JP6259747B2 - Processor device, endoscope system, operating method of processor device, and program - Google Patents

Description

  The present invention relates to a processor device, an endoscope system, an operating method of a processor device, and a program that support lesion diagnosis based on minute blood vessels and gland ducts.

  In recent medical treatments, diagnosis and the like using an endoscope system including a light source device, an electronic endoscope, and a processor device are widely performed. For observation in the specimen using an endoscope system, in addition to screening observations that detect possible lesions such as brownish areas and redness from a distant view, detection of such lesions is also possible. When the zoom lens is used, a magnified observation is performed in which the likely lesion site is enlarged and examined.

  In recent years, “VS classification” has been performed in which the diagnosis is focused on a fine mucosal structure. In this “VS classification”, a fine blood vessel pattern and a gland duct pattern on the surface layer of the mucosa are independently observed to determine whether or not a “normal” pattern is met. If both the vascular pattern and the duct pattern are “normal” patterns, it is determined that the lesion is not a lesion. If either the vascular pattern or the duct pattern is not “normal”, the possibility of a lesion is determined. It is judged that is high.

  In this VS classification, it is difficult to determine whether or not the blood vessel pattern and the gland duct pattern are “normal” patterns unless the doctor is skilled in pattern observation. Therefore, in Patent Document 1, a blood vessel pattern, a gland duct pattern, and a template pattern in an image obtained during endoscopic observation are subjected to pattern matching, and diagnosis support is performed using the pattern matching processing result. Information is displayed on a monitor or the like. By using the method of Patent Document 1, it is considered that the lesion can be diagnosed reliably without depending on the skill level of the doctor.

Patent 5435746

  However, in Patent Document 1, when the image quality of an image obtained during endoscopic observation is not good, pattern matching with a template pattern may not be reliably performed. In addition, since there are a wide variety of blood vessel patterns and gland duct patterns, a large number of templates required for matching are required to increase the accuracy of pattern matching. Therefore, it has been demanded that diagnosis of a lesion part based on a blood vessel and a duct can be supported without using pattern matching based on a blood vessel pattern or a duct pattern of the template.

  It is an object of the present invention to provide a processor device, an endoscope system, an operation method of a processor device, and a program that support diagnosis of a lesion site based on blood vessels and gland ducts.

  The processor device of the present invention includes an image signal acquisition unit that acquires an image signal obtained by imaging an observation target, a blood vessel extraction unit that extracts a blood vessel from the image signal and generates a blood vessel image signal, and a gland from the image signal. A duct extraction unit that extracts a duct and generates a duct image signal, an index calculation unit that calculates a blood vessel index from the blood vessel image signal, and calculates a duct index from the duct image signal, and a blood vessel index and a duct index And an image generation unit that generates an image for display for displaying information based on the display unit.

  Preferably, the image generation unit generates a display image for displaying the blood vessel index and the gland duct index on the display unit as information based on the blood vessel index and the gland duct index. It is preferable to have a pattern determination unit that determines a blood vessel pattern related to the shape and running of the blood vessel from the blood vessel index and determines a gland duct pattern related to the shape and running of the gland duct from the gland duct index. The blood vessel pattern preferably includes “normal”, “abnormal”, and “disappearance”, and the gland duct pattern preferably includes “normal”, “abnormal”, and “disappearance”. The image generation unit preferably generates a display image for displaying the determination result of the pattern determination unit as information based on the blood vessel index and the gland duct index.

  It is preferable to have a first lesion determination unit that determines a lesion when at least one of a blood vessel pattern and a gland duct pattern is “abnormal” or “disappear” in the determination result of the pattern determination unit. It is preferable to have a boundary calculation unit that calculates a lesion boundary from the distribution of ductal index and a region of interest setting unit that sets a region including the lesion boundary as a region of interest. The image generation unit preferably generates a display image for displaying at least one of the determination result in the first lesion determination unit and the region of interest on the display unit as information based on the blood vessel index and the gland duct index. .

  The blood vessel pattern includes “normal”, “abnormal 1”, “abnormal 2”, “disappearance”, and the gland duct pattern includes “normal”, “abnormality 1”, “abnormality 2”, “disappearance” It is preferred that A color tone calculation unit that obtains a color tone from an image signal and a determination result of the pattern determination unit that a lesion is present when at least one of a blood vessel pattern and a gland duct pattern is “abnormal 1”, “abnormal 2”, or “disappear” It is preferable to have a second lesion determination unit that determines and determines the lesion type according to the color tone. It is preferable that the color tone is represented by “color tone 1”, “color tone 2”, and “color tone 3” that have different colors or brightness relationships with the surroundings. The image generation unit preferably generates a display image for displaying the determination result of the second lesion determination unit on the display unit as information based on the blood vessel index and the gland duct index.

  The blood vessel index is at least one of the density of the blood vessel, the correlation coefficient between the normal blood vessel pattern and the blood vessel, the traveling direction of the blood vessel, and the frequency component of the blood vessel. The gland duct index is the density of the gland duct, Preferably, at least one of the correlation coefficient between the duct pattern and the duct, the traveling direction of the duct, and the frequency component of the duct.

  The blood vessel extraction unit extracts a region including a pixel having a smaller pixel value than the surrounding pixels in the image signal as the blood vessel pixel to generate a blood vessel image signal, and the gland duct extraction unit includes the image signal in the image signal. Thus, it is preferable to extract a region including a pixel having a pixel value larger than that of surrounding pixels as a gland duct pixel to generate a gland duct image signal. The blood vessel extraction unit generates a blood vessel image signal by performing black hat filter processing on the image signal, and the gland duct extraction unit performs top hat filter processing on the image signal. Is preferably generated.

  The endoscope system including the processor device of the present invention described above includes a light source unit that emits light including short-wave light or short-wave light as light for illuminating an observation target. The short wavelength light is preferably narrow band light including a wavelength range of 400 to 420 nm.

  The operation method of the endoscope system according to the present invention includes a step in which an image signal acquisition unit acquires an image signal obtained by imaging an observation target, and a blood vessel extraction unit extracts a blood vessel from the image signal to obtain a blood vessel image. A signal generating step, a duct extracting unit extracting a duct from the image signal to generate a duct image signal, and an index calculating unit calculating a blood vessel index from the blood vessel image signal, A step of calculating a duct index from the signal, and an image generating unit generating a display image for displaying information based on the blood vessel index and the duct index on the display unit.

  The program according to the present invention includes a computer, an image signal acquisition unit that acquires an image signal obtained by imaging an observation target, a blood vessel extraction unit that extracts a blood vessel from the image signal and generates a blood vessel image signal, and an image signal A gland duct extraction unit that extracts a gland duct from the gland duct, generates a gland duct image signal, calculates a blood vessel index from the blood vessel image signal, calculates a gland duct index from the gland duct image signal, a blood vessel index and a gland It is made to function as an image generation part which produces | generates the image for a display for displaying the information based on a pipe | tube index on a display part.

  ADVANTAGE OF THE INVENTION According to this invention, the diagnosis of the lesion part based on a blood vessel and a gland duct can be supported, without using the pattern matching based on the blood vessel pattern and gland duct pattern of a template.

1 is an external view of an endoscope system according to a first embodiment. It is a block diagram which shows the function of the endoscope system of 1st Embodiment. It is a graph which shows the emission spectrum of purple light V, blue light B, green light G, and red light R. 3 is a graph showing an emission spectrum of purple light V. It is a block diagram which shows the function of a special image process part. It is explanatory drawing which shows the function of the blood vessel extraction part. It is explanatory drawing which shows the function of a duct extraction part. It is an image figure which shows the special image for a display which displays an observation object, a blood vessel index, and a gland duct index. It is a block diagram which shows the function of the special image process part to which the pattern determination part was added. It is an image figure which shows the special image for a display which displays the observation object and the determination result in a pattern determination part. It is a block diagram which shows the function of the special image process part to which the pattern determination part, the 1st lesion determination part, the boundary part calculation part, and the region-of-interest setting part were added. It is an image figure which shows the special image for a display which displays an observation object, the determination result in a 1st lesion determination part, and the region of interest ROI. It is a block diagram which shows the function of the special image process part to which the pattern determination part, the 2nd lesion determination part, and the color tone calculation part were added. It is an image figure which shows the special image for a display which displays the observation object, the determination result in a 2nd lesion determination part, and the region of interest ROI. It is a flowchart which shows a series of flows of this invention. It is a block diagram which shows the function of the endoscope system of 2nd Embodiment. It is a graph which shows the emission spectrum of white light. It is a graph which shows the emission spectrum of special light. It is a graph which shows the spectrum of a blue-violet laser beam. It is a block diagram which shows the function of the endoscope system of 3rd Embodiment. It is a top view which shows a rotation filter. It is a figure which shows the function of the capsule endoscope system of 4th Embodiment. 4 is a graph showing emission spectra of purple light V, blue light B, green light G, and red light R different from those in FIG. 3.

[First Embodiment]
As shown in FIG. 1, the endoscope system 10 according to the first embodiment includes an endoscope 12, a light source device 14, a processor device 16, a monitor 18, and a console 19. The endoscope 12 is optically connected to the light source device 14 and electrically connected to the processor device 16. The endoscope 12 includes an insertion portion 12a to be inserted into a subject, an operation portion 12b provided at a proximal end portion of the insertion portion 12a, a bending portion 12c and a distal end portion 12d provided at the distal end side of the insertion portion 12a. have. By operating the angle knob 12e of the operation unit 12b, the bending unit 12c performs a bending operation. With this bending operation, the tip 12d is directed in a desired direction.

  In addition to the angle knob 12e, the operation unit 12b includes a mode switching SW 13a, a diagnostic information display SW 13b, and a zoom operation unit 13c. The mode switching SW 13a is used for switching operation between two types of modes, a normal observation mode and a special observation mode. The normal observation mode is a mode for displaying a normal image on the monitor 18. The special observation mode is a mode for displaying a special image on the monitor 18. When the diagnostic information display SW 13b is pressed in the special image, diagnostic information for supporting diagnosis of a lesioned part such as an index related to a blood vessel or a gland duct is displayed on the monitor 18 together with the special image. The zoom operation unit 13c enlarges or reduces the observation target being displayed on the monitor 18 by moving the zoom lens 47 (see FIG. 2) between the tele position and the wide position. Even in the normal observation mode, the diagnostic information may be displayed on the monitor 18 when the diagnostic information display SW 13b is pressed.

  The processor device 16 is electrically connected to the monitor 18 and the console 19. The monitor 18 outputs and displays image information and the like. The console 19 functions as a UI (User Interface) that receives input operations such as function settings. The processor device 16 may be connected to an external recording unit (not shown) for recording image information and the like.

  As shown in FIG. 2, the light source device 14 includes a V-LED (Violet Light Emitting Diode) 20a, a B-LED (Blue Light Emitting Diode) 20b, a G-LED (Green Light Emitting Diode) 20c, and an R-LED (Red). Light Emitting Diode) 20d, a light source control unit 21 that controls driving of the four color LEDs 20a to 20d, and an optical path coupling unit 23 that couples the optical paths of the four color lights emitted from the four color LEDs 20a to 20d. . The light coupled by the optical path coupling unit 23 is irradiated into the subject through a light guide (LG (Light Guide)) 41 and an illumination lens 45 inserted into the insertion unit 12a. An LD (Laser Diode) may be used instead of the LED. The “light source unit” of the present invention corresponds to the four-color LEDs 20a to 20d.

  As shown in FIG. 3, the V-LED 20a generates purple light V having a center wavelength of 405 ± 10 nm and a wavelength range of 380 to 420 nm. The B-LED 20b generates blue light B having a center wavelength of 460 ± 10 nm and a wavelength range of 420 to 500 nm. The G-LED 20c generates green light G having a wavelength range of 480 to 600 nm. The R-LED 20d generates red light R having a center wavelength of 620 to 630 nm and a wavelength range of 600 to 650 nm. In the present invention, “narrowband light including a wavelength range of 400 to 420 nm” corresponds to the violet light V.

  The light source control unit 21 lights the V-LED 20a, the B-LED 20b, the G-LED 20c, and the R-LED 20d in any of the normal observation mode and the special observation mode. Accordingly, the observation target is irradiated with light in which four colors of light of purple light V, blue light B, green light G, and red light R are mixed. In the normal observation mode, the light source controller 21 controls the LEDs 20a to 20d so that the light quantity ratio among the violet light V, blue light B, green light G, and red light R is Vc: Bc: Gc: Rc. Control. On the other hand, in the first and second special observation modes, the light source control unit 21 is configured so that the light quantity ratio among the violet light V, the blue light B, the green light G, and the red light R is Vs: Bs: Gs: Rs. Each LED 20a-20d is controlled.

  In the special observation mode, as shown in FIG. 4, the blue light B, the green light G, and the red light R may be turned off and only the purple light V may be emitted. By thus illuminating only the purple light V on the observation target, the contrast of blood vessels and gland ducts can be improved. Thereby, the calculation accuracy of the blood vessel index and gland duct index, which will be described later, can be increased, and the accuracy of lesion determination based on the blood vessel index and gland duct index can also be increased.

  As shown in FIG. 2, the light guide 41 is built in the endoscope 12 and the universal cord (the cord connecting the endoscope 12, the light source device 14, and the processor device 16). The combined light propagates to the distal end portion 12d of the endoscope 12. A multimode fiber can be used as the light guide 41. As an example, a thin fiber cable having a core diameter of 105 μm, a cladding diameter of 125 μm, and a diameter of φ0.3 to 0.5 mm including a protective layer serving as an outer shell can be used.

  The distal end portion 12d of the endoscope 12 is provided with an illumination optical system 30a and an imaging optical system 30b. The illumination optical system 30 a has an illumination lens 45, and light from the light guide 41 is irradiated to the observation target via the illumination lens 45. The imaging optical system 30 b includes an objective lens 46, a zoom lens 47, and an imaging sensor 48. Reflected light from the observation target enters the image sensor 48 through the objective lens 46 and the zoom lens 47. As a result, a reflected image of the observation object is formed on the image sensor 48.

  The imaging sensor 48 is a color imaging sensor that captures a reflection image of a subject and outputs an image signal. The image sensor 48 is preferably a CCD (Charge Coupled Device) image sensor, a CMOS (Complementary Metal-Oxide Semiconductor) image sensor, or the like. The image sensor 48 used in the present invention is a color image sensor for obtaining RGB image signals of three colors of R (red), G (green), and B (blue), that is, an R pixel provided with an R filter. , A so-called RGB imaging sensor including a G pixel provided with a G filter and a B pixel provided with a B filter.

  The image sensor 48 is a so-called complementary color image sensor that includes complementary filters for C (cyan), M (magenta), Y (yellow), and G (green) instead of an RGB color image sensor. May be. When the complementary color imaging sensor is used, CMYG four-color image signals are output. Therefore, it is necessary to convert CMYG four-color image signals into RGB three-color image signals by complementary color-primary color conversion. . Further, the image sensor 48 may be a monochrome image sensor not provided with a color filter. In this case, the light source control unit 21 needs to turn on the blue light B, the green light G, and the red light R in a time-sharing manner, and add a synchronization process in the processing of the imaging signal.

  The image signal output from the image sensor 48 is transmitted to the CDS / AGC circuit 50. The CDS / AGC circuit 50 performs correlated double sampling (CDS (Correlated Double Sampling)) and automatic gain control (AGC (Auto Gain Control)) on an image signal which is an analog signal. The image signal that has passed through the CDS / AGC circuit 50 is converted into a digital image signal by an A / D converter (A / D (Analog / Digital) converter) 52. The A / D converted digital image signal is input to the processor device 16.

  The processor device 16 includes a receiving unit 53, a DSP (Digital Signal Processor) 56, a noise removing unit 58, an image processing switching unit 60, a normal image processing unit 62, a special image processing unit 64, and a video signal generating unit. 66. The receiving unit 53 receives a digital RGB image signal from the endoscope 12. The R image signal corresponds to a signal output from the R pixel of the imaging sensor 48, the G image signal corresponds to a signal output from the G pixel of the imaging sensor 48, and the B image signal is output from the B pixel of the imaging sensor 48. Corresponds to the signal to be transmitted.

  The DSP 56 performs various signal processing such as defect correction processing, offset processing, gain correction processing, linear matrix processing, gamma conversion processing, and demosaicing processing on the received image signal. In the defect correction process, the signal of the defective pixel of the image sensor 48 is corrected. In the offset process, the dark current component is removed from the RGB image signal subjected to the defect correction process, and an accurate zero level is set. In the gain correction process, the signal level is adjusted by multiplying the RGB image signal after the offset process by a specific gain. The RGB image signal after the gain correction process is subjected to a linear matrix process for improving color reproducibility. After that, brightness and saturation are adjusted by gamma conversion processing. The RGB image signal after the linear matrix processing is subjected to demosaic processing (also referred to as isotropic processing or pixel interpolation processing), and a signal of a color that is insufficient at each pixel is generated by interpolation. By this demosaic processing, all the pixels have RGB signals.

  The noise removing unit 58 removes noise from the RGB image signal by performing noise removal processing (for example, a moving average method, a median filter method, etc.) on the RGB image signal subjected to gamma correction or the like by the DSP 56. The RGB image signal from which noise has been removed is transmitted to the image processing switching unit 60. The “image signal acquisition unit” of the present invention corresponds to a configuration including the reception unit 53, the DSP 56, and the noise removal unit 58.

  When the mode switching SW 13a sets the normal observation mode, the image processing switching unit 60 transmits the RGB image signal to the normal image processing unit 62 and is set to the first or second special observation mode. In this case, the RGB image signal is transmitted to the special image processing unit 64.

  The normal image processing unit 62 performs color conversion processing, color enhancement processing, and structure enhancement processing on the RGB image signal. In the color conversion process, a 3 × 3 matrix process, a gradation conversion process, a three-dimensional LUT process, and the like are performed on the digital RGB image signal to convert the digital RGB image signal into a color converted RGB image signal. Next, various color enhancement processes are performed on the RGB image signal that has been subjected to the color conversion process. A structure enhancement process such as spatial frequency enhancement is performed on the RGB image signal that has been subjected to the color enhancement process. The RGB image signal subjected to the structure enhancement processing is input from the normal image processing unit 62 to the video signal generation unit 66 as an RGB image signal of a normal image for display.

  The special image processing unit 64 generates a special image for display, and when the diagnostic information display SW 13b is operated, a display including diagnostic information for supporting diagnosis of a lesion such as an index related to a blood vessel or a gland duct Generate special images for Details of the special image processing unit 64 will be described later. The RGB image signal of the special image for display generated by the special image processing unit 64 is input to the video signal generation unit 66. Note that the special image processing unit 64 may perform color conversion processing, color enhancement processing, and structure enhancement processing in the same manner as the normal image processing unit 62.

  The video signal generation unit 66 converts the RGB image signal of the display image input from the normal image processing unit 62 or the special image processing unit 64 into a video signal for display as an image that can be displayed on the monitor 18. Based on this video signal, the monitor 18 displays a normal image and a special image.

  As shown in FIG. 5, the special image processing unit 64 includes a switching unit 68, a blood vessel extraction unit 70, a gland duct extraction unit 72, an index calculation unit 74, and an image generation unit 76. When the diagnostic information display SW 13 b is not operated, the switching unit 68 transmits the RGB image signal from the image processing switching unit 60 only to the image generation unit 76. On the other hand, when the diagnostic information display SW 13 b is operated, the switching unit 68 transmits the RGB image signal from the image processing switching unit 60 not only to the image generation unit 76 but also to the blood vessel extraction unit 70 and the gland duct extraction unit 72. .

  The blood vessel extraction unit 70 performs black hat filter processing on the B image signal among the RGB image signals from the switching unit 68 to generate a blood vessel image signal obtained by extracting blood vessels from the B image signal. As shown in FIG. 6, the B image signal before being input to the blood vessel extraction unit 70 includes a gland duct having a pixel value larger than that of the surrounding pixels, in addition to the pixel of the blood vessel V having a pixel value smaller than that of the surrounding pixels. S pixels are included. By applying a black hat filter to the B image signal, a blood vessel image signal in which only pixels of the blood vessel V having a pixel value smaller than that of surrounding pixels is extracted can be obtained. On the other hand, the duct extraction unit 72 performs top hat filter processing on the B image signal among the RGB image signals from the switching unit 68. By applying this top hat filter, as shown in FIG. 7, a duct image signal in which only pixels of the duct S having a pixel value larger than that of surrounding pixels is extracted is obtained.

  In the blood vessel extraction unit 70, it is possible to generate a blood vessel image signal similar to the above by performing inversion processing after performing top hat filter processing on the B image signal. By performing the same top hat filter processing in the blood vessel extraction unit 70 and the gland duct extraction unit 72 in this way, it is possible to reduce the processing load. The duct extraction unit 72 can generate a duct image signal similar to the above by performing black hat filter processing on the B image signal and performing inversion processing.

  The index calculation unit 74 calculates a blood vessel index from the blood vessel image signal. Examples of the blood vessel index include the density of blood vessels per unit area in the blood vessel image signal, the blood vessel pattern on the blood vessel image signal and the correlation coefficient between the normal blood vessel pattern, the traveling direction of the blood vessel on the blood vessel image signal, and the blood vessel image signal. It is the frequency component of the upper blood vessel. Similarly, the index calculation unit 74 calculates a duct index from the duct image signal. Examples of the duct index include, for example, the density of the duct per unit area in the duct image signal, the correlation between the duct pattern on the duct image signal and the normal duct pattern, and the gland on the duct image signal. It is a frequency component of the duct on the traveling direction of the duct and the duct image signal.

  The traveling direction of the blood vessel on the blood vessel image signal and the traveling direction of the gland duct on the gland duct image signal can be obtained, for example, by applying a Gabor filter to the blood vessel image signal or the gland duct image signal. When this Gabor filter is used, when there are components in all directions, the blood vessel or gland duct is made to run randomly. Moreover, the frequency component of the blood vessel on the blood vessel image signal and the frequency component of the gland duct on the gland duct image signal can be obtained, for example, by performing Fourier transform on the blood vessel image signal or the gland duct image signal.

  The image generation unit 76 generates a special image for display based on the RGB image signal from the switching unit 68, the blood vessel index, and the gland duct index. As a special image for display, for example, as shown in FIG. 8, an image 78 to be observed such as a blood vessel V and a gland duct S is displayed in the center, and a blood vessel index and a gland duct index 79 are displayed in the upper right.

  It should be noted that based on the blood vessel index and gland duct index calculated by the index calculation unit 74, the blood vessel pattern related to the blood vessel shape and running and the gland duct shape and gland duct pattern related to running may be determined. In this case, as shown in FIG. 9, a pattern determination unit 80 is provided between the index calculation unit 74 and the image generation unit 76.

  For example, when the observation target is the stomach, the pattern determination unit 80 determines whether the blood vessel pattern corresponds to “normal”, “abnormal”, or “disappearance” from the blood vessel index, and from the glandular duct index. It is determined whether the duct pattern corresponds to “normal”, “abnormal”, or “disappearance”. These “normal”, “abnormal”, and “disappearance” correspond to “regular”, “irregular”, and “absent” of VS classification, respectively (for example, “NBI Endoscopy Atlas (edited by Manabu Muto / Kenji Yao) / Nano Sano) (published by Nankodo) ", pages 114 and 115). The determination result 82 in the pattern determination unit 80 is displayed on a special image for display together with an image 78 to be observed such as a blood vessel V and a gland duct S as shown in FIG.

  Specifically, the pattern determining unit 80 determines that the blood vessel pattern is “disappeared” when the blood vessel density is equal to or lower than a certain value or the frequency component of the blood vessel has only the power of the low frequency component. The pattern determination unit 80 also determines that the blood vessel running direction is “normal” when the duct density is not less than a certain value and significantly different from the normal value, and the correlation coefficient with the “normal” blood vessel pattern is not more than a certain value. If any one of the cases in which the traveling direction of the blood vessel pattern does not match, the blood vessel pattern is determined as “abnormal”. The pattern determining unit 80 determines that the blood vessel pattern is “normal” when neither the “disappearance” condition nor the “abnormal” condition is satisfied. Note that “disappearance”, “abnormality”, and “normal” of the ductal pattern are also determined by the same method as the blood vessel pattern.

  For example, when the observation target is the large intestine, the pattern determination unit 80 determines whether the blood vessel pattern corresponds to “normal”, “abnormal 1”, “abnormal 2”, or “disappear” from the blood vessel index. From the duct index, it is determined whether the duct pattern corresponds to “normal”, “abnormal 1”, “abnormal 2”, or “disappearance”. Among these, “abnormal 1”, “abnormal 2”, and “disappearance” are classified according to NICE (NBI International Colorectal Endoscopic classification) classification (for example, see pages 164 and 165 of the above-mentioned “NBI Endoscopic Atlas”). . The determination result in the pattern determination unit 80 when the observation target is the large intestine is displayed on the special image for display together with the image of the observation target such as the blood vessel V and the gland duct S as described above.

  The “abnormality 1” of the blood vessel pattern indicates “a lace-like isolated blood vessel that does not exist or travels across a lesion”. The “abnormality 2” in the blood vessel pattern indicates “a brown blood vessel surrounding the white gland duct structure”. The “disappearance” of the blood vessel pattern indicates “there is a region where the blood vessel has been divided or disappeared”. On the other hand, “abnormality 1” of the duct pattern indicates that “a dark or white spot having the same size exists or a similar duct structure does not exist”. “Abnormal 2” of the duct pattern indicates “an oval, tubular, or branched white duct structure surrounded by a brown blood vessel”. The “disappearance” of the duct pattern indicates “an intangible or disappeared duct pattern”.

  Note that it is possible to determine whether or not a lesion such as cancer exists using the determination result in the pattern determination unit 80 and to display the lesion region on the monitor 18 as a region of interest. For example, when the observation target is the stomach, as shown in FIG. 11, the first lesion determination unit 84, the region of interest setting unit 86 provided between the pattern determination unit 80 and the image generation unit 76, and an index A boundary calculation unit 88 provided between the calculation unit 74 and the region of interest setting unit 86 is used to determine a lesion and set a region of interest.

  The first lesion determination unit 84 determines that “the lesion does not exist” when both the blood vessel pattern and the gland duct pattern are “normal”. In this case, “no lesion exists” is displayed on the display special image. On the other hand, the lesion determination unit 84 determines that “the lesion is present” when either one of the blood vessel pattern and the gland duct pattern is “abnormal” or “disappear” (that is, not “normal”). To do. In this case, as shown in FIG. 12, a guidance display indicating that “a lesion may exist” is displayed on the special image for display.

  In addition, the boundary calculation unit 88 calculates a lesion boundary (demarcation line) indicating a boundary between a lesion and a non-lesion. In this boundary calculation unit 88, among the gland duct indices calculated by the index calculation unit 74, for example, the density distribution and / or shape distribution discontinuous points of the gland duct using the density distribution and / or shape distribution of the gland duct. Is calculated. The discontinuous point is a portion where the density and / or shape of the gland duct changes rapidly, and can be calculated by edge detection or the like. And the boundary part calculation part 88 connects the calculated discontinuous point, and produces | generates a closed curve. The generated closed curve becomes a lesion boundary. Then, the region of interest setting unit 86 sets the lesion boundary calculated by the boundary calculation unit 88 as a region of interest ROI (Region Of Interest) including the lesion. As shown in FIG. 12, the set region of interest ROI is displayed on a special image for display together with the image 90 to be observed.

  When the observation target is the large intestine, as shown in FIG. 13, lesion determination is performed using a color tone calculation unit 92 and a second lesion determination unit 94. The color tone calculation unit 92 receives the RGB image signal from the second switching unit 68, and calculates the color tone of the observation target based on the received RGB image signal. Further, the color tone calculation unit 92 sets the calculated color tone as “tone equal to or brighter than the surroundings” as the color tone 1, “brown compared to the surrounding colors” as the color tone 2, and “brown brighter than the surroundings” as the color tone 3 Classify as

なお、観察対象が大腸である場合においても、血管指標又は腺管指標が特定の条件を満たす領域を、病変が含まれる関心領域ＲＯＩ（Region Of Interest）として、表示用の特殊画像上に表示するようにしてもよい。 The second lesion determination unit 94 is provided between the pattern determination unit 80 and the image generation unit 76. The second lesion determination unit 94 determines that “the lesion does not exist” when both the blood vessel pattern and the gland duct pattern are “normal”. In this case, “no lesion exists” is displayed on the display special image. On the other hand, the second lesion determination unit 94 determines that “a lesion is present” when one of the blood vessel pattern and the gland duct pattern is “abnormal 1”, “abnormal 2”, or “disappearance”. Is determined. When it is determined that “the lesion is present” as described above, the second lesion determination unit 94 further adds the determination result of the pattern determination unit 80 and the color tone calculated by the color tone calculation unit 94, and the lesion Determine the type. The types of lesions are “type 1”, “type 2”, and “type 3”, and are classified according to Table 1 below.

Even when the observation target is the large intestine, a region where the blood vessel index or gland duct index satisfies a specific condition is displayed as a region of interest ROI (Region Of Interest) including a lesion on a special image for display. You may do it.

  Next, a series of flows of the present invention will be described with reference to the flowchart shown in FIG. First, screening is performed from a distant view state in the normal mode. At the time of this screening, when a site (possible site of lesion) such as a brownish area or redness is detected, the zoom operation unit 13c is operated to perform enlarged observation to display the site of possible lesion. . In accordance with this, the mode switching SW 13a is operated to switch to the special mode. As a result, the special image is displayed on the monitor 18.

  Then, during the display of the special image, the diagnostic information display SW 13b is operated in order to diagnose from the state of the blood vessel V and the gland duct S whether the possible lesion site is a lesion. In response to this operation, the blood vessel extraction unit 70 performs black hat filter processing on the B image signal of the special image, thereby generating a blood vessel image signal from which the blood vessel V is extracted. In addition, the duct extraction unit 72 performs a top hat filter process on the B image signal of the special image to generate a duct image signal from which the duct S is extracted. Then, the index calculation unit 74 calculates a blood vessel index from the blood vessel image signal and calculates a gland duct index from the gland duct image signal. The calculated blood vessel index and gland duct index are displayed on the monitor 18 together with the observation target including the blood vessel V and the gland duct S.

  As described above, the objective blood vessel index and the gland duct index such as the blood vessel density and the gland duct density related to the blood vessel V and the gland duct S are displayed on the monitor 18 during the enlarged observation, so that the blood vessel pattern and the gland duct pattern are displayed. Even a doctor who is not skilled in observation can surely diagnose a lesion.

[Second Embodiment]
In the second embodiment, the observation target is illuminated using a laser light source and a phosphor instead of the four-color LEDs 20a to 20d shown in the first embodiment. The rest is the same as in the first embodiment.

  As shown in FIG. 16, in the endoscope system 100 according to the second embodiment, in the light source device 14, a blue laser light source that emits blue laser light having a central wavelength of 445 ± 10 nm instead of the four color LEDs 20 a to 20 d (see FIG. 16). 16, a blue-violet laser light source (indicated as “405LD” in FIG. 16) 106 that emits blue-violet laser light having a central wavelength of 405 ± 10 nm is provided. Light emission from the semiconductor light emitting elements of these light sources 104 and 106 is individually controlled by the light source control unit 108, and the light quantity ratio between the emitted light of the blue laser light source 104 and the emitted light of the blue-violet laser light source 106 is freely changeable. It has become. In the present invention, “narrowband light including a wavelength range of 400 to 420 nm” corresponds to blue-violet laser light.

  The light source control unit 108 drives the blue laser light source 104 in the normal observation mode. In contrast, in the special observation mode, both the blue laser light source 104 and the blue violet laser light source 106 are driven, and the emission ratio of the blue laser light is made larger than the emission ratio of the blue violet laser light. I have control. The laser light emitted from each of the light sources 104 and 106 enters the light guide (LG) 41 through optical members (all not shown) such as a condenser lens, an optical fiber, and a multiplexer.

  Note that the full width at half maximum of the blue laser beam or the blue-violet laser beam is preferably about ± 10 nm. As the blue laser light source 104 and the blue-violet laser light source 106, a broad area type InGaN laser diode can be used, and an InGaNAs laser diode or a GaNAs laser diode can also be used. In addition, a light-emitting body such as a light-emitting diode may be used as the light source.

  In addition to the illumination lens 45, the illumination optical system 30a is provided with a phosphor 110 on which blue laser light or blue-violet laser light from the light guide 41 is incident. When the phosphor 110 is irradiated with the blue laser light, the phosphor 110 emits fluorescence. Some of the blue laser light passes through the phosphor 110 as it is. The blue-violet laser light is transmitted without exciting the phosphor 110. The light emitted from the phosphor 110 is irradiated into the specimen through the illumination lens 45. The “light source unit” of the present invention corresponds to a configuration including the blue laser light source 104, the blue-violet laser light source 106, and the phosphor 110.

  Here, in the normal observation mode, mainly the blue laser light is incident on the phosphor 110. Therefore, as shown in FIG. 17, the blue laser light and the fluorescence excited and emitted from the phosphor 110 are combined by the blue laser light. White light is irradiated to the observation target. On the other hand, in the special observation mode, since both the blue-violet laser beam and the blue laser beam are incident on the phosphor 110, the phosphor is generated by the blue-violet laser beam, the blue laser beam, and the blue laser beam as shown in FIG. Special light obtained by combining fluorescence excited and emitted from 110 is irradiated into the specimen.

The phosphor 110 absorbs a part of the blue laser light and emits a plurality of types of phosphors that emit green to yellow light (for example, YAG phosphors or phosphors such as BAM (BaMgAl ₁₀ O ₁₇ )). It is preferable to use what is comprised including. If a semiconductor light emitting element is used as an excitation light source for the phosphor 110 as in this configuration example, high intensity white light can be obtained with high luminous efficiency, and the intensity of white light can be easily adjusted, and the color of white light can be easily adjusted. Changes in temperature and chromaticity can be kept small.

  In the special observation mode, as shown in FIG. 19, the blue laser light may be turned off and only the blue-violet laser light may be emitted. By illuminating the observation target with only this blue-violet laser beam, the contrast of blood vessels and gland ducts can be improved. Thereby, the calculation accuracy of the blood vessel index and gland duct index described above can be increased, and the accuracy of lesion determination based on the blood vessel index and gland duct index can also be increased.

[Third Embodiment]
In the third embodiment, the observation target is illuminated using a broadband light source such as a xenon lamp and a rotary filter instead of the four-color LEDs 20a to 20d shown in the first embodiment. Further, instead of the color image sensor 48, the observation target is imaged by a monochrome image sensor. The rest is the same as in the first embodiment.

  As shown in FIG. 20, in the endoscope system 200 of the third embodiment, the light source device 14 is provided with a broadband light source 202, a rotary filter 204, and a filter switching unit 205 instead of the four color LEDs 20 a to 20 d. Yes. The imaging optical system 30b is provided with a monochrome imaging sensor 206 that is not provided with a color filter, instead of the color imaging sensor 48. The “light source unit” of the present invention corresponds to a configuration including the broadband light source 202 and the rotation filter 204.

  The broadband light source 202 is a xenon lamp, a white LED, or the like, and emits white light having a wavelength range from blue to red. The rotary filter 204 includes a normal observation mode filter 208 provided inside and a special observation mode filter 209 provided outside (see FIG. 21). The filter switching unit 205 moves the rotary filter 204 in the radial direction, and when the normal switching mode is set by the mode switching SW 13a, the normal observation mode filter 208 of the rotary filter 204 is inserted into the white light path. When the special observation mode is set, the special observation mode filter 209 of the rotation filter 204 is inserted into the optical path of white light.

  As shown in FIG. 21, the normal observation mode filter 208 includes, along the circumferential direction, a B filter 208a that transmits blue light of white light, a G filter 208b that transmits green light of white light, and white light. Among them, an R filter 208c that transmits red light is provided. Therefore, in the normal observation mode, the rotation filter 204 is rotated, so that blue light, green light, and red light are alternately irradiated on the observation target.

  The special observation mode filter 209 includes, in the circumferential direction, a B filter 209a that transmits special blue light in a band different from that of blue light and white light, and special green light in a band different from green light among white light. A G filter 209b that transmits light and an R filter 209c that transmits special red light in a band different from red light among white light are provided. Therefore, in the special observation mode, the rotation filter 204 is rotated so that special blue light, special green light, and special red light are alternately irradiated onto the observation target.

  In the endoscope system 200, in the normal observation mode, the inside of the sample is imaged by the monochrome imaging sensor 206 every time the observation target is irradiated with blue light, green light, and red light. Thereby, RGB image signals of three colors are obtained. Based on the RGB image signals, a normal image is generated by the same method as in the first embodiment.

  On the other hand, in the special observation mode, the inside of the specimen is imaged by the monochrome imaging sensor 206 each time the observation object is irradiated with special blue light, special green light, and special red light. Thereby, RGB image signals of three colors are obtained. A special image is generated based on an image signal of at least one color among the RGB image signals of these three colors. Other than that, a special image is generated in the same manner as in the first embodiment.

  In the special observation mode, only the special blue light may be emitted by being fixed to the B filter 209a of the rotary filter 204. In this case, only the narrow-band special blue light (corresponding to the “narrow-band light including the wavelength range of 400 to 420 nm” of the present invention) is illuminated on the observation target, so that the contrast of blood vessels and gland ducts can be reduced. Can be improved. Thereby, the calculation accuracy of the blood vessel index and gland duct index described above can be increased, and the accuracy of lesion determination based on the blood vessel index and gland duct index can also be increased.

[Fourth Embodiment]
In the fourth embodiment, instead of the insertion type endoscope 12 and the light source device 14, a swallowable capsule endoscope is used to acquire RGB image signals necessary for generating a normal image and a special image.

  As illustrated in FIG. 22, the capsule endoscope system 300 according to the fourth embodiment includes a capsule endoscope 302, a transmission / reception antenna 304, a capsule receiving device 306, a processor device 16, and a monitor 18. . The capsule endoscope 302 includes an LED 302a, an image sensor 302b, an image processing unit 302c, and a transmission antenna 302d. The processor device 16 is the same as that in the first embodiment, but in the fourth embodiment, a mode switching SW 308 for switching between the normal observation mode and the special observation mode and a diagnostic information display SW 310 are newly provided.

  The LEDs 302 a emit white light, and a plurality of LEDs 302 a are provided in the capsule endoscope 302. Here, as the LED 302a, it is preferable to use a white LED or the like that includes a blue light source and a phosphor that emits fluorescence by converting the wavelength of light from the blue light source. An LD (Laser Diode) may be used instead of the LED. White light emitted from the LED 302a is illuminated on the observation target. The “light source unit” of the present invention corresponds to the LED 302a.

  The imaging sensor 302b is a color imaging sensor, images an observation target illuminated with white light, and outputs RGB image signals. Here, as the imaging sensor 302b, a CCD (Charge Coupled Device) imaging sensor or a CMOS (Complementary Metal-Oxide Semiconductor) imaging sensor is preferably used. The RGB image signal output from the imaging sensor 302b is processed by the image processing unit 302c so as to be a signal that can be transmitted by the transmission antenna 302d. The RGB image signal that has passed through the image processing unit 302c is wirelessly transmitted from the transmission antenna 302d to the transmission / reception antenna 304.

  The transmission / reception antenna 304 is attached to the body of the subject and receives the RGB image signal from the transmission antenna 302d. The transmission / reception antenna 304 transmits the received RGB image signal to the capsule receiver 306 wirelessly. The capsule receiving device 306 is connected to the receiving unit 53 of the processor device 16 and transmits the RGB image signal from the transmitting / receiving antenna 304 to the receiving unit 53.

In the above embodiment, four colors of light having an emission spectrum as shown in FIG. 3 are used, but the emission spectrum is not limited to this. For example, as shown in FIG. 23, the green light G and the red light R have the same spectrum as that in FIG. 3, while the violet light V ^* has a center wavelength of 410 to 420 nm and the purple light V in FIG. Light having a wavelength range slightly longer than the wavelength may be used. Further, the blue light B ^* may be light having a central wavelength of 445 to 460 nm and a wavelength range slightly shorter than the blue light B in FIG.

  The present invention is applicable to various medical image processing apparatuses in addition to the processor apparatus incorporated in the endoscope system as in the first to third embodiments and the capsule endoscope system as in the fourth embodiment. It is possible to apply.

10, 100, 200 Endoscope system 70 Blood vessel extraction unit 72 Gland duct extraction unit 74 Index calculation unit 76 Image generation unit 80 Pattern determination unit 84 First lesion determination unit 86 Region of interest setting unit 88 Boundary unit calculation unit 92 Color tone calculation Unit 94 second lesion determination unit 300 capsule endoscope system

Claims (19)

An image signal acquisition unit for acquiring an image signal obtained by imaging an observation target;
A blood vessel extraction unit that extracts blood vessels from the image signal and generates a blood vessel image signal;
A duct extraction unit for extracting a duct from the image signal and generating a duct image signal;
A blood vessel index from the blood vessel image signal, an index calculation unit for calculating a gland duct index from the gland duct image signal;
An image generating unit that generates an image for display for displaying information based on the blood vessel index and the gland duct index on a display unit;
A processor device comprising:   The processor device according to claim 1, wherein the image generation unit generates a display image for displaying the blood vessel index and the gland duct index on a display unit as information based on the blood vessel index and the gland duct index.   The pattern determining unit according to claim 1, further comprising: a pattern determining unit that determines a blood vessel pattern related to the shape and running of the blood vessel from the blood vessel index and determines a gland duct pattern related to the shape and running of the gland duct from the gland duct index. Processor device.   The processor device according to claim 3, wherein the blood vessel pattern includes “normal”, “abnormal”, and “disappearance”, and the gland duct pattern includes “normal”, “abnormal”, and “disappearance”.   5. The processor device according to claim 3, wherein the image generation unit generates a display image for displaying a determination result in the pattern determination unit as information based on the blood vessel index and the gland duct index.   5. A first lesion determination unit that determines that a lesion is present when at least one of the blood vessel pattern and the gland duct pattern is “abnormal” or “disappear” in the determination result of the pattern determination unit. The processor device as described. A boundary calculation unit for calculating a lesion boundary from the distribution of the duct index;
The processor device according to claim 6, further comprising a region-of-interest setting unit that sets a region including the lesion boundary as a region of interest.   The image generation unit displays a display image for displaying at least one of the determination result in the first lesion determination unit and the region of interest on the display unit as information based on the blood vessel index and the gland duct index. The processor device according to claim 7 to be generated.   The blood vessel pattern includes “normal”, “abnormal 1”, “abnormal 2”, and “disappearance”. The gland duct pattern includes “normal”, “abnormal 1”, “abnormal 2”, “disappearance”. 4. The processor device according to claim 3, wherein: A color tone calculation unit for obtaining a color tone from the image signal;
When at least one of the blood vessel pattern and the gland duct pattern is “abnormal 1”, “abnormal 2”, “disappear” in the determination result of the pattern determination unit, it is determined that the lesion is present, and according to the color tone The processor device according to claim 9, further comprising a second lesion determination unit that determines a type of the lesion.   The processor according to claim 10, wherein the color tone is represented by “color tone 1”, “color tone 2”, and “color tone 3” having different colors or brightness relationships with surroundings.   The said image generation part produces | generates the image for a display for displaying the determination result in a said 2nd lesion determination part on a display part as information based on the said blood vessel index and the said gland duct index. The processor device as described.   The blood vessel index is at least one of the density of the blood vessel, a correlation coefficient between a normal blood vessel pattern and the blood vessel, a traveling direction of the blood vessel, and a frequency component of the blood vessel, The density of the ducts, the correlation coefficient between the duct pattern in the normal state and the ducts, the traveling direction of the ducts, and the frequency components of the ducts are at least one. The processor device as described. The blood vessel extraction unit extracts, as the blood vessel pixel, a region including a pixel having a pixel value smaller than that of surrounding pixels in the image signal, and generates the blood vessel image signal,
The gland duct extraction unit extracts a region including a pixel having a pixel value larger than that of surrounding pixels in the image signal as a pixel of the gland duct, and generates the gland duct image signal. The processor device according to claim 1. The blood vessel extraction unit generates the blood vessel image signal by performing black hat filter processing on the image signal,
15. The processor device according to claim 14, wherein the duct extraction unit generates the duct image signal by performing top hat filter processing on the image signal. An endoscope system comprising the processor device according to any one of claims 1 to 15,
An endoscope system having a light source unit that emits light including short wavelength light or light having the short wavelength as light for illuminating the observation target.   The endoscope system according to claim 16, wherein the short-wavelength light is narrow-band light including a wavelength range of 400 to 420 nm. An image signal acquisition unit acquiring an image signal obtained by imaging an observation target;
A blood vessel extraction unit extracting a blood vessel from the image signal to generate a blood vessel image signal;
A duct extraction unit to extract a duct from the image signal and generate a duct image signal;
An index calculator calculates a blood vessel index from the blood vessel image signal, and calculates a gland duct index from the gland duct image signal;
An image generating unit generating a display image for displaying information based on the blood vessel index and the gland duct index on a display unit;
A method for operating a processor device comprising: Computer
An image signal acquisition unit for acquiring an image signal obtained by imaging an observation target;
A blood vessel extraction unit that extracts blood vessels from the image signal and generates a blood vessel image signal;
A duct extraction unit for extracting a duct from the image signal and generating a duct image signal;
A blood vessel index from the blood vessel image signal, an index calculation unit for calculating a gland duct index from the gland duct image signal;
The program for functioning as an image generation part which produces | generates the image for a display for displaying the information based on the said blood vessel index and the said gland duct index on a display part.

Images