information (FIG. 6A) is converted into real space information (FIG. 6B) using the Fourier transform and the dispersion relation of sound waves. Take advantage of what you can do. That is, a forward Fourier transform is first performed on the time series of measured sound waves to obtain a spectrum on a time scale. Then, after converting the time / space scale from the sound wave dispersion relationship (ω = ck, ω: time frequency, k: spatial frequency), the spatial distribution of the sound source is obtained as an image by performing the inverse Fourier transform. FIG. 6C shows an example of image reconstruction using the FTA method. As the sound wave source, the same data of FIG. 4C as in the case of the DnS method is used. In an image obtained when this FTA method is used, there is a tendency that artifacts 104 in the parallel direction as shown in FIG. 6C are generated.

As described above, the time domain method tends to produce artifacts in the vertical direction with respect to the signal measurement surface, whereas the Fourier domain method tends to produce artifacts in the parallel direction. The present inventors paid attention to such a difference in tendency and obtained the idea of reducing artifacts by combining the results obtained by the two image reconstruction methods. That is, by using not only the Fourier domain method but also the time domain method, an image with fewer artifacts than an image using only the Fourier domain method can be obtained. Specifically, in the biological information imaging apparatus of the present invention, the first image reconstructed by the time domain method and the second image reconstructed by the Fourier domain method are arithmetically processed (synthesized), Acquire diagnostic images with few artifacts.

  In general, the image reconstruction processing by the time domain method takes longer processing time than the image reconstruction processing by the Fourier domain method. Therefore, it is preferable to reduce the number of data points of a signal used for processing by the time domain method (than the number of data points used for processing by the Fourier domain method). Thereby, the processing speed of the time domain method can be improved. The image obtained by the time domain method is mainly used for the purpose of reducing artifacts of the image obtained by the Fourier domain method.

[Embodiment]
An overview of a biological information imaging apparatus according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a diagram illustrating a configuration example of a biological information imaging apparatus according to the present embodiment.

  The biological information imaging apparatus according to the present embodiment images information on the distribution of optical characteristic values in a living body, in particular information on the light absorption coefficient value distribution, and information on the concentration distribution of substances constituting the biological tissue obtained from the information. It is possible. This apparatus is used, for example, for diagnosis of malignant tumors, vascular diseases, etc., and for follow-up of chemotherapy. In particular, this apparatus can be preferably applied to an apparatus for obtaining a diagnostic image of breast cancer or the like.

  The biological information imaging apparatus of the present embodiment generally includes a light source 4, an ultrasonic detector 8, a signal processing device 6, and an image display device 5.

The light source 4 is a means for generating pulsed light 2 for irradiating the living body 1 as a subject. Here, the living body means any place of the human body such as breast, finger, foot or neck. Usually, the pulsed light generated from the light source 4 is applied to the surface of the living body through the light propagation device 3 such as an optical fiber or a liquid light guide. However, when the energy of the pulsed light exceeds the limit of the optical propagation path such as an optical fiber, the living body can be irradiated with the pulsed light using a mirror or a lens. In order to reduce the irradiation energy per unit area on the surface of the living body, it is preferable to spread the pulsed light rather than condense it. Typically, it is about several cm ² , but the area is designed to an optimum value depending on the energy of the pulsed light and the light irradiation limit to the human body.

  The ultrasonic detector 8 is means for detecting an ultrasonic wave 7 generated by the light absorber 9 in the living body absorbing a part of the energy of light and converting it into an electric signal. The light absorber 9 is, for example, a tumor, a blood vessel, or a tissue similar to these. It is preferable to use an ultrasonic detector 8 having a plurality of ultrasonic receiving elements. In that case, the ultrasonic receiving elements may be arranged in a two-dimensional array. The signal processing device 6 is means for analyzing an electrical signal obtained by the ultrasonic detector 8 and performing image reconstruction processing. The signal processing device 6 includes a time domain processing unit 61, a Fourier domain processing unit 62, and an image calculation unit 63. The image display device 5 is means for displaying an image based on the processing signal of the signal processing device 6.

(light source)
The light source 4 irradiates light having a specific wavelength that is absorbed by a specific component among the components constituting the living body. The light source 4 is preferably a pulsed light source capable of generating pulsed light on the order of several nanometers to several hundred nanoseconds. A laser is preferable as the light source, but a light emitting diode or the like may be used instead of the laser. As the laser, various lasers such as a solid laser, a gas laser, a dye laser, and a semiconductor laser can be used. In order to measure the difference of the optical characteristic value distribution, particularly the light absorption coefficient value distribution, depending on the wavelength, a light source capable of oscillating light of different wavelengths may be used instead of oscillating only light of a single wavelength. As a light source at that time, a dye that can change an oscillation wavelength, a laser using an optical crystal such as OPO (Optical Parametric Oscillators), titanium sapphire, and alexandrite can be used. Regarding the wavelength to be used, a region of 700 nm to 1100 nm, which is less absorbed in the living body, is preferable. When obtaining the optical characteristic value distribution of the living tissue relatively near the surface of the living body, it is also possible to use a wavelength region having a wider range than the above wavelength region, for example, a wavelength region of 400 nm to 1600 nm.

(Ultrasonic detector)
The ultrasonic detector 8 needs to detect the ultrasonic wave 7 generated by the light absorber 9 that has absorbed a part of the energy of the light propagated in the living body and convert it into an electric signal. Therefore, it is desirable that the frequency band that can be received by the ultrasonic detector 8 is optimized according to the size of the light absorber 9 in the living body. The ultrasonic detector 8 can detect any sound wave as long as it can detect an acoustic wave signal, such as a transducer using a piezoelectric phenomenon, a transducer using optical resonance, or a transducer using a change in capacitance. A vessel may be used. When receiving ultrasonic waves generated from light absorbers of various sizes, a transducer using a change in capacitance with a wide detection frequency band is preferable.

  In the present embodiment, an example is shown in which one ultrasonic detector 8 is arranged in the vicinity of the surface of the living body. However, the present invention is not limited to such an arrangement, and it is only necessary that ultrasonic waves can be detected at a plurality of locations. . That is, since the same effect can be obtained if sound waves can be detected at a plurality of locations, one sound wave detector may be scanned two-dimensionally on the surface of the living body, and ultrasonic receiving elements are arranged one-dimensionally or two-dimensionally. The array type may be arranged at a plurality of locations. The two-dimensional array of ultrasonic detectors can detect ultrasonic signals at a plurality of positions at the same time, and therefore has an advantage that the measurement time can be shortened as compared with a scan type detector.

  When the electrical signal obtained from the ultrasonic detector 8 is small, it is preferable to amplify the signal intensity using an amplifier. Moreover, it is desirable to use an acoustic impedance matching agent for suppressing reflection of ultrasonic waves between the ultrasonic detector and the biological material to be measured.

(Signal processing device)
The signal processing device 6 can construct an image from the electric signal output from the ultrasonic detector 8, and can thereby derive information on the optical characteristic value distribution of the living body (particularly information on the light absorption characteristic value distribution).

  As shown in FIG. 1, the signal processing device 6 according to the present embodiment includes a time domain processing unit 61 that performs image reconstruction processing by the time domain method, and a Fourier domain processing unit that performs image reconstruction processing by the Fourier domain method. 62. The signal processing device 6 configures an image with each of the time domain processing unit 61 and the Fourier domain processing unit 62 based on the electrical signal obtained from the ultrasonic detector 8. Then, the image calculation unit 63 calculates the images obtained from both the processing units 61 and 62 and outputs the result to the image display device 5.

  The signal processing device 6 refers to at least a storage unit that stores the intensity of the ultrasonic wave and its temporal change, and refers to the storage unit to convert the signal obtained from the ultrasonic detector 8 into data related to the optical characteristic value. It is realizable with the apparatus which comprises the calculating part to convert. For example, the signal processing device 6 can be configured by a computer and software that can analyze data, hardware that can analyze data, or both.

(Image display device)
The image display device 5 only needs to display an image obtained by analyzing the ultrasonic signal, and various devices such as a liquid crystal display can be used.

In addition, when irradiating a living body with light of a plurality of wavelengths, information related to the optical characteristic value distribution in the living body, for example, the light absorption coefficient value distribution can be calculated for each wavelength. It is also possible to image the concentration distribution of substances constituting the living body by comparing these values with the wavelength dependence inherent in the substances constituting the living tissue (glucose, collagen, oxidized / reduced hemoglobin, etc.). .

  By using the biological information imaging apparatus shown in such an embodiment, it is possible to obtain a diagnostic image with fewer artifacts than in the prior art, and to perform a highly reliable diagnosis. .

(Image construction method)
Next, an image construction method by the biological information imaging apparatus of this embodiment will be described with reference to FIG. FIG. 3 is a flowchart of a process for imaging information related to a light absorption coefficient value distribution that is an optical characteristic value distribution in a living body. The following (S1) to (S5) correspond to the processes S1 to S5 of the flowchart shown in FIG. Each process will be described in the order of this number.

  (S1) The light source 4 irradiates the surface of the living body 1 with the pulsed light 2.

  (S2) The light absorber 9 in the living body absorbs part of the energy of the pulsed light and generates an ultrasonic wave. This ultrasonic signal is detected by the ultrasonic detector 8. Normally, an ultrasonic signal generated by a spherical light absorber shows an N-shaped waveform as shown in FIG. 2 when time is taken on the horizontal axis and intensity of the ultrasonic wave is taken on the vertical axis.

  (S3) The time domain processing unit 61 reconstructs an image from the signal obtained in (S2) by the time domain method described above. As the time domain method, the above-described DnS method, CBP method, HTA method, or the like can be selected.

  (S4) Similarly, the Fourier domain processing unit 62 reconstructs an image from the signal obtained in (S2) by the Fourier domain method described above.

  (S5) Next, the image computing unit 63 computes both an image obtained by the time domain method and an image obtained by the Fourier domain method, thereby outputting a final diagnostic image. As the calculation of the image, for example, sum (SUM), product (PRODUCT) and the like can be considered.

  7A to 7E are diagrams illustrating examples of images. FIG. 7A is a diagram schematically showing two circular sound wave sources (light absorbers) existing in a living body. FIG. 7B is an image configured using the DnS method as the time domain method, and FIG. 7C is an image configured using the Fourier domain method. As can be seen from FIG. 7B and FIG. 7C, there is a tendency that generated artifacts are different between an image obtained by the time domain method and an image obtained by the Fourier domain method. Thus, artifacts can be effectively eliminated or reduced by calculating (combining) two images. FIG. 7D shows the result of the sum operation of two images. It can be seen that the artifacts are less noticeable than the sound source. FIG. 7E shows the result of product operation of two images. It can be seen that the outline of the acoustic wave source is clear and the artifacts are effectively eliminated. The same effect can be obtained even when the CBP method or the HTA method is used as the time domain method. This is because the properties of artifacts generated by the CBP method and the HTA method are similar to those of the DnS method.

In the present embodiment, the example of obtaining the light energy absorption density distribution in the living body or the characteristic value distribution related to the light absorption coefficient value based on the flowchart of FIG. 3 has been described, but the present invention is limited to the processing of this flowchart. Is not to be done. The essence of the present invention is to generate a plurality of images from an ultrasonic signal using a plurality of image reconstruction methods having different artifact properties, and to reduce the artifacts by combining the plurality of images. That is, in this embodiment, the final diagnostic image is composed of two images, but the diagnostic image may be composed of three or more images. Further, the calculation of the image is not limited to the sum and product, and other calculation processing may be applied.

(Example)
An embodiment of the present invention will be described. An example of a biological information imaging apparatus for the purpose of obtaining an image representing information related to the distribution of light absorption coefficient values in a living body will be described. Here, an experimental example in which a phantom simulating a living body is imaged is shown.

  A Q-switched Nd: YAG laser capable of oscillating 1064 nm nanosecond pulsed light was used as the light source. The pulse width is about 5 nanoseconds and the repetition rate is 10 Hz. For example, the energy of one pulse is 120 mJ. This pulsed light is guided to the surface of the living body using a dielectric reflecting mirror. In addition, the phantom used here embeds a rubber tube having a diameter of 1 mm as a light absorber in an intralipid solidified with agar.

  The ultrasonic signal measured by the ultrasonic detector after the light irradiation is sent to a signal processing device constituted by a computer. The signal processing apparatus reconstructs two images using a time domain method and a Fourier domain method, and outputs a product of the two images as a diagnostic image. Here, the DnS method was used as the time domain method.

  Using such a device, when the light absorption energy density distribution of the entire phantom is imaged, it is possible to obtain an image of a light absorber with fewer artifacts than when the Fourier domain method is used alone. Became.

FIG. 1 is a diagram illustrating a configuration example of a biological information imaging apparatus according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of an ultrasonic signal detected in the biological information imaging apparatus. FIG. 3 is a flowchart showing an example of an image construction method by the biological information imaging apparatus. 4A to 4D are diagrams illustrating an outline of an image configuration by the time domain method (DnS method). FIG. 5A and FIG. 5B are diagrams for explaining the outline of the image configuration by the time domain method (CBP method). 6A to 6C are diagrams illustrating an outline of an image configuration by the Fourier domain method. 7A to 7E are diagrams showing the effects of the present invention.

Explanation of symbols

1: Living body 2: Pulse light 3: Light propagation device 4: Light source 5: Image display device 6: Signal processing device 7: Ultrasonic wave generated from light absorber 8: Ultrasonic detector 9: Light absorber 61: Time domain Processing unit 62: Fourier domain processing unit 63: Image calculation unit 101: Point sound source signal 102: Sound wave source 103, 104: Artifact

Claims (12)

A light source for irradiating a living body with light;
An ultrasonic detector that receives ultrasonic waves generated by light absorption and outputs signals;
A time domain processing unit for reconstructing a first image from a signal output from the ultrasonic detector by a time domain method;
A Fourier domain processing unit for reconstructing a second image from a signal output from the ultrasonic detector by a Fourier domain method;
An image calculation unit that generates a diagnostic image from the first image and the second image;
A biological information imaging apparatus comprising:   The time domain processing unit reconstructs the first image from a signal having a smaller number of data points than a signal used by the Fourier domain processing unit to reconstruct the second image. Item 2. The biological information imaging apparatus according to Item 1.   The biological information imaging apparatus according to claim 1, wherein the image calculation unit generates the diagnostic image by taking a sum or a product of the first image and the second image. .   The biological information imaging apparatus according to claim 1, wherein the time domain processing unit reconstructs the first image by a DnS method.   The biological information imaging apparatus according to claim 1, wherein the time domain processing unit reconstructs the first image by a CBP method.   The biological information imaging apparatus according to claim 1, wherein the time domain processing unit reconstructs the first image by an HTA method. An image construction method in a biological information imaging apparatus for irradiating a living body with light, receiving ultrasonic waves generated by light absorption, and imaging it,
Reconstructing a first image from a signal obtained by receiving the ultrasonic wave by a time domain method;
Reconstructing a second image from a signal obtained by receiving the ultrasonic wave by a Fourier domain method;
Generating a diagnostic image from the first image and the second image;
An image construction method comprising:   In the step of reconstructing the first image, the first image is reconstructed by the time domain method from a signal having a smaller number of data points than the signal used to reconstruct the second image. The image construction method according to claim 7.   The said diagnostic image is produced | generated in the process of producing | generating the said diagnostic image by taking the sum or product of the said 1st image and the said 2nd image. Image construction method.   The image constructing method according to claim 7, wherein in the step of reconstructing the first image, the first image is reconstructed by a DnS method.   The image constructing method according to claim 7, wherein in the step of reconstructing the first image, the first image is reconstructed by a CBP method.   The image constructing method according to claim 7, wherein in the step of reconstructing the first image, the first image is reconstructed by an HTA method.

Images