JP2012137405A - Diagnosis support device and method, and device and method for detecting lesion part - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a diagnosis support device which, when an area having fibrillation suspected of having a lesion part in progress exists in an in vivo inner wall part having a layer structure when the biliary tract and pancreatic duct or the like are normal, performs diagnostic support by automatically detecting such an area.SOLUTION: Based on light intensity data obtained by the optical coherence tomographic measurement, an arithmetic processing unit 90 of an OCT device 1 adds the light intensity data within a range of a predetermined data in the same depth relating to the depth direction of the inner wall part from a measurement point set on the surface of the inner wall part, and when the added value shows a predetermined change along the depth direction, the measurement point is detected as the lesion part and displayed.

Description

  The present invention relates to a diagnosis support apparatus, a diagnosis support method, a lesion detection apparatus, and a lesion detection method, and particularly uses light intensity data acquired by an OCT apparatus using optical coherence tomography (OCT) measurement. The present invention relates to a diagnosis support apparatus, a diagnosis support method, a lesion detection apparatus, and a lesion detection method that perform diagnosis support.

  Conventionally, an optical tomographic image of a living tissue is acquired using an optical tomographic image acquisition apparatus (OCT apparatus) using OCT measurement. This OCT apparatus divides low-coherent light emitted from a light source into measurement light and reference light, and then combines backscattered light and reference light from the measurement object when the measurement light is irradiated onto the measurement object. An optical tomographic image is acquired based on the intensity of the interference light between the backscattered light and the reference light (Patent Document 1). The OCT measurement is roughly classified into two types: TD-OCT (Time Domain OCT) measurement and FD-OCT (Fourier Domain OCT) measurement. TD-OCT measurement acquires the backscattered light intensity distribution corresponding to the position in the depth direction of the measurement target (hereinafter referred to as the depth position) by measuring the interference light intensity while changing the optical path length of the reference light. It is a method to do.

  On the other hand, in the FD-OCT measurement, the interference light intensity is measured for each spectral component of the light without changing the optical path lengths of the reference light and the signal light, and the spectral interference intensity signal obtained here is Fourier transformed by a computer. This is a method of obtaining a reflected light intensity distribution corresponding to a depth position by performing a representative frequency analysis. In recent years, it has attracted attention as a technique that enables high-speed measurement by eliminating the need for mechanical scanning existing in TD-OCT.

  As a typical apparatus configuration for performing FD-OCT measurement, there are two types, that is, an SD-OCT (Spectral Domain OCT) apparatus and an SS-OCT (Swept Source OCT) apparatus. The SD-OCT apparatus uses broadband low-coherent light such as SLD (Super Luminescence Diode), ASE (Amplified Spontaneous Emission) light source, white light as a light source, and uses Michelson interferometer to generate broadband low-coherent light. After the measurement light and the reference light are divided, the measurement light is irradiated onto the measurement object, the backscattered light returning at that time and the reference light are caused to interfere, and this interference light is separated into each frequency component using a spectrometer. By disassembling and measuring the interference light intensity for each frequency component using a detector array in which elements such as photodiodes are arranged in an array, the resulting spectrum interference intensity signal is Fourier transformed by a computer, An optical tomographic image is constructed.

  Such an OCT apparatus can acquire the three-dimensional structure information (referred to as three-dimensional structure information) of the measurement object by two-dimensionally scanning the optical axis of the measurement light. Conventionally, three-dimensional structure information of a part in a body cavity of a human body is acquired by an OCT apparatus, and the three-dimensional structure information is imaged (visualized) to display a three-dimensional image on a monitor, or three-dimensional structure information is analyzed and a lesion portion is analyzed. It has been proposed to support image diagnosis by automatically detecting (see, for example, Patent Documents 2 and 3).

  By the way, the biliary tract and the pancreatic duct are known as regions having a very low cure rate when cancer is caused. For diagnosis of biliary tract and pancreatic duct, first, abdominal ultrasound and blood tests are performed as primary screening, X-ray CT (Computed Tomography) as secondary screening, MRCP (Magnetic Resonance Cholangiopancreatography) imaging diagnosis, endoscopic cholangiography ( Modalities (medical devices) such as ERCP (Endoscopic Retrograde Cholangio-pancreatography), endoscopic ultrasound EUS (Endoscopic Ultrasoundscopy), and intraluminal ultrasonography (IDUS) are used to identify the location of cancer, Diagnosis of depth and progress is being made. When it is determined that surgical resection is effective as a result of comprehensive judgment, a region to be resected is determined and surgically resected. The excised tissue is subjected to a histopathological search, and a diagnosis is made as to whether the cancer has spread to the resected margin. If cancer remains at the resection stump, it is highly presumed that there is leftover cancer and additional resection or radiation chemotherapy is required.

JP 2007-225349 A JP 2010-145297 A JP 2010-158343 A

  However, abdominal ultrasound and endoscopic ultrasound EUS are not suitable for observing changes in the shape of the tube wall. In X-ray CT and ERCP, the shape of the tube wall is imaged, and if there is a large tumor of 1 mm or more, it can be diagnosed with these, but it is difficult to discern changes below that. IDUS is a method of inserting a probe directly into a tube, and is suitable for observation of a tube wall, but has a resolution of about 100 μm. On the other hand, the morphological change of cancer that has spread laterally in the epithelium of the bile duct or pancreatic duct may be about 20 μm, and the resolution is insufficient to observe it.

  For this reason, the accuracy of lateral development is poor with the conventional diagnostic method, and there are cases in which the surgical margin is positive, that is, the cancer remains uncut by the histopathological diagnosis after surgical resection. Things were a problem. Therefore, highly sensitive diagnosis of the lateral development of cancer and determination of highly accurate surgical resection line are important issues.

  On the other hand, the OCT device has high resolution and is effective for image diagnosis of the biliary tract and pancreatic duct, and automatic diagnosis that automatically detects the lesion by analyzing the three-dimensional structure information of the biliary tract and pancreatic duct is also useful as a diagnostic support for the doctor. It is considered a thing. However, at present, no automatic diagnostic attempt has been made to distinguish between normal and abnormal sites in the biliary tract and pancreatic duct. Patent Documents 2 and 3 propose automatic diagnostic trials, but are mainly intended for the large intestine and do not detect morphological changes in lesions peculiar to the biliary tract and pancreatic duct.

  Here, in the inner wall of the biliary tract / pancreatic duct, it is known that fibrosis has progressed as a morphological change (characteristic structure) suspected of being cancerous. Other suspected morphological changes such as cancer are known, but if at least the above morphological changes can be detected automatically, it is useful for diagnosis support of the biliary tract and pancreatic duct.

  Regardless of biliary tract or pancreatic duct, the same morphological changes due to cancer, etc. can be automatically detected in organs having a layer structure under normal conditions such as bronchi, pharynx, esophagus, stomach, large intestine, uterus and ureter. If it is possible, it will be useful for diagnosis support.

  The present invention has been made in view of such circumstances, and in the inner wall portion of a living body having a layer structure at normal time such as a biliary tract and a pancreatic duct, a region where fibrosis is in progress is suspected of being a lesion. Is provided, a diagnosis support apparatus, a diagnosis support method, a lesion detection apparatus, and a lesion detection method that automatically detect the region and provide diagnosis support.

  In order to achieve the object, the diagnosis support apparatus according to claim 1 acquires light intensity data for acquiring light intensity data of the inner wall obtained by optical coherence tomography with respect to the inner wall of the living body having a layer structure. And a predetermined range of light intensity data at the same depth with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall based on the light intensity data acquired by the light intensity data acquiring means An addition means for adding the detection point, a detection means for detecting the measurement point as a lesion when the addition value calculated by the addition means shows a predetermined change along the depth direction, and the light intensity data And a lesion part display means for displaying information indicating the lesion part detected by the detection means on the image visualized.

  According to the present invention, when there is a “region in which fibrosis is progressing” in the inner wall portion of a living body having a layer structure at a normal time such as a biliary tract or a pancreatic duct, there is a suspected lesion such as cancer. The layer structure of the inner wall portion disappears, and as a result, an added value obtained by adding the light intensity data of a predetermined range existing at the same depth position as the reference position with each position in the depth direction of the inner wall portion as a reference position. Since it shows a certain amount of attenuation along the depth direction, it is possible for the observer to easily grasp the suspected area of the lesion by automatically detecting and displaying such an area. .

  According to a second aspect of the present invention, in the diagnosis support apparatus according to the first aspect of the invention, the detection means is within a range in which the addition value calculated by the addition means can be regarded as being attenuated uniformly along the depth direction. In some cases, the measurement point is detected as a lesion.

  According to the present invention, it is possible to prevent omission of detection of a region where the layer structure has disappeared.

  According to a third aspect of the present invention, there is provided the diagnosis support apparatus according to the first or second aspect, wherein the detection means includes a measurement point at which the addition value calculated by the addition means indicates a predetermined change along the depth direction. In the case where a region having a predetermined size is detected and the set of measurement points is continuously formed, the region is detected as a lesioned part.

  In the present invention, when the region where the added value exhibits constant attenuation is limited to a very small range, the region does not correspond to the “region in which fiberization is progressing”, and is therefore excluded.

  According to a fourth aspect of the present invention, in the diagnosis support apparatus according to any one of the first to third aspects, the lesioned part display unit is arranged in a predetermined area on the lesioned part on the image obtained by visualizing the light intensity data. It is characterized by being displayed with a color.

  According to the present invention, it becomes possible for an observer to grasp the detected lesion area clearly.

The diagnosis support apparatus according to claim 5 is the invention according to any one of claims 1 to 4,
The light intensity data acquisition means acquires light intensity data of a three-dimensional area of the inner wall.

  In a preferred embodiment of the present invention, the light intensity data acquisition means preferably acquires light intensity data of a three-dimensional area of the inner wall portion.

  The diagnosis support apparatus according to a sixth aspect is the invention according to any one of the first to fifth aspects, wherein the lesion display means is a three-dimensional image obtained by a perspective projection process as an image obtained by visualizing the light intensity data. At least one of a three-dimensional image obtained by parallel projection processing and a tomographic image at a predetermined cross section is displayed.

  Since the present invention acquires light intensity data of a three-dimensional area of the inner wall, various forms of images can be generated, and a three-dimensional image to be displayed in three dimensions or a tomographic image of only a specific cross section can be displayed. Indicates that it is possible. In any image, the lesion can be displayed.

  The diagnosis support apparatus according to claim 7 is the invention according to any one of claims 1 to 6, wherein the inner wall portion has a biliary tract, pancreatic duct, bronchi, pharynx, esophagus, stomach, large intestine, uterus, or ureter It is characterized by being the inner wall part.

  The invention according to claims 1 to 6 limits the part in the living body that is particularly effective.

  In order to achieve the above object, a diagnostic support method according to claim 8 is the light intensity data for acquiring the light intensity data of the inner wall obtained by optical coherence tomography with respect to the inner wall of the living body having a layer structure. Based on the light intensity data acquired by the acquisition step and the light intensity data acquisition step, the light intensity in a predetermined range at the same depth with respect to the depth direction of the inner wall portion from the measurement point set on the surface of the inner wall portion An addition step of adding data, a detection step of detecting the measurement point as a lesion when the addition value calculated in the addition step shows a predetermined change along the depth direction, and the light intensity A lesion portion display step of displaying information indicating the lesion portion detected by the detection step on an image in which data is visualized.

  In order to achieve the above object, a lesion detection apparatus according to claim 9 is a light intensity for obtaining light intensity data of an inner wall obtained by optical coherence tomography with respect to an inner wall of a living body having a layer structure. Based on the light intensity data acquired by the data acquisition means and the light intensity data acquisition means, a predetermined range of light at the same depth with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall Addition means for adding intensity data; and detection means for detecting the measurement point as a lesion when the addition value calculated by the addition means shows a predetermined change along the depth direction. It is characterized by that.

  In order to achieve the above object, a lesion detection method according to claim 10 is the light intensity for acquiring the light intensity data of the inner wall obtained by optical coherence tomography measurement on the inner wall of the living body having a layer structure. Based on the light intensity data acquired in the data acquisition step and the light intensity data acquisition step, the light within a predetermined range at the same depth with respect to the depth direction of the inner wall portion from the measurement point set on the surface of the inner wall portion An addition step of adding intensity data; and a detection step of detecting the measurement point as a lesion when the addition value calculated in the addition step shows a predetermined change along the depth direction. It is characterized by that.

  According to the present invention, when there is a `` region where fibrosis is progressing '' in which there is a suspicion of a lesion in an inner wall portion of a living body having a layer structure at a normal time such as a biliary tract or a pancreatic duct, the region is It can be detected automatically and can provide diagnosis support.

Configuration diagram showing the overall configuration of the OCT apparatus Medical network diagram to which the OCT device is connected Sectional drawing which shows the cross section containing the long axis of an OCT probe Explanatory drawing used to describe the operation when placing the OCT probe in the biliary tract and pancreatic duct Explanatory drawing used to describe the operation when placing the OCT probe in the biliary tract and pancreatic duct The figure which showed the aspect which has arrange | positioned the balloon to the OCT probe Diagram showing normal layer structure of the inner wall of the biliary tract and pancreatic duct Explanatory drawing used to explain lesion detection processing type 1 Explanatory drawing used for description of lesion detection processing type 2 Explanatory drawing used to explain lesion detection processing type 3 Explanatory drawing used to explain lesion detection type 4 Explanatory drawing used to explain lesion detection processing type 5 Explanatory drawing used for description of lesion detection processing type 6 Explanatory drawing used for description of lesion detection processing type 7 Explanatory drawing used to explain lesion detection processing type 8 Explanatory drawing used to explain lesion detection processing type 9 Explanatory drawing used for explanation of detection processing of contact area and non-contact area of sheath Flow chart showing the processing procedure of the first embodiment of the diagnosis support function Diagram showing the configuration of the display screen displayed on the monitor in the diagnosis support function Diagram showing the configuration of the display screen displayed on the monitor in the diagnosis support function Flow chart showing the processing procedure of the second embodiment of the diagnosis support function BC tomogram displayed in the second embodiment of the diagnosis support function Explanatory drawing used to explain the display mode of 3D structure data Explanatory drawing used to describe the endoscope mode Explanatory drawing used to explain perspective mode Explanatory drawing used for explanation of long axis section mode Illustration used to explain the surface mode Illustration used to explain the surface mode Flow chart showing processing procedure of detection processing type 1 Flow chart showing processing procedure of detection processing type 2 Flow chart showing processing procedure of detection processing type 3 Flow chart showing processing procedure of detection processing type 4 Flow chart showing processing procedure of detection processing type 5 Flowchart showing processing procedure of detection processing type 6 Flow chart showing processing procedure of detection processing type 7 Flowchart showing processing procedure of detection processing type 8 Flowchart showing processing procedure of detection processing type 9 The flowchart which showed the process sequence in the case of changing the threshold value of a detection process type 2 according to the contact area | region and non-contact area | region of a sheath.

  Hereinafter, the best mode for carrying out the diagnosis support apparatus according to the present invention will be described.

  FIG. 1 is a block diagram showing a configuration of an OCT apparatus used for measuring tomographic information of an inner wall portion of a biliary tract / pancreatic duct to be measured in the present embodiment. As shown in FIG. 1, the OCT apparatus 1 acquires tomographic information to be measured by, for example, SS-OCT measurement centered on a wavelength of 1.3 μm, and displays an OCT image by the acquired tomographic information. The OCT light source 10, the OCT interferometer 30, the arithmetic processing unit (computer) 90, the monitor 100 and the like are configured.

  The OCT light source 10 is a light source that emits laser light L in the infrared region while sweeping the frequency at a constant period.

  The laser light L emitted from the OCT light source 10 is demultiplexed into the measurement light L1 and the reference light L2 by the optical demultiplexing unit 3 in the OCT interferometer 30. The optical demultiplexing unit 3 is composed of, for example, an optical coupler having a branching ratio of 90:10, and demultiplexes at a ratio of measurement light: reference light = 90: 10.

  In the OCT interferometer 30, the reference light L2 demultiplexed by the optical demultiplexing unit 3 is reflected after the optical path length is adjusted by the optical path length adjusting unit 80 via the circulator 5a.

  The optical path length adjustment unit 80 changes the optical path length of the reference light L2 in order to adjust the position at which tomographic image acquisition is started, and includes collimator lenses 81 and 82 and a reflection mirror 83. The reference light L2 from the circulator 5a passes through the collimator lenses 81 and 82 and then is reflected by the reflection mirror 83. The return light L2a of the reference light L2 is incident on the circulator 5a again through the collimator lenses 81 and 82.

  Here, the reflection mirror 83 is disposed on the movable stage 84, and the movable stage 84 is provided so as to be movable in the arrow X direction by the mirror moving unit 85. When the movable stage 84 moves in the direction of arrow A, the optical path length of the reference light L2 is changed. Then, the return light L2a of the reference light L2 from the optical path length adjustment unit 80 is guided to the optical multiplexing / demultiplexing unit 4 via the circulator 5a.

  On the other hand, the measurement light L1 demultiplexed by the optical demultiplexing unit 3 is guided to the light emitting end at the tip of the OCT probe 40 by the optical fiber FB2 via the circulator 5b, the optical fiber FB1, and the optical rotary connector unit 41. The light is emitted from the light emitting end and irradiated onto the measurement target T. The OCT probe 40 guides the backscattered light L4 from the measurement target T when the measurement light L1 is irradiated on the measurement target T.

  The OCT probe 40, which will be described in detail later, has an optical fiber FB2 inserted and disposed in the probe outer cylinder, and the optical fiber FB2 ahead of the optical rotary connector 41 is moved outside the probe by a motor (not shown) in the optical scanning unit 42. It is configured to rotate in the cylinder. Further, the measurement light L1 emitted from the light emission end toward the measurement target T is in a direction deflected with respect to the optical axis of the optical fiber FB2 (substantially perpendicular direction), and the optical fiber FB2 rotates. Accordingly, the emission direction of the measurement light L1 emitted from the light emission end is also rotated around the long axis of the OCT probe 40.

  Here, in the present embodiment, the emission direction of the measurement light L1 emitted from the light emission end of the OCT probe 40 (the direction of the optical axis direction of the measurement light L1 and the depth direction of the measurement target T) is set. The A-axis direction or A-axis, the long-axis direction of the OCT probe 40 (optical-axis direction of the optical fiber FB2) is the C-axis direction or C-axis, and the direction perpendicular to the AC plane composed of the A-axis and C-axis is the B-axis direction or B axis. Further, in the case of the configuration in which the emission direction of the measurement light L1 from the light emission end is rotated around the major axis of the OCT probe 40 as in the OCT probe 40 of the present embodiment, the A axis direction is also around the major axis. In order to rotate, when it is referred to as the A-axis direction, it indicates the radiation direction when viewed from the long axis of the OCT probe 40, and when it is referred to as the B-axis direction, it indicates the circumferential direction around the long axis of the OCT probe 40.

  According to this, when the optical fiber FB2 rotates as described above, the measurement light L1 is scanned in the B-axis direction, so that the tomographic information on the AB plane composed of the A-axis and the B-axis is acquired. Note that scanning in the B-axis direction of the measurement light L1 is referred to as B scanning.

  The light exit end of the OCT probe 40 is moved forward and backward in the C-axis direction together with the optical fiber FB2 by a motor (not shown) in the optical scanning unit 42. As a result, the measurement light L1 is scanned in the C-axis direction, so that tomographic information on the AC plane is acquired. Note that scanning in the C-axis direction of the measurement light L1 is referred to as C scanning.

  By performing this C scan simultaneously or alternately with the B scan, tomographic information of a three-dimensional area in a three-dimensional space composed of the A axis, the B axis, and the C axis is acquired. In this specification, the tomographic information of a three-dimensional area is referred to as three-dimensional structure information or three-dimensional structure data.

  The backscattered light L4 is guided to the OCT interferometer 30, and is guided by the OCT interferometer 30 to the optical multiplexing / demultiplexing unit 4 via the circulator 5b. In the optical multiplexing / demultiplexing unit 4, the backscattered light L4 of the measurement light L1 and the return light L2a of the reference light L2 are combined and emitted to the interference information detecting unit 70 side.

  The interference information detection unit 70 detects the interference light L5 between the return light L4 of the measurement light L1 and the return light L2a of the reference light L2 multiplexed by the optical multiplexing / demultiplexing unit 4 at a predetermined sampling frequency. InGaAs photodetectors 71a and 71b that measure the light intensity of the interference light L5, and an interference light detector 72 that performs a balance detection of the detection value of the InGaAs photodetector 71a and the detection value of the InGaAs photodetector 71b are provided. The interference light L5 is divided into two by the optical multiplexing / demultiplexing unit 4, detected by the InGaAs photodetectors 71a and 71b, and output to the interference light detection unit 72. The interference light detection unit 72 detects the intensity of the backscattered light L4 at each position in the A-axis direction by Fourier-transforming the interference light L5 in synchronization with the sweep trigger signal of the OCT light source 10.

  The arithmetic processing unit 90 is a component having the same configuration as that of a general-purpose computer main body, and includes a processor, a storage device, a communication interface, and the like (not shown). Further, an input device 94 such as an operation panel, a keyboard, and a mouse and a monitor 100 are connected to the arithmetic processing unit 90. The arithmetic processing device 90 may be integrated with the OCT light source 10 and the OCT interferometer 30 in the housing of the OCT apparatus 1 main body, or may be connected to the OCT apparatus 1 main body with a cable or the like. It may be a connected computer.

  The arithmetic processing device 90 is a control function that comprehensively controls each part of the OCT apparatus 1 such as the OCT light source 10, the mirror moving unit 85, the optical scanning unit 42, and the intensity of the backscattered light L4 detected by the interference light detecting unit 72. A display (image generation) function for acquiring information on the measurement object T, generating an OCT image such as a tomographic image of the measurement target T and displaying it on the monitor 100, a diagnosis support function for supporting diagnosis performed in combination with the display function, and the like. ing. Necessary operations for implementing these functions are performed using the input device 92, and necessary information is displayed on the monitor 100. Note that the processing unit 90 may also perform processing such as Fourier transform for detecting the intensity of the backscattered light L4 in the interference light detection unit 72.

  In addition, the arithmetic processing unit 90 includes an X-ray fluoroscopic apparatus 110 having a communication function as shown in FIG. 2 and other arbitrary medical devices (endoscope, CT, MRI, EUS, IDUS, MRI, THz, etc.) can communicate with each other. It is also connected to the database 112 via a communication line.

  The database 112 is a filing system that manages data and the like of OCT images generated by the arithmetic processing unit 90. For example, the database 112 manages a large-capacity storage device that stores data and the data stored in the large-capacity storage device. And database software. The arithmetic processing unit 90 can appropriately acquire data stored in the database 112, and can appropriately acquire data such as OCT images stored in the database in the past and data stored in the database by other medical devices. It can be done. In addition, any terminal such as a personal computer can be connected to the database 112 via a communication line. Data such as an OCT image stored in the database 112 is read out by the terminal, and a diagnosis support in the arithmetic processing unit 90 is performed. Processing such as functions can also be performed by software processing of the terminal.

  The database can include, for example, a filing system for other uses such as an electronic medical record in addition to a filing system for image data acquired by an arbitrary medical device.

  Next, the configuration of the OCT probe 40 in the OCT apparatus 1 used for measuring the inner wall of the biliary tract and pancreatic duct to be measured in the present embodiment will be described with reference to FIG.

  As shown in the figure, the distal end portion of the OCT probe 40 has a probe outer tube (sheath) 44, a cap 46, an optical fiber FB2, a spring 48, a fixing member 50, and an optical lens 52. .

  The sheath 44 is a flexible cylindrical member that covers the entire OCT probe 40, and is made of a material that allows the measurement light L1 and the backscattered light L3 to pass therethrough.

  The cap 46 is provided at the distal end of the sheath 44 and closes the distal end of the sheath 44.

  The cap 46 is formed with a guide wire hole 54 for inserting a guide wire. The guide wire hole 54 has one opening 54A on the side surface of the cap 46 and the other opening on the front surface. 54B. The guide wire is arranged in advance at the measurement site and guides the OCT probe 40 to the position. The guide wire arranged at the measurement position is inserted into the guide wire hole 54, and the OCT probe 40 is moved forward. The OCT probe 40 can be guided to the guide wire and moved to the measurement site. By guiding the OCT probe 40 using the guide wire in this way, the OCT probe 40 can be easily placed in the bile duct or pancreatic duct where it is difficult to directly enter the OCT probe 40.

  The optical fiber FB2 is inserted into the sheath 44 and optically connected to the proximal end side of the OCT probe 40 via the optical fiber FB1 and the optical rotary connector 41 as shown in FIG. By this optical fiber FB2, the measurement light L1 emitted from the optical fiber FB1 is guided to the optical lens 52, and the backscattered light L3 from the measurement target T acquired by the optical lens 52 is guided to the optical fiber FB1. .

  The spring 48 is fixed to the outer periphery of the optical fiber FB2, whereby the rotational force applied in the optical rotary connector 41 on the proximal end side is transmitted to the distal end of the optical fiber FB2.

  The optical lens 52 is fixed to the distal end side of the optical fiber FB2 by the fixing member 50, and is formed in a substantially hemispherical shape. By this optical lens 52, the measurement light L1 emitted from the optical fiber FB2 is deflected in the A-axis direction and condensed at a predetermined position. Further, the backscattered light L3 from the measurement target T is collected by the optical lens 52 and is incident on the optical fiber FB2.

  The point where the optical axis of the optical fiber FB2 and the optical axis of the measuring light L1 deflected by the optical lens 52 intersect corresponds to the light emitting end of the OCT probe 40. Further, as described above, the emission direction of the measurement light L1 emitted from the light emission end of the OCT probe 40 (the direction along the optical axis direction of the measurement light L1 and the depth direction of the measurement target T) is the A axis. Direction or A axis, the long axis direction of the OCT probe 40 (optical axis direction of the optical fiber FB2) is the C axis direction or C axis, and the direction perpendicular to the AC plane composed of the A axis and C axis is the B axis direction or B axis. The emission direction of the measurement light L1 deflected by the optical lens 52 is the A-axis direction. In the optical lens 52 according to the present embodiment, the measurement light L1 is deflected in a direction substantially perpendicular to the C-axis direction, so the A-axis direction is a direction substantially perpendicular to the C-axis direction.

  On the proximal end side of the OCT probe 40, the driving device 56 including the optical rotary connector portion 41 and the optical scanning portion 42 shown in FIG.

  In the driving device 56, the optical rotary connector 41 is disposed in a frame 60 disposed in a housing 58. The optical rotary connector 41 includes an optical connector 61 fixed to the frame 60 and an optical connector 61. And a rotating cylinder 62 supported to be rotatable with respect to the connector portion 61.

  The optical connector 61 and the rotating cylinder 62 are formed with a light guide hole penetrating both, and a required optical system is incorporated in the light guide hole, and the optical connector portion 61 side of the light guide hole The end of the optical fiber FB1 shown in FIG. 1 is fixed to the open end of the optical fiber FB1, and the end of the optical fiber FB2 of the OCT probe 40 is fixed to the open end of the light guide hole on the rotary tube 62 side. As a result, the optical fiber FB1 and the optical fiber FB2 are optically connected, the measurement light L1 emitted from the optical fiber FB1 enters the optical fiber FB2, and the backscattered light L3 emitted from the optical fiber FB2 becomes the optical fiber FB1. It is made to enter.

  A spring 48 fixed to the outer periphery of the optical fiber FB2 is fixed to the rotating cylinder 62 together with the optical fiber FB2, and a sheath 44 is fixed to the housing 58. Accordingly, when the rotary cylinder 62 rotates with respect to the optical connector portion 61, the entire optical fiber FB2 rotates with respect to the optical fiber FB1, and also rotates within the sheath 44 and is fixed to the distal end side of the optical fiber FB2. 52 also rotates within the sheath 44.

  On the other hand, a motor 63 that is a component of the optical scanning unit 42 in FIG. 1 is fixed to the frame 60, and this motor 63 is connected to the rotary cylinder 62 via a gear 64. Accordingly, the rotating cylinder 62 is rotated by driving the motor 63, and the optical fiber FB2 and the optical lens 52 are rotated in the direction of the arrow R2 (B-axis direction) in the sheath 44 as described above. Thereby, the measurement light L1 emitted from the optical lens 52 is scanned (B-scan) in the B-axis direction.

  Further, another motor 65 that is a constituent element of the optical scanning unit 42 in FIG. 1 is fixed to the casing 58, and the ball screw 66 for advancing / retreating movement is rotated by the motor 65. The forward / backward moving ball screw 66 is screwed into a screw hole of a support member 67 provided in the frame 60, and the frame 60 moves forward / backward in the optical axis direction of the optical fiber FB2 when the forward / backward moving ball screw 66 rotates. It has become. Therefore, by driving the motor 65, the optical fiber FB2 and the optical lens 52 are moved forward and backward in the directions of arrows S1 and S2 (C-axis direction) in the sheath 44. Accordingly, the measurement light L1 emitted from the optical lens 52 is scanned (C scan) in the C-axis direction.

  The motors 63 and 65 are driven by a drive signal given from the arithmetic processing device 90 according to an operation of the operator by the input device 92 of FIG. 1, for example, or driven by an operation of an operation unit provided in the drive device 56 itself. The B scanning and the C scanning of the measuring light L1 can be appropriately performed. Although not shown in the figure, the driving device 56 is provided with a position sensor that acquires information about the rotation angle of the optical fiber FB2 and information about the position of the optical fiber FB2 in the C-axis direction. Such information is supplied to the arithmetic processing unit 90. Thereby, the arithmetic processing unit 90 controls the scanning position of the measurement light L1 for the B-scan and C-scan, and measures the intensity of the backscattered light L4 in the A-axis direction acquired from the interference light detector 72. The positions in the B-axis direction and the C-axis direction when being performed can be grasped. However, the method for controlling the scanning position and the method for grasping the positions in the B-axis direction and the C-axis direction are not limited to this.

  In addition, markers (platinum markers) 68A and 68B indicating the C scanning range are attached to two locations on the outer peripheral surface of the sheath 44. As will be described later, in the OCT measurement of the biliary tract and pancreatic duct, an X-ray fluoroscopic image is taken using an X-ray fluoroscopic device. At that time, since the X-ray transmittances of the sheath 44 and the markers 68A and 68B are different (the sheath 44 transmits X-rays and the markers 68A and 68B do not transmit), the positions of the markers 68A and 68B are indicated on the X-ray fluoroscopic image. Can be grasped. On the other hand, since these markers 68A and 68B are attached to the positions of the start point and end point of C scanning determined in advance (in the drawing, they are shown at intervals smaller than the actual positions), on the X-ray fluoroscopic image. The range of the C scan can be grasped from the fluoroscopic image by the positions of the markers 68A and 68B. The markers 68A and 68B are attached slightly outside (outside the C scanning range) the positions of the starting point and the ending point of C scanning so as not to block the measuring light L1. Further, which of the markers 68A and 68B indicates the position of the start point and the position of the end point is switched depending on the direction of C scanning. Furthermore, the material of the markers 68 </ b> A and 68 </ b> B may be any material that has an X-ray transmittance different from that of the sheath 44.

  Here, terms used in the present embodiment will be described. For the A, B, and C axes defined above, the plane made by the A and B axes (the plane that is orthogonal to the C axis when the A axis is orthogonal to the C axis) is ', The surface formed by the A axis and the C axis is the' AC surface ', the surface formed by the B axis and the C axis is the' BC surface ', the cross section to be cut by the AB surface is the' AB cross section ', AC The cross section of the cutting target cut along the plane is called “AC cross section”, and the cross section of the cutting target cut along the BC plane is called “BC cross section”.

  1, the intensity value of the backscattered light L4 obtained from the interference light detection unit 72 or a value obtained by performing a predetermined conversion process on the intensity value is expressed as “scattered light intensity” or “scattered light intensity”. "Data" (mainly using scattered light intensity data), the data obtained by associating each scattered light intensity data with the position where each scattered light intensity data was obtained (or a position corresponding thereto), or "tomographic information" or "tomographic data" '(Mainly using tomographic data), AB cross-sectional data is' AB cross-sectional data', AC cross-sectional tomographic data is' AC tomographic data ', BC cross-sectional tomographic data is' BC tomographic data', for example, B scan and C This is tomographic information of a three-dimensional area obtained by performing scanning simultaneously (so-called spiral scanning) or alternately (B scanning while moving a certain distance in the C-axis direction). Assume that each scattered light intensity data is associated with each position as 'three-dimensional structure information' or 'three-dimensional structure data' (mainly using three-dimensional structure data).

  Furthermore, the images of the AB cross section, AC cross section, and BC cross section that can be displayed on the monitor are called “AB tomographic image”, “AC tomographic image”, and “BC tomographic image”, respectively. An image that can be displayed on the monitor generated from the structure data is referred to as a “three-dimensional structure image”, and an image that can be displayed on the monitor obtained by the measurement of the OCT apparatus 1 is referred to as an “OCT image”.

  In the OCT probe 40 of the above-described embodiment, the light emitting end that emits the measurement light L1 in the A-axis direction that is substantially perpendicular to the C-axis direction is the long axis of the OCT probe 40 in the C-axis direction (the light of the optical fiber FB2). The measurement light L1 is scanned in the B-axis direction by rotating around the axis), and the measurement light L1 is scanned in the C-axis direction by moving the light emitting end in the C-axis direction. Not exclusively. For example, the measurement light L1 can be scanned in two different directions by changing at least one of the emission direction or the emission position (position of the light emission end) of the measurement light L1 emitted from the light emission end of the OCT probe 40. The three-dimensional structure data of the three-dimensional region can be acquired. At this time, the emission direction of the measurement light L1 from the light emission end is the A-axis direction or A-axis, one of the two directions in which the measurement light L1 is scanned is the B-axis direction or B-axis, and the other is the C-axis direction or C-axis. Each process described below by defining an axis can be applied in substantially the same manner.

  Next, a procedure for observing the biliary tract / pancreatic duct using the OCT apparatus 1 will be described.

  First, as shown in FIG. 2, the patient is placed on the imaging stand of the X-ray fluoroscopic apparatus 110 in a room where the X-ray fluoroscopic apparatus 110 capable of communicating with the OCT apparatus 1 is installed, so that X-ray fluoroscopy is possible.

  Subsequently, as shown in FIG. 4, the insertion portion 202 of the endoscope 200 is inserted from the mouth of a patient (not shown), and the distal end of the insertion portion 202 is inserted into the duodenum via the esophagus / stomach (duodenal papilla). ) Until the position near.

  Although not described in detail, the endoscope 200 includes an insertion unit 202 that is inserted into a body cavity, a hand operation unit that is provided with an operation member that is held by an operator and performs various operations, as is well known. . A forceps channel through which a treatment tool such as forceps is inserted is provided inside the insertion portion 202, and a forceps port (forceps introduction port) 204 for inserting the treatment tool into the forceps channel is provided in the hand operation unit. A forceps port (forceps outlet port) 206 for leading the treatment tool inserted from the introduction port 204 is provided on the distal end surface of the insertion portion 202.

  In the same figure, the pancreas is shown beside the duodenum, and the liver is shown above it. A gallbladder is shown under the liver.

  As is well known, bile produced by the liver is transported to the duodenum by the bile duct and sent out into the duodenum through the opening of the fatter nipple. The gallbladder is connected to the bile duct by the gallbladder duct and accumulates bile.

  In addition, pancreatic juice produced by the pancreas is transported to the duodenum through the pancreatic duct in the pancreas. The fine pancreatic ducts gather together toward the duodenum, forming two thick ducts called the main pancreatic duct and the accessory pancreatic duct, which are connected to the duodenum. The main pancreatic duct merges with the common bile duct in the duodenal wall and becomes a short common duct, and is opened to the duodenum in the papilla.

  The bile duct is a general term for luminal structures that carry bile from the liver to the duodenum, and is named according to the site. The bile duct in the liver is called the intrahepatic bile duct, and the bile duct from the outside of the liver to the duodenum is called the extrahepatic bile duct. The intrahepatic bile duct is largely divided into two branches, called the left hepatic duct and the right hepatic duct, respectively. The extrahepatic hepatic duct is the part from the hilar region to the duodenum where the left hepatic duct and the right hepatic duct join, and the portion from the hilar part to the three-pipe confluence where the gallbladder duct joins is called the total hepatic duct From the junction of the three pipes to the duodenum is called the common bile duct.

  In addition to the bile ducts (intrahepatic bile ducts, extrahepatic bile ducts), the part including the gallbladder and the papilla is called the biliary tract, and the present invention can be applied to all parts of the biliary tract as targets for diagnosis and the like.

  When the distal end of the insertion portion 202 of the endoscope 200 is inserted to the vicinity of the duodenal papilla as described above, an ERCP (Endoscopic retrograde cholangiopancreatography) catheter is inserted into the forceps channel from the forceps opening (forceps inlet) 204. Derived from a forceps opening (forceps outlet) 206 and inserted into a bile duct or pancreatic duct while confirming a fluoroscopic image. After imaging, a guide wire 210 for guiding the OCT probe 40 of FIG. 3 to a desired position via the catheter is placed in the bile duct or pancreatic duct, and the site to be observed (measured) by the OCT apparatus 1 as shown in FIG. To place.

  Subsequently, the catheter is withdrawn in a state where the guide wire is placed, and the guide wire 210 is passed from the proximal end side to the guide wire hole 54 of the OCT probe 40 shown in FIG. It is inserted into the forceps channel from the introduction port) 204 and led out from the forceps port (forceps outlet port) 206, inserted along the guide wire 210 into the biliary tract or pancreatic duct from the opening of the fermenter papilla, and placed at the site to be observed (FIG. 5). reference).

  Further, during such work, the arithmetic processing unit 90 causes the arithmetic processing unit 90 of the OCT apparatus 1 to execute processing in the real-time observation mode, so that the arithmetic processing unit 90 repeats the B scanning with the OCT probe 40 and the AB tomographic data. And a circular AB tomogram is displayed on the monitor 100 in real time. While observing the AB tomogram displayed in real time, the observer moves the position of the OCT probe 40 and confirms the outline of the center of the lesioned part and the laterally developed region. Then, after determining the position of the OCT probe 40 so that the C scanning range is appropriate, the input device 92 performs an instruction operation for starting the C scanning measurement.

  When an instruction operation for starting C-scan measurement is performed, a trigger signal is given from the OCT apparatus 1 to the X-ray fluoroscopic apparatus 110 via the communication line before the C-scan starts, and the X-ray fluoroscopic apparatus 110 is triggered by the trigger signal. An X-ray fluoroscopic image of the peripheral region of the OCT probe 40 is taken, and the X-ray fluoroscopic image is given to the arithmetic processing unit 90 through a communication line or registered in the database 112. As described above, the sheath 44 of the OCT probe 40 is provided with the markers 68A and 68B indicating the start point and the end point of the C scan, and the markers 68A and 68B are imaged on the X-ray fluoroscopic image. The X-ray fluoroscopic image specifies which range of the entire area is the C-scanning range (measurement region).

  When the photographing and storage of the X-ray fluoroscopic image are completed, the OCT apparatus 1 starts the C scan by moving the light emitting end (optical lens 52) of the OCT probe 40 to the start position of the C scan. At this time, the B scan and the C scan are performed simultaneously (so-called spiral scan), for example, and three-dimensional structure data having a length of 100 mm, for example, is acquired. During the C scan, the observer holds the endoscope 200 and the OCT probe 40 in a stationary state. In the monitor screen, a specific AC tomographic image is developed in accordance with the C scan together with the AB tomographic image being measured. As shown in FIG. 6, it is preferable to install balloons 220 and 222 that can be inflated and contracted, for example, at two locations outside the C scanning range on the outer peripheral surface of the sheath 44 of the OCT probe 40. According to this, it is possible to inflate the balloons 220 and 222 by placing a fluid into the balloons 220 and 222 after placing the OCT probe 40 at the measurement site and to contact the inner wall of the measurement site. As a result, the relative position between the OCT probe 40 and the inner wall of the measurement site can be fixed without being affected by body movement.

  When the C-scan is finished, the OCT apparatus 1 gives a signal to end the measurement visually by the monitor 100 or audibly by the speaker. Then, after the optical lens 52 which is the light emitting end of the OCT probe 40 is returned to the start position of the C scan, the real time observation mode is restored, and the B scan is performed by the OCT probe 40. A tomogram and an AC tomogram are displayed on the monitor 100 in real time. The observer repeats the above operation when measuring another place, and when ending the observation, the observer pulls out the OCT probe 40 from the forceps channel of the insertion portion 202 of the endoscope 200 and Stop operation.

  As a method for inserting the OCT probe 40 into the biliary tract and pancreatic duct, there is a percutaneous liver observation method in addition to the transendoscopic observation method, which may be used. Further, instead of using a guide wire, the OCT probe 40 may be inserted into the biliary tract and pancreatic duct from the papilla part through the inside of the catheter. Any of the methods used for contrast medium injection in an ERCP (Endoscopic retrograde cholangiopancreatography) procedure or insertion of an IDUS (endoscopic retrograde cholangiopancreatography) probe can be used.

  In the above description, the case where the X-ray fluoroscopic apparatus 110 is used together with the OCT apparatus 1 has been described. However, instead of the X-ray fluoroscopic apparatus 110, other fluoroscopic apparatuses such as a CT, MRI, and THz imaging apparatus may be used. Images captured by the fluoroscopic devices are not only displayed in real time on a monitor included in the fluoroscopic devices, but also captured by the OCT device 1 via the communication line in the OCT device 1 and displayed on the monitor 100 of the OCT device 1 together with the OCT image. It may be displayed in real time.

  The three-dimensional structure data acquired by the OCT apparatus 1 as described above is image-processed by the arithmetic processing unit 90 of the OCT apparatus 1 and visualized as a three-dimensional structure image, and displayed on the monitor 100. The three-dimensional structure image can be displayed by switching to a different display form by various visualization techniques as will be described later.

  The arithmetic processing unit 90 of the OCT apparatus 1 also has a diagnosis support function for automatically detecting a lesioned part (a characteristic structure suspected as a lesioned part). The lesion part can be displayed on the monitor 100 by automatic detection.

  Further, the three-dimensional structure data acquired by the OCT apparatus 1 or the three-dimensional structure image generated thereby is stored in the database 112 connected by a communication line in association with various incidental information such as patient information. The patient information is information for identifying a patient such as a patient ID and a patient name.

  In addition, regarding the processing of the display function for displaying the three-dimensional structure image and the diagnosis support function for supporting diagnosis in the arithmetic processing unit 90 after acquiring the three-dimensional structure data, other than the OCT apparatus 1 connected to the database 112 via the communication line. It can also be implemented in other terminals (such as computers of electronic medical record systems). That is, software (OCT display software) for executing processing similar to that of the arithmetic processing unit 90 of the OCT apparatus 1 is executed in a terminal other than the arithmetic processing unit 90, and the three-dimensional structure data is acquired from the database 112 to obtain the arithmetic processing unit. 90 processes can be performed.

  Next, a diagnosis support function in the OCT apparatus 1 (arithmetic processing apparatus 90) will be described.

  When the three-dimensional structure data of a predetermined part of the biliary tract / pancreatic duct is acquired as described above, a three-dimensional structure image or the like is displayed by the display function of the arithmetic processing unit 90 (described later). When a lesion is present, the observer identifies the lesion and, when resecting, determines a resection range by determining a resection line.

  On the other hand, the arithmetic processing unit 90 has a diagnosis support function that automatically detects a lesioned part (part suspected as a lesioned part) and displays information on the region of the lesioned part together with the three-dimensional structure image. In this case, the diagnosis support function is appropriately executed by a predetermined operation of the input device 92 so that it can be used as a reference for specifying a lesioned part or determining a resection line.

  Note that the display function for displaying a three-dimensional structure image or the like is also implemented in the diagnosis support function in the same manner, and is performed in the description of the diagnosis support function.

  First, before explaining the diagnosis support function, the structure of the inner wall of the biliary tract / pancreatic duct and the morphological change when a lesion such as cancer occurs will be described.

  The structure of the inner wall portion in the biliary tract is formed by a mucosal epithelial layer, a fibromuscular layer, and a serous lower layer having different tissue structures in order from the lumen side surface (surface of the inner wall portion) to the outside as shown in FIG. In the present embodiment, each layer is referred to as a first layer, a second layer, and a third layer in order from the lumen side to the outside, and the inner wall surface (the surface of the first layer) is referred to as the outermost surface (or epithelial cell). In the three-dimensional structure data obtained by the above OCT measurement, the strongest scattered light intensity data is obtained in the second layer, and then the strong scattered light intensity data is obtained in the first layer if it is a normal site (normal tissue). The weakest scattered light intensity data is obtained in the third layer.

If there is a lesion on the inner wall of the biliary tract or pancreatic duct that requires excision such as cancer or a precancerous condition that is likely to change to cancer in the future, the normal site as shown in FIG. Morphological changes such as lesion types 1 to 6 are known pathologically.
1. A papillary ridge is formed on the outermost surface.
2. The first layer is thickened.
3. Fiberization proceeds.
4. The outermost cells (epithelial cells) grow randomly.
5. Cancerous ducts develop.
6. The shape of blood vessels changes. Or a new blood vessel is formed.

The arithmetic processing unit 90 detects the characteristic structure (form) indicated by the lesions of the above lesion types 1 to 6 from the three-dimensional structure data by executing the diagnosis support function, and extracts it as a feature region. As the detection processing method, the following nine types of detection processing types 1 to 9 are conceivable. Each of the detection processing types 1 to 9 described below has the following different types of lesions that can be detected. The lesion type can be detected.
Detection processing type 1: lesion types 1 to 3
Detection processing type 2: lesion types 1, 2, 4
Detection processing type 3: lesion type 1, 4
Detection processing type 4: lesion type 1
Detection processing type 5: lesion type 1
Detection processing type 6: lesion type 5
Detection processing type 7: lesion type 3
Detection processing type 8: lesion type 3
Detection processing type 9: lesion type 6
First, a description will be given of a process performed as a preprocess by the arithmetic processing unit 90 when any of the detection process types 1 to 9 is performed. A noise removal process for removing noise from the three-dimensional structure data acquired by OCT measurement. Is done. This noise removal processing is based on, for example, blurring processing.

  Here, when the three-dimensional structure data is represented by a voxel space obtained by subdividing a three-dimensional space with minimum unit voxels, the scattered light intensity data at each position in the measurement region corresponding to the position of each voxel is the voxel value of each voxel. As assigned. The voxel region to which the scattered light intensity data is assigned has a substantially cylindrical solid shape, similar to the measurement region, and includes an A axis (A axis direction), a B axis (B axis direction), and a C axis (C axis direction). ) Is defined similarly in the voxel space.

  In the blurring process, each of all the voxels to which the three-dimensional structure data is assigned is set as the target voxel in order, and voxels in a predetermined area centered on the target voxel (for example, 9 × 9 × 9 = adjacent to the target voxel). The average value of the scattered light intensity data of 27 voxels) is converted as the scattered light intensity data of the target voxel. As a result, noise is removed from the three-dimensional structure data.

  Subsequently, a sheath removal process for removing scattered light intensity data at the sheath 44 of the OCT probe 40 from the three-dimensional structure data is performed. In this sheath removal process, first, a signal indicating the value of scattered light intensity data at a position where each line in the A-axis direction passes (each position in the A-axis direction) is referred to as an A-axis direction scattered light intensity signal. At each position, the A-axis direction scattered light intensity signal is differentiated to generate a differential signal.

  Assuming that lines in the A-axis direction are present, there are many such lines at different positions in the B-axis direction and the C-axis direction, and when focusing on all the lines when focusing on the A-axis direction, And “at each position on the BC plane”.

  Next, at each position on the BC plane, a position in the A-axis direction where the differential signal has the maximum value is detected. The position where the differential signal indicates the maximum value indicates the position of the inner wall surface of the sheath 44, and the range from the position of the maximum value to a position spaced outward in the A-axis direction by the known sheath thickness is the sheath 44. Scattered light intensity data at. Then, the scattered light intensity data in the range is converted into a value reduced to the noise level. As a result, the scattered light intensity data at the sheath 44 is removed from the three-dimensional structure data.

  After the above preprocessing is performed, detection processing according to the following detection processing types 1 to 9 is performed by the arithmetic processing device 90.

1. Detection processing type 1
The detection processing type 1 processing is performed when there is a portion where the scattered light intensity of the first layer is out of a predetermined value range in the inner wall portion (internal wall portion) of the biliary tract and pancreatic duct from which the three-dimensional structure data has been acquired. This is processing for detecting a site as a lesion (characteristic region) having a characteristic structure.

  FIG. 8A shows the three-dimensional structure data of the inner wall of the living body having the lesion types 2 and 3 among the lesion types 1 to 3 detected by this detection processing type 1, and the AC tomogram passing through the lesion site. This is illustrated by an image. FIG. 8A shows a site where fiberization has progressed in the first layer, and in FIG. 8B in which a part of FIG. 8A is enlarged, the first layer is thicker than the surroundings. Sites are shown, and these sites correspond to lesion types 2 and 3 lesions.

  In order to detect such a lesion as a feature region, in the detection processing type 1, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, similarly to the sheath removal process in the pre-process, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S50).

  Next, at each position on each BC plane, a position in the A-axis direction where the obtained differential signal has the maximum value is detected. The position where the differential signal shows the maximum value is the position of the surface of the first layer of the living body wall portion, that is, the position of the outermost surface, and as a result, the living body wall portion at each position on the BC plane as shown in FIG. The position of the outermost surface in the A-axis direction is detected (step S52).

  Note that the position in the C-axis direction is fixed as the order in which the above processing is performed at each position on the BC plane, the position in the B-axis direction is sequentially changed, and the above processing is performed at all positions in the B-axis direction. Although the case where the same position is repeated by shifting the position in the axial direction and vice versa can be considered, any order may be used.

  When the position of the outermost surface of the living body inner wall is detected, next, at each position on the BC plane, as shown in FIG. 8B, a range corresponding to a predetermined length on the outer side in the A axis direction from the position of the outermost surface (for example, , The A-axis direction scattered light intensity signal (scattered light intensity data) in the 100 μm length range including the second layer is integrated (step S54). As a result, an integrated value of the scattered light intensity in a range from the outermost surface of the living body wall portion to a predetermined depth in the tissue is obtained. FIG. 8C illustrates an integrated value distribution obtained as a result.

  Next, a position where the integrated value is out of the predetermined value range is detected (step S56). That is, a position where the integrated value is larger than the predetermined upper limit value and a position smaller than the predetermined lower limit value are detected. When papillary ridges are formed on the outermost surface of the inner wall as in lesion types 1 and 2 or when the first layer is thickened, the integrated value decreases, and when fibrosis progresses as in lesion type 3 Since the integrated value becomes large, if a threshold value (upper limit value and lower limit value) is set to an appropriate value, a position that may be a lesion is detected by this detection.

  Then, detection of a region (a continuous region having a predetermined size or more) having a certain three-dimensional spread formed by the position detected as the integrated value is out of the predetermined value range is performed, If detected, the area is finally determined to be a lesion (lesion area) (step S58).

  The processing of the detection processing type 1 is performed by the above processing procedure, and the lesion part corresponding to any one of the lesion types 1 to 3 is detected, and the region of the detected lesion part is a lesion part described later. It is extracted as a feature area at the time of display.

  As a detection process similar to this detection process type 1, as described above, the A-axis direction scattered light intensity signal is not integrated in the range of a predetermined length outside the position of the outermost surface in the A-axis direction. When the position where the scattered light intensity data shows a value larger than a predetermined threshold is detected, and the detected position forms an area having a certain three-dimensional spread, that area is set as the area of the lesioned part. It may be detected and extracted as a feature region.

2. Detection processing type 2
The processing of the detection processing type 2 is characterized when there is a portion where the thickness of the first layer exceeds a predetermined threshold in the inner wall portion (biological wall portion) of the biliary tract and pancreatic duct from which the three-dimensional structure data is obtained. This is processing for detecting a lesion (characteristic region) having a typical structure.

  FIG. 9A shows the three-dimensional structure data of the inner wall of the living body having the lesion type 2 of the lesion types 1, 2, and 4 detected by this detection processing type 2, and the AC passing through the lesion unit. This is illustrated by a tomographic image. The figure shows a site where the first layer is thickened, and this site corresponds to a lesion of lesion type 2.

  In order to detect such a lesion, the detection processing type 2 refers to the flowchart of FIG. 30 in which each processing is performed according to the following procedure.

  First, similarly to the sheath removal process in the preprocessing, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S70).

  Next, at each position on the BC plane, based on the differential signal, the position of the outermost surface of the living body inner wall and the position of the boundary between the first layer and the second layer are detected as shown in FIG. 9B. For example, the change in the A-axis direction scattered light intensity signal is large between the position of the outermost surface of the living body inner wall and the position of the boundary between the first layer and the second layer, and the differential signal shows a relatively large maximum value. In the signal, a position indicating the largest local maximum value existing on the innermost side in the A-axis direction and the largest local maximum value existing on the outermost side is detected as a corresponding position (step S72).

  Next, the distance between the position of the outermost surface of the living body inner wall and the position of the boundary between the first layer and the second layer is obtained. Thereby, the thickness of the first layer at each position on the BC surface is obtained (step S74).

  Next, as shown in FIG. 9C, a position on the BC plane where the thickness of the first layer shows a value larger than a predetermined threshold is detected (step S76). As a result, the position of the lesion part thickened as in the lesion type 2 is detected. In addition, since the thickness of the first layer is increased for lesion types 1 and 4 as well, the positions of those lesions are also detected.

  Then, detection of a region (a continuous region having a predetermined size or more) having a three-dimensionally constant spread formed by the position detected as the thickness of the first layer being larger than the predetermined threshold is performed, If detected, the area is finally determined to be a lesion (lesion area) (step S78).

  The processing of detection processing type 2 is performed by the above processing procedure, the lesion part corresponding to one of the lesion types 1, 2, and 4 is detected, and the detected lesion part region is a lesion described later. It is extracted as a feature area when displaying a part.

  As a detection process similar to this detection process type 2, when a position indicating scattered light intensity data in a certain range is detected, and the detected position forms an area having a certain three-dimensional spread. The area may be detected as a lesion area and extracted as a feature area.

3. Detection processing type 3
The processing of detection processing type 3 is performed when a portion where the surface roughness of the outermost surface exceeds a predetermined threshold exists in the inner wall portion (biological wall portion) of the biliary tract and pancreatic duct from which the three-dimensional structure data is obtained. This is processing for detecting a lesion (characteristic region) having a characteristic structure.

  FIG. 10A shows an AC tomogram passing through the three-dimensional structure data of the living body wall portion having a lesion type 4 of the lesion types 1 and 4 detected by this detection processing type 3. Is exemplified. This figure shows a lesion type 4 lesion in which the outermost surface cells are randomly grown.

  In order to detect such a lesion, in the detection processing type 3, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, similarly to the sheath removal process in the pre-process, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S90).

  Next, at each position on the BC plane, the position in the A-axis direction where the obtained differential signal shows the maximum value is detected, and as shown in FIG. 10B, A on the outermost surface of the living body inner wall at each position on the BC plane. A position in the axial direction is detected (step S92). Here, this detected position is referred to as a measurement position.

  As the order in which the above processing is performed at each position on the BC plane, the position in the C-axis direction is fixed, the position in the B-axis direction is sequentially changed, and the above processing is performed at all positions in the B-axis direction. Although the case where the same position is repeated by shifting the position in the axial direction and vice versa can be considered, any order may be used.

  Next, as shown in FIG. 10B, the average position of the outermost surface in the A-axis direction is calculated (step S94). Specifically, a calculation area having a predetermined size is set on the BC plane with the position of the BC plane as a focus position and the focus position as the center. Then, the average value of the measurement positions (coordinate values) in the A-axis direction on the outermost surface in the calculation area is calculated. The average value calculated in this way is obtained as the average position in the A-axis direction of the outermost surface at the position of interest on the BC plane. This process is repeated while sequentially changing the position of interest on the BC plane, and the average position of the outermost surface in the A-axis direction is calculated over the entire range of the BC plane.

  Next, a difference (difference value) with respect to the average position of the measurement positions on the outermost surface is calculated at each position on the BC plane (step S96). At this time, when the coordinate value of the measurement position on the outermost surface is shifted so that the coordinate value in the A-axis direction of the average position is 0 (constant position) at each position on the BC plane, the surface of the outermost surface as shown in FIG. The curve value indicating the measurement position indicates the difference value.

  Next, as shown in FIG. 10C, the BC plane is divided into regions of a predetermined size (the B-axis direction and the C-axis direction are divided by a predetermined length unit), and the maximum difference value for each region. And the minimum value are obtained (step S98). The difference is set as a value (evaluation value) indicating the surface roughness of the outermost surface in each region as shown in FIG.

  Next, an area (position) in which the surface roughness of the outermost surface in each area shows a value larger than a predetermined threshold is detected (step S100). As a result, the position of the lesion part in which the outermost cells randomly proliferate as in the lesion type 4 is detected. Further, when the nipple-like bulge is generated on the outermost surface as in the lesion type 1, the surface roughness is increased, so that the position of the lesion part of the lesion type 1 is also detected.

  Then, a region having a certain spread (a continuous region having a predetermined size or more) formed by the detected region (position) is detected, and when the region is detected, the region is finally a lesioned part. It is determined (region of lesion) (step S102).

  The processing of the detection processing type 3 is performed by the above processing procedure, the lesion part corresponding to one of the lesion types 1 and 4 is detected, and the area of the detected lesion part is the lesion part described later. It is extracted as a feature area at the time of display.

  Note that the BC plane is not divided into regions of a predetermined size as in step S98, but equally spaced lines are set in either the B direction or the C direction, and a range of a predetermined length in each line. The difference between the maximum value and the minimum value of the difference value (difference value with respect to the average position of the measurement position on the outermost surface) may be obtained for each time, and the difference thus obtained may be used as the surface roughness value in the range. Further, the method of obtaining the surface roughness value is not limited to the above method. For example, after obtaining a difference value from an average value in an area of a predetermined size as shown in FIG. 10C, an average value of absolute deviation from the average line in the predetermined area may be used as an evaluation value. Good. Or, a method of calculating an evaluation value from the square value of the deviation, a method of setting the maximum value or the minimum value as an evaluation value, a method of obtaining an evaluation value from the standard deviation or variance of the difference value, a method of setting the number of peaks as an evaluation value, Any method may be used as long as it obtains a value indicating the surface roughness. In order to detect random cell growth of lesion type 4, since the diameter of one cell is about 10 μm, the size of the predetermined region is preferably at least 40 μm on one side. On the other hand, in order to detect a papillary ridge of lesion type 1, it is desirable that at least one side is 120 μm or more in order to detect a ridge of about 30 μm. In this manner, different lesion types may be detected by changing the size of the predetermined region.

  This detection processing type 3 is not limited to the biliary tract and pancreatic duct, but other organs having a flat epithelial structure at normal time, such as the bronchi, pharynx, esophagus, stomach, large intestine, esophagus, and urine. This is effective for detection of a lesion in a tube or the like, and when performing diagnosis support for these parts, this detection processing type 3 is used as a detection process for a lesion (feature region), and the detected lesion is displayed. Diagnosis support can be performed using the method as in the embodiment.

4). Detection processing type 4
The processing of detection processing type 4 is characterized when there is a part having a substantially spherical tissue form in the vicinity of the outermost surface in the inner wall part (in vivo wall part) of the biliary tract / pancreatic duct from which the three-dimensional structure data is obtained. This is processing for detecting a lesion (characteristic region) having a structure.

  FIG. 11A illustrates the three-dimensional structure data of the living body wall portion having the lesion portion of the lesion type 1 detected by the present detection processing type 4 as an AC tomographic image passing through the lesion portion. In the figure, a lesion type 1 lesion having a papillary ridge on the outermost surface is shown.

  In order to detect such a lesion, in the detection processing type 4, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, the position of the outermost surface of the living body inner wall and the position of the boundary between the first layer and the second layer are detected by the same processing as in the detection processing type 2 (steps S110 and S112).

  Next, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S114). In addition, you may use what memorize | stored the differential signal calculated | required by step S110.

  Next, the detection of the position where the slope of the A-axis direction scattered light intensity signal rises steeply in the position range outside the outermost surface and inside the boundary between the first layer and the second layer is the differential at each position on the BC plane. This is performed based on the signal (step S116). For example, when the A-axis direction scattered light intensity signal at the positions indicated by reference signs a and b in FIG. 11A is shown in FIG. 11B, the A-axis direction scattered light intensity at the position indicated by reference sign b where no lesion exists. The signal does not have a position where the slope rises steeply in the above position range. On the other hand, in the A-axis direction scattered light intensity signal at the position of the symbol a where the lesion exists, there is a position where the inclination rises steeply in the position range, and the position is detected.

  Then, an area formed by the detected position and having a certain three-dimensional extent (a continuous area having a predetermined size or more) is detected, and if it is detected, the area eventually becomes a lesion. Is determined to be a region (region of lesion) (step S118).

  The processing of the detection processing type 4 is performed by the above processing procedure, the lesion part corresponding to the lesion type 1 is detected, and the region of the detected lesion part is a feature in displaying the lesion part described later. Extracted as a region.

5. Detection processing type 5
In the processing of detection processing type 5, there is a portion having a substantially circular (bubble shape) high scattering region in the BC cross section near the outermost surface in the inner wall portion (internal wall portion) of the biliary tract and pancreatic duct from which the three-dimensional structure data is obtained. In this case, the part is detected as a lesion (characteristic region) having a characteristic structure. As a result, the lesion part of lesion type 1 in which a papillary ridge is generated on the outermost surface is detected.

  In the detection process type 5, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, as in the detection processing type 1, the position in the A-axis direction of the outermost surface of the living body inner wall at each position on the BC plane is detected (steps S130 and S132).

  Next, as shown in FIG. 12A, a BC tomographic image (reslice image) of the BC cross section at a position spaced apart from the position of the outermost surface by a certain distance in the A-axis direction is generated as shown in FIG. Scattered light intensity data in the cross section is extracted) (step S134).

  Next, the BC cross section is divided into regions of a predetermined size, and the tomographic image (scattered light intensity data) is Fourier transformed for each region to obtain the frequency component of the tomographic image of each region (step S136).

  Next, a region having a frequency component exceeding a predetermined threshold is detected (step S138). For example, when the tomographic image of the region indicated by reference signs a and b in FIG. 12B is subjected to Fourier transform, the value of the frequency component of each frequency is shown in FIG. There is no value that exceeds the threshold value in the frequency component in the region. On the other hand, when a bubble-shaped pattern is present in the BC tomogram near the outermost surface, the lesion is a lesion type 1 in which a papillary ridge has occurred on the outermost surface, and the region indicated by symbol b in which the lesion exists. There is a frequency peak showing a value that exceeds the threshold value in the frequency component at, and the frequency peak is detected by the above processing.

  Then, a region having a certain spread (continuous region having a predetermined size or more) formed by the region detected as having a frequency component exceeding a predetermined threshold is detected, and finally, when detected, Therefore, it is determined that the region is a lesion (region of the lesion) (step S140).

  The processing of detection processing type 5 is performed by the above processing procedure, and the lesion part corresponding to the lesion type 1 is detected, and the detected lesion part region is a feature in displaying the lesion part described later. Extracted as a region.

6). Detection processing type 6
The detection processing type 6 processing is performed when there is a part having a spherical or hemispherical low-scattering region having a diameter exceeding a predetermined value in the inner wall (biological wall) of the biliary tract and pancreatic duct from which the three-dimensional structure data is obtained. This is processing for detecting the site as a lesion (characteristic region) having a characteristic structure.

  FIG. 13A illustrates the three-dimensional structure data of the living body wall portion having a lesion portion of lesion type 5 detected by this detection processing type 6 as an AC tomographic image passing through the lesion portion. This figure shows a lesion type 5 lesion where a cancerous gland duct has developed.

  In order to detect such a lesion, in the detection processing type 6, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, the position of the outermost surface of the living body inner wall is detected by the same processing as that of the detection processing type 2 (steps S150 and 152).

  Next, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S154). In addition, you may use what memorize | stored the differential signal calculated | required by step S150.

  Next, the detection of the position where the slope of the A-axis direction scattered light intensity signal rises steeply in the living body is performed based on the differential signal at each position on the BC plane (step S156). For example, when the A-axis direction scattered light intensity signal at the positions indicated by reference signs a and b in FIG. 13 (A) is shown in FIG. 13 (B), A at the position of reference sign b where no lesion (cancerous gland duct) exists. In the axial scattered light intensity signal, there is no position where the slope rises steeply in the above position range. On the other hand, in the A-axis direction scattered light intensity signal at the position of the symbol a where the lesion exists, there is a position where the inclination rises steeply in the position range, and the position is detected.

  Then, an area formed by the detected position and having a certain three-dimensional extent (a continuous area having a predetermined size or more) is detected, and if it is detected, the area eventually becomes a lesion. Is determined to be a region (region of lesion) (step S158).

  The processing of the detection processing type 6 is performed by the above processing procedure, the lesion part corresponding to the lesion type 5 is detected, and the region of the detected lesion part is a feature in displaying the lesion part described later. Extracted as a region.

7). Detection processing type 7
The detection processing type 7 is performed in the A-axis direction from the outermost surface to the position where the scattered light intensity is at the signal limit (noise level) in the inner wall (internal wall) of the biliary tract and pancreatic duct from which the three-dimensional structure data is obtained. Is a process of detecting a site as a lesion (characteristic region) having a characteristic structure when a site having a distance shorter than a predetermined threshold exists.

  FIG. 14A illustrates the three-dimensional structure data of the living body wall portion having the lesion portion of the lesion type 3 detected by the present detection processing type 7 as an AC tomographic image passing through the lesion portion. This figure shows a lesion type 3 lesion in which fibrosis has progressed.

  In order to detect such a lesion, in the detection processing type 7, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, the position in the A-axis direction of the outermost surface of the living body inner wall at each position on the BC plane is detected by the same processing as in detection processing type 1 (steps S170 and S172).

  Next, at each position on the BC plane, the signal value of the scattered light intensity signal in the A-axis direction is a value indicating a signal limit equivalent to the noise level outside a certain position in the A-axis direction (smaller than a predetermined threshold as a signal limit) (Value) is extracted as shown in FIG. 14B (step S174). As a result, the position of the innermost boundary in the A-axis direction of the signal limit region is detected at each position on the BC plane.

  Next, at each position on the BC plane, the length in the A-axis direction (signal effective length) from the position of the outermost surface of the living body inner wall to the boundary position of the signal limit region is calculated as shown in FIG. (Step S176).

  Next, as shown in FIG. 14C, the position on the BC plane where the signal effective length in the A-axis direction indicates a value smaller than a predetermined threshold is detected (step S178). As fibrosis progresses as in lesion type 3, the intensity of backscattered light at that portion increases, and the signal effective length in the A-axis direction decreases accordingly, so that the position of the lesion portion of lesion type 3 is Detected.

  Then, a region having a certain spread formed by the detected position (a continuous region having a predetermined size or more) is detected, and when the region is detected, the region is finally determined as a lesioned portion (lesioned portion). (Step S180).

  That is, in the detection processing type 7, the signal value of the scattered light intensity signal with respect to the A-axis direction from the arbitrary measurement point set on the BC plane perpendicular to the A-axis direction (that is, the inner wall surface) indicates the signal limit. An evaluation value corresponding to the length to the boundary point that becomes smaller (that is, the noise level) is calculated, and a measurement point where the evaluation value is smaller than a predetermined threshold is detected. If the set of detected measurement points forms a continuous area having a predetermined size or more, the area is detected as a lesion.

  The processing of the detection processing type 7 is performed by the above processing procedure, the lesion part corresponding to the lesion type 3 is detected, and the region of the detected lesion part is a feature in displaying the lesion part described later. Extracted and set as an area.

8). Detection processing type 8
The processing of the detection processing type 8 is characterized when there is a site where the normal layer structure has disappeared in the inner wall (biological wall) of the biliary tract / pancreatic duct from which the three-dimensional structure data is obtained. This is processing for detecting a lesion (characteristic region) having a structure. Thereby, the detection of the lesion part of the lesion type 3 in which the fibrosis progresses and the layer structure disappears is performed.

  In this detection process type 8, each process is performed in the following procedure. Reference is made to the flowchart of FIG.

  First, scattered light intensity data is added in the B-axis direction at each position on the AC plane (step S190). As a result, at each position in the C-axis direction, an added value of the scattered light intensity data at each position in the A-axis direction is obtained. FIG. 15A shows the added value at each position in the A-axis direction at a certain position in the C-axis direction when this processing is performed on a normal inner wall portion having the first layer and the second layer. FIG. 15B shows an A axis at a certain position in the C axis direction when this processing is performed on an inner wall portion having a lesion portion in which the fibrosis has progressed and the layer structure has disappeared as in lesion type 5 The added value at each position in the direction is shown.

  Next, at each position in the C-axis direction, it is determined whether or not the added value shows a constant attenuation toward the outside in the A-axis direction (in the case of FIG. 15B). A position in the axial direction is detected (step S192). In this case, it is preferable to treat the addition value as affirmative even when the added value is in a range where it can be assumed that the attenuation is constant toward the outside in the A-axis direction. Thereby, it becomes possible to prevent the detection omission of the region where the layer structure has disappeared.

  Then, a region having a certain spread formed by the detected position (a continuous region having a predetermined size or more) is detected, and when the region is detected, the region is finally determined as a lesioned portion (lesioned portion). (Step S194). As a result, as shown in FIG. 15B, a site where the layer structure is lost is detected as a lesion.

  The processing of the detection processing type 8 is performed by the above processing procedure, the lesion part corresponding to the lesion type 3 is detected, and the detected lesion part region is a feature in displaying the lesion part described later. Extracted as a region.

  As a modification of detection processing type 8, processing may be performed with the B axis and C axis reversed. That is, in this modification, first, scattered light intensity data is added in the C-axis direction at each position on the AB plane. As a result, at each position in the B-axis direction, an added value of the scattered light intensity data at each position in the A-axis direction is obtained. Next, at each position in the B-axis direction, it is determined whether or not the added value shows a constant attenuation toward the outside in the A-axis direction, and the position in the B-axis direction when an affirmative is detected. Then, a region having a certain spread formed by the detected position (a continuous region having a predetermined size or more) is detected, and when the region is detected, the region is finally determined as a lesioned portion (lesioned portion). Area).

  Further, the adding direction of the scattered light intensity data is not limited to the B-axis direction or the C-axis direction, and may be a direction at least parallel to the BC plane. That is, you may make it add to the diagonal direction which is a direction parallel to BC surface, and is not orthogonal to a B-axis direction and a C-axis direction.

  That is, in the detection processing type 8, the scattered light intensity data in a predetermined range at the same depth in the A-axis direction is added from any measurement point set on the BC plane perpendicular to the A-axis direction (that is, the inner wall surface). If the added value is within a range where it can be assumed that the value is attenuated to the outside in the A-axis direction, the measurement point is detected. If the set of detected measurement points forms a continuous area having a predetermined size or more, the area is detected as a lesion.

9. Detection processing type 9
The processing of detection processing type 9 is performed when there is a portion where a low scattering region that is distributed linearly or in a net shape is present in the inner wall (biological wall) of the biliary tract / pancreatic duct from which three-dimensional structure data is obtained. This is processing for detecting the site as a lesion (characteristic region) having a characteristic structure.

  FIG. 16A illustrates three-dimensional structure data of a living body wall portion having a lesion part of lesion type 6 detected by this detection processing type 7 as an AC tomographic image passing through the lesion part. In the same figure, a lesion part of lesion type 6 in which blood vessels (new blood vessels) are generated in the second layer is shown (see FIG. 16C).

  In order to detect such a lesion, in the detection process type 9, each process is performed according to the following procedure. Reference is made to the flowchart of FIG.

  First, the position of the inner wall surface is detected by the same processing as that of the detection processing type 2 (steps S210 and S212).

  Next, a differential signal obtained by differentiating the A-axis direction scattered light intensity signal is obtained at each position on the BC plane (step S214). In addition, you may use what memorize | stored the differential signal calculated | required by step S210.

  Next, the detection of the position where the slope of the A-axis direction scattered light intensity signal rises steeply in the living body is performed based on the differential signal at each position on the BC plane (step S216). For example, in FIG. 16A, the A-axis direction scattered light intensity signal at the positions indicated by reference signs a and b in FIG. 16A is shown in FIG. There is no position where the slope rises steeply in the above position range. On the other hand, the A-axis direction scattered light intensity signal at the position of the symbol a where the blood vessel exists has a position where the inclination rises steeply in the above position range, and the position is detected.

  Then, a region formed by the detected position and having a certain three-dimensional extent (a continuous region having a predetermined size or more) and a region distributed in a linear or net shape (FIG. 16C )) Is detected, and if it is detected, the area is finally determined to be a lesion (region of the lesion) (step S218).

  The processing of the detection processing type 9 is performed by the above processing procedure, and the lesion part corresponding to the lesion type 9 is detected, and the detected lesion part region is a feature in displaying the lesion part described later. Extracted as a region.

  Note that there may be a portion where the sheath 44 of the OCT probe 40 is in contact with the outermost surface of the living body wall portion and a portion where it is not in contact during OCT measurement. When the sheath 44 is in contact with the outermost surface of the living body inner wall portion, the inner wall portion is compressed by the pressing effect. Therefore, when the thickness of the first layer is larger than a predetermined threshold as in the detection processing type 2, the lesioned portion In such a case, there is a possibility that even if a lesioned part is generated, it is not determined to be a lesioned part.

  Therefore, a region in which the sheath 44 is in contact with the outermost surface of the living body wall portion and a region in which the sheath 44 is not in contact are detected and separated, and a threshold corresponding to the contact state or non-contact state of the sheath 44 for each region. May be set.

  Detection of the area (contact area) where the sheath 44 is in contact with the area (non-contact area) where the sheath 44 is not in contact can be performed as follows. FIG. 17A illustrates the three-dimensional structure data before removing the scattered light intensity data of the sheath 44 as an AC tomographic image. In FIG. 17A, the contact area and the non-contact area of the sheath 44 are shown. ing. In this three-dimensional structure data, for example, the scattered light intensity signal in the A-axis direction at the position of the non-contact region indicated by the symbol a in FIG. 17A shows a signal as shown in FIG. The A-axis direction scattered light intensity signal at the position of the contact area indicated by the symbol b shows a signal as shown in FIG. According to this, in the non-contact region, as shown in FIG. 17B, the inner wall 1 and outer wall m of the sheath 44, the outermost surface n of the living body inner wall portion, and the boundary o between the first layer and the second layer are characteristic. Changes. On the other hand, in the contact region, since the positions of the outer wall m and the outermost surface n of the sheath 44 are substantially coincident as shown in FIG. 17C, there are no four characteristic changes as in the non-contact region.

  Therefore, the sheath 44 is obtained by differentiating the A-axis direction scattered light intensity signal at each position on the BC plane and detecting the positions of l, m, n, and o of the A-axis direction scattered light intensity signal based on the differential signal. The contact area and the non-contact area can be detected.

  Hereinafter, the processing procedure in the arithmetic processing unit 90 when the threshold value of the detection processing type 2 is changed according to the contact area and the non-contact area of the sheath 44 will be described with reference to the flowchart of FIG.

  For example, before the detection process type 2 process is performed, the following process for determining the contact area and the non-contact area of the sheath 44 is performed. First, an A-axis direction scattered light intensity signal at each position on the BC plane is formed from the three-dimensional structure data including the scattered light intensity data of the sheath 44 before the sheath removal process is performed as a pre-process. At the position, the A-axis direction scattered light intensity signal is differentiated to generate a differential signal (step S300).

  Next, the range of the sheath 44 is detected based on the differential signal at each position on the BC plane (step S302). That is, the position in the A-axis direction where the differential signal shows the maximum value (or the position of the maximum value existing on the innermost side in the A-axis direction) is detected as the position of the inner wall surface of the sheath 44, and the known sheath 44 is detected from that position. A range up to the position of the outer peripheral surface (outer wall surface) of the sheath 44 that is separated to the outside in the A-axis direction by the thickness is detected as the range of the sheath 44.

  Next, at each position on the BC plane, whether the differential signal has a signal value of a large value (a value larger than a predetermined threshold value) corresponding to the outermost surface of the living body inner wall portion outside the range (outer peripheral surface) of the sheath 44. Thus, it is determined whether the sheath 44 is a non-contact area or a contact area (step S304). That is, in the non-contact region of the sheath 44, an A-axis direction scattered light intensity signal as shown in FIG. 17B is obtained, so that the differential signal has a large value (predetermined value) outside the position of the outer peripheral surface of the sheath 44. Signal value greater than the threshold value). On the other hand, since the A-axis direction scattered light intensity signal as shown in FIG. 17C is obtained in the contact region of the sheath 44, the differential signal has a large value (predetermined threshold value) outside the position of the outer peripheral surface of the sheath 44. Does not have a signal value of (greater value). Therefore, when it is determined that the differential signal has a signal value of a large value (a value larger than a predetermined threshold value) corresponding to the outermost surface of the living body wall outside the outer peripheral surface of the sheath 44, the differential signal is obtained. It is determined that the position of the BC surface is a non-contact region (non-contact position) of the sheath 44, and the differential signal is larger than the outer peripheral surface of the sheath 44 and has a large value corresponding to the outermost surface of the living body wall (predetermined value). If it is determined that the signal value is not greater than the threshold value, it is determined that the position on the BC plane from which the differential signal is obtained is the contact region (contact position) of the sheath 44.

  Next, for the position of the BC plane determined as the non-contact region of the sheath 44, the threshold used for the detection processing type 2 process is set as the first threshold (step S306). On the other hand, for the position of the BC surface determined as the contact area of the sheath 44, the threshold used for the detection processing type 2 process is set as the second threshold (step S308).

  When the above processing is completed, detection processing type 2 processing is performed. At that time, when performing the process of step S76 of FIG. 30, the thickness of the first layer is compared with the first threshold at the position of the BC surface determined as the non-contact region of the sheath 44, and the thickness of the first layer is determined. The position of the BC plane showing a value greater than the first threshold is detected as a candidate lesion. On the other hand, at the position of the BC surface determined as the contact area of the sheath 44, the thickness of the first layer is compared with the second threshold value, and the position of the BC surface showing the value of the first layer thickness larger than the second threshold value is Detected as a candidate lesion.

  In the contact region of the sheath 44, the thickness of the first layer is compressed by the sheath pressing effect as compared to the non-contact region in both the normal site and the lesioned part (because the thickness is reduced). The second threshold value is set to a value smaller than the first threshold value.

  Thereby, in the processing of the detection processing type 2, the threshold for determining whether or not the thickness of the first layer is normal is set to an appropriate value depending on whether the sheath 44 is in the contact area or non-contact area, Is properly detected.

  The detection processing type 2 process is not limited to the biliary tract and pancreatic duct, but other layers having a layer structure similar to the biliary tract and pancreatic duct, and the first layer having a substantially constant thickness, such as bronchi and pharynx at normal times. It is effective for detecting lesions in the esophagus, ureters, etc., and when performing diagnosis support for these parts, detection processing type 2 is used as a detection process for lesions (characteristic regions), and the detected lesions are displayed. The diagnosis support can be performed using the method in the same manner as in the present embodiment. At that time, if the threshold value is changed between the contact region and the non-contact region of the sheath 44 as described above, the lesioned part is appropriately detected.

  In addition, the process of determining the contact area and the non-contact area of the sheath 44 and changing the threshold value between the contact area and the non-contact area as described above can be effectively applied to the detection process type 7 process. That is, in the processing of the detection processing type 7, as shown in step S178 of the flowchart of FIG. 35, the position on the BC plane where the signal effective length in the A-axis direction is smaller than a predetermined threshold is detected as the position of the lesioned part. Is done. In the contact area of the sheath 44, the effective signal length in the A-axis direction in the normal region and the lesioned part is smaller than that in the non-contact area due to the pressing effect of the sheath 44. Therefore, the threshold value to be compared with the signal effective length is also in the contact area of the sheath 44. It is preferable to set a value smaller than the non-contact area.

  In the above description of detection processing types 1 to 9, in the above description, the scattered light intensity data is stronger in the second layer than in the first layer, but the opposite may occur depending on the measurement conditions. Therefore, in that case, the processing of each detection processing type 1 to 9 may be changed in consideration of that.

  Next, a diagnosis support function in the arithmetic processing unit 90 of the OCT apparatus 1 will be described.

  FIG. 18 is a flowchart showing the processing procedure of the first embodiment of the diagnosis support function. Hereinafter, the first embodiment of the diagnosis support function will be described in accordance with this flowchart. Note that the flowchart in the figure also shows functional blocks in the arithmetic processing device 90, and the processing in each step also represents processing as each processing unit in the arithmetic processing device 90 (the same applies to the flowchart in FIG. 21).

  First, the arithmetic processing unit 90 acquires the three-dimensional structure data of the inner wall (biological inner wall) of the biliary tract / bile duct to be diagnosed (step S10). In addition to the case of acquiring directly after OCT measurement, it is also possible to acquire the three-dimensional structure data stored in the database 112.

  Next, when the observer instructs the execution of the diagnosis support function using the input device 92, the arithmetic processing device 90 executes a feature extraction process (step S12). In this feature extraction processing, the processing of each of the detection processing types 1 to 9 described above is executed, and the feature region suspected of being a lesion is extracted as described above. However, it is not always necessary to execute the processes of all the detection process types 1 to 9, and any one or a plurality of detection process types may be executed. The operator may be able to select whether to do this. When only one detection processing type process is executed, the color corresponding to the feature region is one color in the coloring process in step S16 described later, and therefore the feature classification process in the next step S14 is not necessary.

  Next, the arithmetic processing unit 90 executes feature classification processing (step S14). In this feature classification process, each feature region extracted by the feature extraction process is classified by a predetermined classification item. For example, depending on the type of detection processing type that extracted each feature region, the method of classifying the type of extraction target (type of characteristic shape) as a classification item, or the change in mucosal morphology with a strong suspicion of a lesion Extracted (extracted by detection processing types 1 to 5, 8) and weakly suspected lesions (detection processing type 6) extracted by the presence of a gland duct or blood vessel network that may exist even in a normal part , 9) and the like. These classification methods may be appropriately changed by the operator.

  Next, the arithmetic processing unit 90 executes a coloring process (step S16). In this coloring process, color information is associated with a different color for each feature region belonging to each classification item classified by the feature classification process.

  Next, the arithmetic processing unit 90 performs a composition process of the three-dimensional structure data and the colored feature regions (step S18). In this synthesis process, the color information associated with each feature area is assigned to the voxel corresponding to each feature area of the 3D structure data, and the 3D structure data with the feature area information (referred to as 3D structure data with feature area information). Is generated.

  Next, the arithmetic processing unit 90 performs a display process (step S20) for visualizing the three-dimensional structure data with the feature region information and displaying the data on the monitor 100. In this display process, for example, as shown in FIG. 19 or FIG. 20, a display screen that displays a plurality of three-dimensional structure images is generated and displayed on the monitor 100.

  When comparing the configuration of the display screens of FIG. 19 and FIG. 20, all the components in FIG. 19 are included in FIG. 20, and therefore, the same reference numerals as those in FIG. The configuration of the display screen in FIG. 20 will be described.

  20 includes a character information display unit 302 that displays character information, a three-dimensional image display unit 304 that displays three-dimensional structure data as a three-dimensional image, and a BC tomogram display unit that displays a BC tomogram. 306, an AC tomogram display unit 308 that displays an AC tomogram, an AB tomogram display unit 310 that displays an AB tomogram, and a modality image display unit 312 that displays a modality image are provided.

  The character information display unit 302 displays various character information such as the date on which the measurement was performed and patient information.

  The three-dimensional image display unit 304 includes a three-dimensional image obtained by viewing the entire three-dimensional structure data from an oblique direction outside (an “perspective mode” image described later), and an endoscope inserted into the lumen of the biliary tract / pancreatic duct. A three-dimensional structure image that displays the three-dimensional structure data in a three-dimensional manner, such as a three-dimensional image (an “endoscope mode” image described later) as if viewed, is displayed. Although a three-dimensional image in the “perspective mode” is shown in the figure, it is possible to switch the form (display mode) of the three-dimensional image displayed by the observer operating the input device 92. The three-dimensional image is displayed on the three-dimensional image display unit 304 by performing a visualization process in the selected display mode on the three-dimensional structure data with the feature area information by the display process of the arithmetic processing unit 90. The The three-dimensional image displayed on the three-dimensional image display unit 304 includes a BC cross-sectional line 320 indicating the position and range of the BC tomogram displayed on the BC tomogram display unit 306, and an AC tomogram display unit 308. AC cross sectional line 322 indicating the position and range of the AC tomographic image to be displayed, and AB cross sectional line 324 indicating the position and range of the AB tomographic image displayed on the AB tomographic image display unit 310 are combined and displayed on the three-dimensional structure image. .

  The BC tomogram display unit 306 displays the voxel value (scattered light intensity) of the three-dimensional structure data with feature region information at the position of the BC section line 320 displayed on the three-dimensional image display unit 304 by the display processing of the arithmetic processing unit 90. Data and color information) are visualized, and a BC tomogram in FIG. The BC tomogram shown in the figure includes two feature regions 330 and 332 extracted by the feature extraction process (step S12 in FIG. 18) and classified into different classification items by the feature classification process (step S14 in FIG. 18). In the feature regions 330 and 332, the associated colors are translucent and displayed superimposed on the image of the living body inner wall. Such coloration display of the feature region is not shown in the drawing, but the three-dimensional image of the three-dimensional image display unit 304, the AC tomogram of the AC tomogram display unit 308, and the AB tomogram of the AB tomogram display unit 310. In case of including a feature region, the same operation is performed. In addition, if the colored display of the feature area obstructs the observation of the image of the living body inner wall, it is possible to erase the color by a predetermined operation of the observer. In this case, each feature area colored with the three-dimensional structure data The image of each display unit is regenerated and displayed based on the three-dimensional structure data before performing the combining process. That is, even when the diagnosis support function is not executed, a display screen configured in the same manner as in FIGS. 19 and 20 is displayed by the display function of the arithmetic processing unit 90. At that time, the three-dimensional structure data is used as it is, and the image of each display unit is displayed. Even after the diagnosis support function is executed, it is possible to appropriately switch to the display when such a diagnosis support function is not used.

  The position of the BC cross section line 320 can be changed by an operation of the input device 92 of the observer. For example, the BC cross section line 320 of the three-dimensional image display unit 304 is dragged using the mouse. By doing so, the position of the BC cross section line 320 can be changed in the directions of the A axis and the B axis, the BC tomogram display unit 306 is selected, and the slider 342 of the position selection bar 340 at the bottom of the screen is moved to the left and right. By moving, the position of the BC section line 320 can be changed in the B-axis direction. When the position of the BC cross section line 320 is changed, the BC tomogram at the changed position is regenerated from the three-dimensional structure data with feature region information, and the display of the BC tomogram on the BC tomogram display unit 306 is updated. It is also possible to change the size of the range of the BC section line 320 by operating the input device 92 of the observer to enlarge or reduce the area of the BC section displayed as the BC section image.

  The AC tomogram display unit 308 displays the voxel value (scattered light intensity) of the three-dimensional structure data with feature region information at the position of the AC cross section line 322 displayed on the three-dimensional image display unit 304 by the display processing of the arithmetic processing unit 90. Data and color information) are visualized and an AC tomogram is displayed.

  Further, the position of the AC cross section line 322 can be changed by the observer's operation of the input device 92 as in the case of the BC cross section line 320. For example, the position of the 3D image display unit 304 can be changed using a mouse. By dragging the AC cross section line 322, the position of the AC cross section line 322 can be changed in the direction of the B axis, and the AC tomogram display unit 308 is selected, and the slider of the position selection bar 340 at the bottom of the screen. By moving 342 left and right, the position of the AC cross section line 322 can be changed in the B-axis direction. When the position of the AC cross section line 322 is changed, the AC tomogram at the changed position is regenerated from the three-dimensional structure data with feature region information, and the AC tomogram on the AC tomogram display unit 308 is updated. Note that the AC cross-sectional area displayed as an AC cross-sectional image can be enlarged or reduced by changing the size of the range of the AC cross-sectional line 322 by operating the input device 92 of the observer.

  Further, the AC tomogram display unit 308 displays a scale indicating the actual dimension of the AC tomogram in the C-axis direction, and two cut lines described later are displayed on the AC tomogram.

  The AB tomogram display unit 310 displays the voxel value (scattered light intensity) of the three-dimensional structure data with feature region information at the position of the AB cross-sectional line 324 displayed on the three-dimensional image display unit 304 by the display processing of the arithmetic processing unit 90. Data and color information) are visualized and an AB tomogram is displayed.

  Further, the position of the AB cross section line 324 can be changed by the observer's operation of the input device 92 as in the case of the AC cross section line 322 and the like. For example, the three-dimensional image display unit 304 is used using a mouse. By dragging the AB cross section line 324, the position of the AB cross section line 324 can be changed in the direction of the C axis. Further, the position of the AB cross section line 324 can be changed in the C-axis direction by setting the AB tomogram display unit to the selected state and moving the slider 346 of the position selection bar 344 in the AB tomogram display unit 310 to the left and right. It is like that. It is also possible to enlarge or reduce the area of the AB cross section displayed as the AB cross section image by changing the size of the range of the AB cross section line 324 by operating the input device 92 of the observer.

  The modality image display unit 312 displays a modality image of the same patient or the like acquired by a medical device (modality) other than the OCT apparatus 1 such as a CT fluoroscopy apparatus, an MRI apparatus, or an EUS apparatus. The modality image to be displayed can be selected by the observer from the image data stored in the database 112 connected via the communication line, and the arithmetic processing unit 90 selects the selected image. Image data is acquired from the database 112 and displayed on the modality image display unit 312.

  The display screen as described above is displayed on the monitor 100 by the display processing of the arithmetic processing unit 90, and the observer can identify the lesioned part with reference to various three-dimensional structure images (three-dimensional images, tomographic images) displayed on the display screen. Identify. At this time, the position and size of the BC cross-section line 320, the AC cross-section line 322, and the AB cross-section line 324 are changed, and a tomogram at an arbitrary position in the measurement region is observed in detail to identify a lesioned part. Can do. Further, the length of the measurement region in the C-axis direction usually reaches a range of 10 to 1000 times the length in the A-axis direction. Therefore, for example, a measurement region including the C axis as an image indicating which range of the entire measurement region is an enlarged and displayed image such as a BC tomogram, an AC tomogram, and an AB tomogram. The simultaneous display of the entire image on the three-dimensional image display unit 304 greatly contributes to improvement in operability. Furthermore, since the characteristic region suspected of being a lesion is colored and displayed at a glance, the observation time is significantly reduced as compared with the case where the entire measurement region is observed. The color of the feature area is also displayed with a different color for each feature area classified by a predetermined classification item, thereby reducing the work burden on the observer. It is more effective if each feature region is displayed in different colors according to the strength of the suspicion of the lesion.

  In addition, the observer mainly observes the boundary area between the normal area and the characteristic area, which are particularly important among the characteristic areas extracted by the characteristic extraction process (step S12), and the boundary of the lesion area or surgery. Determine the resection line. At this time, the observer can set the positions of the two excision lines 350 and 352 displayed on the AC tomographic image display unit 308 of FIG. 20 to desired positions using a mouse or the like. ing. The two excision lines 350 and 352 indicate both ends of the region to be excised, and are set at the positions of both ends of the AC tomogram in a defined state. However, each can be set by moving to a desired position. is there. The observer can mark the positions of both ends of the ablation area sequentially determined by the ablation lines 350 and 352.

  The modality image display unit 312 has an X-ray fluoroscope around the measurement site photographed by the X-ray fluoroscope 110 when the OCT probe 40 of the OCT apparatus 1 is inserted into the measurement site of the biliary tract and pancreatic duct. It is possible to display an image. At this time, since the markers 68A and 68B indicating the start point and the end point of the C scan given to the OCT probe 40 are contrasted in the X-ray fluoroscopic image, the measurement region is set with the positions of the markers 68A and 68B as marks. It is possible to grasp the position of the biliary tract or the entire pancreatic duct. Since the positions of the resection lines 350 and 352 set in the AC tomogram display unit 308 in FIG. 20 are known in the measurement region by the scale, the resection region is the biliary tract / pancreatic duct. It is possible to grasp the entire range. As processing of the arithmetic processing unit 90, the positions of the resection lines 350 and 352 set in the AC tomographic image display unit 308 with the positions of the markers 68A and 68B on the X-ray fluoroscopic image of the modality image display unit 312 as marks are used as X-rays. It is also possible to display it superimposed on a fluoroscopic image. In addition, the final determined positions of the resection lines 350 and 352 are stored in the database 112, or the display screen of the monitor 100 is pasted on the electronic medical chart as it is so that it can be confirmed on the electronic medical chart. It is also possible to easily carry out explanations to patients and confirmation between doctors.

  Next, a second embodiment of the diagnosis support function will be described according to the flowchart showing the processing procedure of FIG.

  First, the arithmetic processing unit 90 acquires the three-dimensional structure data of the inner wall (biological wall) of the biliary tract / bile duct to be diagnosed (step S30). In addition to the case of acquiring directly after OCT measurement, it is also possible to acquire the three-dimensional structure data stored in the database 112.

  Next, when the observer instructs the execution of the diagnosis support function through the input device 92, the arithmetic processing device 90 executes a feature extraction process (step S32). In this feature extraction processing, the processing of each of the detection processing types 1 to 9 described above is executed, and the feature region suspected of being a lesion is extracted as described above. However, it is not always necessary to execute the processes of all the detection process types 1 to 9, any one of the plurality of detection process types may be executed, and the operator may determine which detection process type is to be executed. May be selectable.

  Next, the arithmetic processing unit 90 performs a feature classification process (step S34). In this feature classification process, each feature region extracted by the feature extraction process is classified by a predetermined classification item. For example, depending on the type of detection processing type that extracted each feature region, the method of classifying the type of extraction target (type of characteristic shape) as a classification item, or the change in mucosal morphology with a strong suspicion of a lesion Extracted (extracted by detection processing types 1 to 5, 8) and weakly suspected lesions (detection processing type 6) extracted by the presence of a gland duct or blood vessel network that may exist even in a normal part , 9) and the like. These classification methods may be appropriately changed by the operator.

Next, the arithmetic processing unit 90 executes accuracy estimation processing (step S36). In this accuracy estimation process, a score (1 to 10) corresponding to the degree of possibility of a lesion is assigned to each classification item in which each feature area is classified by the feature classification process. For example, assume that the following a to d are adopted as classification items.
a. Feature region where layer structure has disappeared b. Feature region with thickened first layer c. Feature area of surface roughness abnormality d. At this time, 10 feature areas belonging to the classification item a, 3 feature areas belonging to the classification item b, and 3 classification items depending on the degree of possibility that the feature area is a lesion (cancer) The feature area belonging to c is 3 points, and the feature area belonging to the classification item d is 2 points.

  In each detection processing type 1 to 9, the score may be changed in consideration of a value (feature amount) finally used for determining whether or not the region is a feature region.

  Next, the arithmetic processing unit 90 performs score processing (step S38). In this score processing, the degree of risk is set based on the score assigned to each feature area. For example, if the number of points assigned to each feature region is 10 points or more, it is set to a risk A indicating that the possibility of a lesion (cancer) is high, and if it is 5 points or more and less than 10 points, Set to risk level B indicating that the possibility of the lesion is moderate, and if it is 3 points or more and less than 5 points, it is set to the risk level C indicating that the possibility of the lesion is low but attention is required Is done. In addition, for the overlap region where a plurality of feature regions overlap, the points assigned to these feature regions are added and the added value is assigned. Then, risk levels A to C are set in the superposed area based on the points assigned to the area.

  FIG. 22A shows a BC tomogram displayed on a display unit corresponding to the BC tomogram display unit 306 in FIG. 20, and the feature region extracted by the feature extraction process in the BC tomogram is shown. Assume that these feature areas are classified by the classification items a to d by the feature classification process. In the figure, for example, in the feature area of the classification item b, 3 points in the non-overlapping area 300 that does not overlap with the other feature areas, 3 + 3 = 6 points in the overlap area 302 that overlaps only the feature area of the classification item c, 3 + 2 = 5 points are assigned to the overlapping area 304 that overlaps the feature area of the classification item d, and 3 + 3 + 2 = 8 points are assigned to the overlapping area 306 that overlaps the feature areas of both the classification item b and the classification item c. Similarly, in other feature areas, points are assigned to the non-overlapping area and the overlapping area. Then, for each non-overlapping area and overlapping area to which points are assigned (hereinafter, each of the non-superimposing area and the overlapping area is referred to as an evaluation area), risk levels A to C corresponding to the points are set. Each evaluation area in which the risk levels A to C are set in this way is as shown in FIG.

  Note that the score assigned to each evaluation area before conversion to the risk level may be used as the value of the risk level.

  Next, the arithmetic processing unit 90 executes a coloring process (step S40). In this coloring process, color information of different colors corresponding to the risk level is associated with each evaluation area in which the risk level is set.

  Next, the arithmetic processing unit 90 executes a synthesis process of the three-dimensional structure data and the colored evaluation areas. In this synthesis process, color information associated with each evaluation region is assigned to voxels corresponding to each evaluation region of the three-dimensional structure data, and three-dimensional structure data with feature region information is generated.

  Next, the arithmetic processing unit 90 executes display processing for visualizing the three-dimensional structure data with feature region information and displaying it on the monitor 100. In this display processing, for example, a display screen that displays a plurality of types of three-dimensional structure images as shown in FIG. 19 or FIG. 20 is generated and displayed on the monitor 100 as in the first embodiment. Is called.

  Operations such as the configuration of the display screen, diagnosis using it, setting of ablation lines, and the like are the same as in the first embodiment, and thus the description thereof will be omitted. However, in FIGS. A three-dimensional image displayed, a BC tomogram displayed on the BC tomogram display unit 306, an AC tomogram displayed on the AC tomogram display unit 308, and an AB tomogram displayed on the AB tomogram display unit 310 The coloring of the feature area is different from that of the first embodiment. In the second embodiment, for example, the BC tomogram of the BC tomogram display unit 306 is evaluated according to the risk level indicating the degree of the possibility of a lesion (cancer) as shown in FIG. The area is displayed with different colors. Therefore, for the observer, the region to be observed with emphasis is obvious at a glance, and the work load is reduced.

  Next, among the display forms (display modes) of the image that is displayed on the monitor 100 by visualizing the three-dimensional structure data, it is particularly preferable when the feature area is colored with the three-dimensional structure data with the characteristic area information. A display mode to be considered will be described. Note that the display screen displayed on the monitor 100 is not necessarily limited to the configuration shown in FIGS. 19 and 20, and how an image in each display mode described below is displayed on the screen of the monitor 100 is limited to a specific mode. Not. Further, the present invention is not limited to the case where the feature region is displayed with a color, but can be applied to a display when the diagnosis support function is not used.

  As shown in FIG. 23A, the three-dimensional structure data (three-dimensional composition data with feature area information) has a substantially cylindrical three-dimensional shape as in the measurement area when shown in the voxel space. The central axis 400 of the three-dimensional structure data is the position (the long axis of the OCT probe 40) where the light emitting end (optical lens 52) has moved during C scanning, and the OCT probe is located in a cylindrical region near the central axis 400. The scattered light intensity data at the 40 sheaths 44 exists, and the scattered light intensity data of the inner wall exists in the outer region. Assuming that the scattered light intensity data of the cavity has a small value and is treated as non-existing, the three-dimensional structure data excluding the cavity region of the inner cavity portion are the region 402 of the inner cylindrical sheath 44 and the outer cylindrical inner wall portion. Area 404. The arithmetic processing unit 90 of the OCT apparatus 1 generates various three-dimensional structure images that can be displayed on the monitor 100 by performing various visualization processes (rendering processes) on the three-dimensional structure data. indicate. In particular, for example, the following (1) “endoscopic mode”, (2) “perspective mode”, and (3) “long-axis cross-sectional mode” are suitable as the types of display modes when performing colored display of feature areas in the diagnosis support function. ", (4)" surface mode ".

  In the “endoscopic mode” and the “perspective mode”, as shown in FIG. 23A, from the three-dimensional structure data in which the scattered light intensity data (data in the region 402) of the sheath 44 exists, as shown in FIG. A three-dimensional structure image is generated using the three-dimensional structure data obtained from the scattered light intensity data of the sheath 44 as described above. The process for removing the scattered light intensity data of the sheath 44 is as described above.

(1) “Endoscope mode”
The endoscope mode is a mode in which a three-dimensional image is displayed as if the measurement region is viewed with an endoscope inserted into the lumen of the biliary tract / pancreatic duct. In the display screens of FIGS. This corresponds to a three-dimensional image displayed on the image display unit 304. In this endoscope mode, the projection center 500 is set on (or in the vicinity of) the central axis 400 and the projection plane 502 is set perpendicular to the central axis 400 with respect to the three-dimensional structure data as shown in FIG. Then, a projection (perspective projection or central projection) process is performed, and a three-dimensional image similar to the image photographed by the endoscope is generated and displayed as shown in FIG.

  In addition, when the diagnosis support function is executed, when an image in the endoscope mode is generated using the 3D structure data with the feature area information as the 3D structure data, the extracted features as shown in FIG. An area 504 is displayed with a color (corresponding color).

  In the endoscope mode, an image obtained by projecting only the surface of the three-dimensional structure data (surface rendering) is usually displayed. However, scattered light intensity data existing on the path of light rays from the projection center 500 to the projection plane 502 is displayed. Integrated projection that integrates (integrates) and projects, maximum value projection (MIP: maximum intensity projection) that projects the maximum value of scattered light intensity data existing on the ray path, or scattering that exists on the ray path You may enable it to switch and display the image produced | generated by the minimum value projection (MinIP: minimum intensity projection) which projects the minimum value of light intensity data. Further, the observer may be able to change the position of the projection center 500 and the position and orientation of the projection plane 502.

(2) "Perspective mode"
The strabismus mode is a mode for displaying a three-dimensional image in which the entire measurement area of the biliary tract and pancreatic duct is viewed from an oblique direction on the outside, and is displayed on the three-dimensional image display unit 304 in the display screens of FIGS. 19 and 20. It corresponds to a dimensional image. In this perspective mode, as shown in FIG. 25A, a projection surface 520 that is not parallel to the central axis 400 and not orthogonal to the three-dimensional structure data is set, and parallel projection processing is performed. As shown in B), a three-dimensional image in which the entire cylindrical three-dimensional structure data is viewed obliquely is generated and displayed. Further, in the perspective mode, the entire three-dimensional structure data is displayed semi-transparently (volume rendering processing or the like), and portions overlapping the three-dimensional structure data can be seen through.

  Further, when an image in the perspective mode is generated using the 3D structure data with the feature area information as the 3D structure data, the color (corresponding color) is extracted from the extracted feature area 504 as shown in FIG. ) Is displayed.

  Note that, similarly to the endoscope mode, in this perspective mode, there is an integrated projection for integrating and projecting scattered light intensity data existing on the path of the light beam onto the projection plane 520, or on the path of the light beam. Generated by maximum intensity projection (MIP), which projects the maximum value of scattered light intensity data, and by minimum intensity projection (MinIP), which projects the minimum value of scattered light intensity data existing on the ray path It is also possible to switch the displayed images. Further, the observer may be able to change the position and orientation of the projection plane 520.

(3) “Long-axis section mode”
The long-axis cross-sectional mode is a mode for displaying a tomographic image (AC tomographic image) of a cross-section obtained by re-slicing the measurement region of the biliary tract / pancreatic duct in the long-axis direction (C-axis direction) of the OCT probe 40. The display screen corresponds to an image displayed on the AC tomogram display unit 308. In this long-axis cross-sectional mode, the plane (AC plane) having the central axis 400 as one side is cut with respect to the three-dimensional structure data including scattered light intensity data of the sheath 44 (region 402) as shown in FIG. Set as surface 530. Then, the three-dimensional structure data (scattered light intensity data) of the AC cross section cut by the cut surface 530 is visualized, and an AC tomogram is displayed together with the scale as shown in FIG.

  In addition, when a long-axis cross-section mode image is generated using the three-dimensional structure data with feature region information as the three-dimensional structure data, the feature region display unit 540 is displayed below the AC tomogram as shown in FIG. Is displayed, and a color (corresponding color) is added to the position corresponding to the feature region. Further, the position of the cutting plane 530 may be automatically detected and set at the position of the cross section having the most characteristic region in the entire circumference, or a function of automatically moving the cutting plane 530 is provided. Also good.

  Even in this long-axis cross-sectional mode, not only the scattered light intensity data in the cross section but also the image obtained by integrating and visualizing the scattered light intensity data in the direction perpendicular to the cross section, and the maximum of the scattered light intensity data in the cross section. It may be possible to switch and display an image obtained by visualizing only the value or the minimum value. Further, the observer may be able to change the position of the cut surface 530 (AC cross section) in the B-axis direction, the range displayed on the monitor 100 as an AC tomographic image, the display magnification, and the like.

(4) "Surface mode"
The surface mode is a mode for displaying a BC cross-sectional image of a cross-section obtained by re-slicing the surface (outermost surface) of the inner wall portion of the measurement region of the biliary tract / pancreatic duct or the vicinity thereof. In FIG. Corresponds to the image. In this surface mode, as shown in FIG. 27A, voxels (BC) arranged at the same position in the A-axis direction (depth direction) with respect to the three-dimensional structure data including scattered light intensity data of the sheath 44 (region 402). The scattered light intensity data of the cross-section voxels) is developed as voxel values of voxels on the same plane. In addition, scale conversion is performed so that the length in the B-axis direction at each position in the A-axis direction becomes the reference length, and rectangular parallelepiped solid structure data as shown in FIG. 27B is generated. That is, in the cylindrical three-dimensional structure data in FIG. 27A, each line in the A-axis direction passing through each position on the BC plane is defined as a plane, the BC plane is a plane, and each line in the A-axis direction is parallel to each line. The arrangement of the scattered light intensity data is the rectangular parallelepiped three-dimensional structure data shown in FIG.

  Subsequently, by detecting the position of the outermost surface of the inner wall portion (region 404) and flattening the outermost surface, the scattered light intensity data of the voxels arranged on the line in the A-axis direction at each position of the BC plane is converted to the A-axis. As shown in FIG. 27C, the scattered light intensity data on the outermost surface is arranged in the voxel of the BC cross section at a predetermined position in the A-axis direction. Then, the scattered light intensity data in the BC cross section or in the vicinity (inner wall portion side) of the BC cross section 550 is visualized, and a BC tomogram is displayed as shown in FIG.

  Further, when the surface mode image is generated using the 3D structure data with the feature area information as the 3D structure data, the color (corresponding color) is displayed on the extracted feature area 504 as shown in FIG. ) Is displayed.

  In addition, the process which detects the outermost surface of an inner wall part with respect to the cylindrical three-dimensional structure data of FIG. 27 (A), expand | deploys the detected outermost surface to a plane, and visualized the scattered light intensity data of the outermost surface May be displayed. According to this, when a stenosis part exists in the lumen in the measurement region and the outermost surface having the shape as shown in FIG. 28A is detected, an image as shown in FIG. 27B is displayed. Therefore, the stenosis site can be grasped instantly. When the 3D structure data with feature area information is used as the 3D structure data, the feature area 504 is displayed with a color (corresponding color) as shown in FIG.

  Also in the surface mode, only the maximum value or the minimum value of the scattered light intensity data on the surface or the image obtained by integrating and visualizing the scattered light intensity data in the direction orthogonal to the tube inner wall surface (cross section) was visualized. Images may be switched and displayed. The observer may be able to change the position of the BC cross section 550 in the A axis direction, the range displayed on the monitor 100 as a BC cross section image, the display magnification, and the like.

  DESCRIPTION OF SYMBOLS 1 ... OCT apparatus, 30 ... OCT interferometer, 40 ... OCT probe, 44 ... Sheath (probe outer cylinder), 52 ... Optical lens, 68A, 68B ... Marker, 90 ... Arithmetic processing unit, 92 ... Input device, 100 ... Monitor 110 ... X-ray fluoroscopic apparatus, 200 ... Endoscope, 210 ... Guide wire

Claims (10)

A light intensity data acquisition means for acquiring light intensity data of the inner wall portion obtained by optical coherence tomography measurement on the inner wall portion inside the living body having a layer structure;
Based on the light intensity data acquired by the light intensity data acquisition means, the light intensity data of a predetermined range at the same depth is added with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall. Adding means;
When the added value calculated by the adding means shows a predetermined change along the depth direction, detecting means for detecting the measurement point as a lesioned part,
A lesion portion display means for displaying information indicating the lesion portion detected by the detection means on an image obtained by visualizing the light intensity data;
A diagnostic support apparatus comprising:   The detection means detects the measurement point as a lesion when the addition value calculated by the addition means is in a range that can be regarded as being attenuated uniformly along the depth direction. The diagnosis support apparatus according to claim 1.   The detection means detects a measurement point in which the addition value calculated by the addition means shows a predetermined change along the depth direction, and forms a region having a predetermined size or more in which the set of measurement points is continuous. The diagnosis support apparatus according to claim 1, wherein the region is detected as a lesion part.   The diagnosis support apparatus according to any one of claims 1 to 3, wherein the lesion part display means displays the lesion part on the image obtained by visualizing the light intensity data with a predetermined color. .   The diagnosis support apparatus according to claim 1, wherein the light intensity data acquisition unit acquires light intensity data of a three-dimensional area of the inner wall portion.   The lesion display means displays at least one of a three-dimensional image by a perspective projection process, a three-dimensional image by a parallel projection process, and a tomographic image at a predetermined section as an image obtained by visualizing the light intensity data. The diagnosis support apparatus according to any one of claims 1 to 5.   The diagnostic support according to any one of claims 1 to 6, wherein the inner wall is the inner wall of the biliary tract, pancreatic duct, bronchus, pharynx, esophagus, stomach, large intestine, uterus, or ureter. apparatus. A light intensity data acquisition step of acquiring light intensity data of the inner wall part obtained by optical coherence tomography measurement on the inner wall part inside the living body having a layer structure;
Based on the light intensity data acquired in the light intensity data acquisition step, light intensity data in a predetermined range at the same depth is added with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall. Adding step;
When the addition value calculated by the addition step indicates a predetermined change along the depth direction, a detection step of detecting the measurement point as a lesion part;
A lesion display step for displaying information indicating the lesion detected by the detection step on the image obtained by visualizing the light intensity data;
A diagnostic support method characterized by comprising: A light intensity data acquisition means for acquiring light intensity data of the inner wall portion obtained by optical coherence tomography measurement on the inner wall portion inside the living body having a layer structure;
Based on the light intensity data acquired by the light intensity data acquisition means, the light intensity data of a predetermined range at the same depth is added with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall. Adding means;
When the added value calculated by the adding means shows a predetermined change along the depth direction, detecting means for detecting the measurement point as a lesioned part,
A lesion detection apparatus comprising: A light intensity data acquisition step of acquiring light intensity data of the inner wall part obtained by optical coherence tomography measurement on the inner wall part inside the living body having a layer structure;
Based on the light intensity data acquired in the light intensity data acquisition step, light intensity data in a predetermined range at the same depth is added with respect to the depth direction of the inner wall from the measurement point set on the surface of the inner wall. Adding step;
When the addition value calculated by the addition step indicates a predetermined change along the depth direction, a detection step of detecting the measurement point as a lesion part;
A lesion detection method characterized by comprising: