Electronic endoscopy system

Abstract

An electronic endoscope system includes an electronic endoscope having an imaging element that images a living tissue at a distal end portion and an ultrasonic probe that applies an ultrasonic wave to the living tissue to obtain an echo signal, an imaging image processor having an image processing section that generates an imaging image by processing an imaging signal output from the imaging element, and an ultrasonic image processor having an ultrasonic image processing section that generates an ultrasonic image by processing the echo signal output from the ultrasonic probe, a noise detection section that detects a noise component included in the echo signal and periodically generated at a threshold level or more set in advance, and a noise suppression section that performs a suppression process on the detected noise component.

Description

Technical Field

[0001] This invention relates to an electronic endoscope system for acquiring ultrasound images. Background Technology

[0002] Electronic endoscope systems are used for the observation and treatment of living tissue inside the human body. As an imager of living tissue, an electronic endoscope system, in addition to acquiring optical images of the subject using an imaging element, can also acquire ultrasound images (ultrasound computed tomography) using an ultrasound probe. A processor connected to the ultrasound endoscope is used for examination and diagnosis as an ultrasound diagnostic device. Hereinafter, an endoscope equipped with an imaging element and an ultrasound probe will be referred to as an ultrasound endoscope (or electronic endoscope).

[0003] An ultrasonic endoscope includes an imaging element and an ultrasonic probe. Within a flexible tube extending from the ultrasonic endoscope towards the processor, an imaging element is positioned at the front end of the insertion section, and an imaging signal transmission line connects to the connector for the processor, transmitting imaging signals via this line. Additionally, within the same flexible tube, an ultrasonic probe is positioned at the front end of the insertion section, and an ultrasonic signal transmission line connects to the connector for the processor, transmitting ultrasonic signals via this line.

[0004] During ultrasound examination and diagnosis, the processor powers the ultrasound probe, which sends ultrasound waves towards the living tissue and receives the reflected waves. The reflected waves received by the ultrasound probe are signaled as echo signals, which are then sent to the processor via an ultrasound signal transmission line. The processor then processes the signals to generate an ultrasound image.

[0005] In addition to a signal processing unit that processes data using transmitted signals (ultrasound signals and camera signals) from the ultrasound endoscope, and a control unit that controls image display, the processor also includes a switching power supply. The switching power supply generates and provides the necessary voltage to operate the various components of the ultrasound endoscope and the processor. The processor is connected to a display for showing camera and ultrasound images.

[0006] The ultrasound image obtained from the echo signal of the ultrasound probe and displayed on the monitor may contain noise components, such as noise generated in the ultrasound endoscope or processor, or external noise overlapping with the AC power supply. These noise components include, for example, noise caused by the switching of the power supply, or noise caused by mutual interference between the transmission lines. For example, in a flexible tube, because the camera signal transmission line and the ultrasound signal transmission line are located close to each other, the electrostatic or electromagnetic coupling between the transmission lines is enhanced, and the pulse control signal controlling the camera element, etc., interferes with the ultrasound probe or the ultrasound signal transmission line, introducing noise components into the echo signal.

[0007] In addition, ultrasound images sometimes contain noise unique to ultrasound, known as artifacts (virtual images that do not actually exist). After ultrasound waves are generated, the echoes reflected from within the biological body are received to obtain the echo signal. However, virtual images are generated due to side lobe artifacts, grating lobes, and multiple reflections, resulting in artifact noise. Furthermore, high-frequency noise generated by the switching power supply is superimposed on the received ultrasound signal, thus artifacts sometimes appear in ultrasound images created from diagnostic ultrasound images.

[0008] To address the noise components in these ultrasound images, Japanese Patent Application Publication No. 2014-003801 discloses an ultrasound image processor that can remove periodic noise caused by the operation of a DC / DC converter. It includes: a main converter that receives power from a power input section and outputs a constant voltage power supply; and multiple slave converters that receive the constant voltage power supply and output power to the circuitry constituting the ultrasound image processor. By synchronizing the switching operations of the main converter and the slave converters, spike noise is reduced.

[0009] Japanese Patent Application Publication No. 2019-076707 discloses a method for changing the switching frequency of a switching power supply by a preset variation range to suppress the increase of switching noise in ultrasound images. This is because when the switching frequency is an integer multiple of the pulse repetition frequency (that is, the frequency at which ultrasonic pulses are transmitted during scanning), switching noise caused by switching will appear in ultrasound images based on ultrasound image data generated by performing B (brightness) mode scanning or M (motion) mode scanning.

[0010] Japanese Patent Application Publication No. 2017-080040 discloses a method for noise reduction, which includes an image generation unit, a detection unit, and a control unit. When the detection unit detects a specific change in the output of an external device or an ultrasound probe in the time direction, the control unit responds to the detection of the specific change by displaying a reference image on the display unit, including a medical image at a position substantially the same as the ultrasound image. Summary of the Invention

[0011] The problem that the invention aims to solve

[0012] In electronic endoscope systems equipped with ultrasound endoscopes and processors, systems capable of extracting echo signals from noise and performing image processing with high precision are preferred for achieving high-accuracy diagnosis. Furthermore, to reduce the physical burden on patients undergoing intracavitary ultrasound endoscope insertion, future efforts will focus on further reducing the diameter of flexible tubes. While conventional transmission lines used for ultrasound and imaging signals have suppressed noise entry due to their narrow diameter, the shielding performance is forced to decrease, making it easier for noise components to be generated.

[0013] Furthermore, if external noise is superimposed on the power supply (AC power supply) driving the ultrasound endoscope and processor, the noise can be reduced to some extent by the power supply circuit and filtering circuit. However, if the noise is significant, or if EMI noise or other noises are superimposed on the echo signal through unexpected paths, it will cause noise components to appear in the ultrasound image, resulting in a decrease in image quality. To further improve clarity, it is also necessary to detect minute echo signals. How to suppress noise components in ultrasound images has become a major problem.

[0014] The purpose of this invention is to provide an electronic endoscope system that can effectively detect periodically generated noise components in ultrasound images and suppress noise components when acquiring ultrasound images using an ultrasound probe, thereby producing high-precision ultrasound images.

[0015] Technical solutions for solving the problem

[0016] One aspect of the present invention is an electronic endoscope system for acquiring ultrasound images, characterized by comprising: an electronic endoscope having an imaging element at its front end for imaging living tissue, and an ultrasound probe for imparting ultrasound waves to the living tissue to obtain an echo signal; an image processor having an image processing unit that processes the imaging signal output from the imaging element to generate an image; and an ultrasound image processor having an ultrasound image processing unit, a noise detection unit, and a noise suppression unit, wherein the ultrasound image processing unit processes the echo signal output from the ultrasound probe to generate an ultrasound image, the noise detection unit detects noise components in the echo signal that are periodically generated at or above a predetermined threshold level, and the noise suppression unit suppresses the detected noise components.

[0017] Preferably, the noise detection unit includes a periodic amplitude detection unit, which calculates the noise period generated periodically by the noise component and calculates the periodic amplitude based on the minimum and maximum values ​​of the noise period. The noise detection unit uses the noise period and the periodic amplitude to detect the noise component.

[0018] Preferably, the noise suppression unit performs gain modification processing, which changes the amplification gain value between the pixel value at the noise pixel location in the ultrasound image of the detected noise component and the pixels other than the noise pixel at the noise pixel location.

[0019] Preferably, the noise suppression unit includes a frequency band removal filter, which calculates a frequency band corresponding to the noise period based on the noise period and the period amplitude, and uses the frequency band to remove the noise component from the echo signal.

[0020] Preferably, the noise suppression unit performs a correction process, replacing the pixel value at the noise pixel location in the ultrasound image of the detected noise component with an interpolated pixel value generated based on the pixel values ​​of the surrounding pixels at the noise pixel location.

[0021] Preferably, the noise suppression unit changes at least one of the operating frequencies of the plurality of component devices of the camera image processor and the ultrasound image processor in order to separate the generation period of the noise component generated periodically by the echo signal.

[0022] Preferably, the operating frequency includes at least one of the ultrasonic frequency of the ultrasonic probe and the operating frequency of the imaging element.

[0023] The ultrasound image processor includes an A / D converter as a component device, which converts the echo signal of the analog signal into a digital signal through sampling;

[0024] Preferably, the operating frequency includes the sampling frequency of the echo signal in the A / D converter.

[0025] The ultrasound image processor includes a DC / DC converter as a component device. This DC / DC converter is a switching power supply used to drive the electronic endoscope, the camera image processor, and the ultrasound image processor.

[0026] Preferably, the operating frequency includes the switching frequency in the DC / DC converter.

[0027] The ultrasound image processor includes an AC / DC converter as a component device. The AC / DC converter is a switching power supply used to convert AC power received from an external commercial power source into DC power and to supply the DC power to the DC / DC converter.

[0028] Preferably, the operating frequency includes the switching frequency in the AC / DC converter.

[0029] The DC / DC converter and the AC / DC converter include a battery that supplies DC power.

[0030] Preferably, when the noise suppression unit detects the noise component, it switches from the DC / DC converter and the AC / DC converter to the battery.

[0031] The electronic endoscope system includes an electronic endoscope position measuring device, which uses magnetism to determine the position of the electronic endoscope when it is inserted into the body cavity.

[0032] The endoscope position measuring device includes a drive unit that winds a transmitting coil around the electronic endoscope and a positioning unit that determines the position of the electronic endoscope using a signal from the transmitting coil when the electronic endoscope is inserted into a body cavity.

[0033] Preferably, when the noise suppression unit detects the noise component, it changes the operating frequency of the transmitting coil or stops the generation of the magnetism.

[0034] The noise detection unit,

[0035] Using the noise-free ultrasound images and the noisy ultrasound images as learning data, a pre-learned inference model for the presence or absence of the noise component is created.

[0036] Preferably, the presence or absence of the noise component is determined by inputting the ultrasound image generated in the image processing unit into the noise component inference model.

[0037] The noise suppression unit creates a noise countermeasure inference model based on machine learning, relating the operational environment information of the electronic endoscope, the camera image processor, or the ultrasound image processor, the feature quantities of the ultrasound image, and information on the effectiveness of measures to suppress the noise components corresponding to the operational environment information. Preferably, when the noise detection unit detects the noise components in a newly generated ultrasound image, it extracts the feature quantities of the ultrasound image and inputs the operational environment information when detecting the noise components and the extracted feature quantities into the created noise countermeasure inference model to set effective noise suppression countermeasures.

[0038] The motion environment information includes the operating frequencies of the multiple component devices of the camera image processor and the ultrasound image processor.

[0039] Preferably, the operating frequency includes the ultrasonic frequency of the ultrasonic probe and the operating frequency of the imaging element, the sampling frequency of the echo signal in the A / D converter, the switching frequency in the DC / DC inverter and the AC / DC inverter, and the operating frequency of the transmitting coil of the electronic endoscope position measuring device.

[0040] Invention Effects

[0041] According to the electronic endoscope system, when using an ultrasonic probe to acquire ultrasonic images, it is possible to effectively detect periodically generated noise components in the ultrasonic images and suppress these noise components, thereby creating high-precision ultrasonic images. Attached Figure Description

[0042] Figure 1This is an example block diagram of the overall structure of an electronic endoscope system according to one embodiment.

[0043] Figure 2 This is an example diagram of an electronic endoscope with an ultrasonic probe used in an electronic endoscope system according to one embodiment.

[0044] Figure 3 This is a block diagram illustrating the schematic structure of an electronic endoscope and a camera image processor used in an electronic endoscope system according to one embodiment.

[0045] Figure 4 The imaging principle of ultrasound images in an electronic endoscope system is described in one embodiment.

[0046] Figure 5 This is an example diagram illustrating the signal processing before the echo signal is displayed as a B-mode ultrasound image when acquiring an ultrasound image.

[0047] Figure 6 It means Figure 5 The diagram shows an example of the echo signal after passing through an amplification circuit and an integration circuit in Mode A, as well as the brightness after brightness modulation.

[0048] Figure 7 This is an example diagram of a B-mode ultrasound image acquired by an electronic endoscope system according to one embodiment.

[0049] Figure 8 This is an example diagram showing the scanning position of the ultrasound beam and the noise generation part in a B-mode ultrasound image obtained by an electronic endoscope system according to one embodiment.

[0050] Figure 9 This is an example diagram showing the frequency of noise components appearing in B-mode ultrasound images obtained by an electronic endoscope system according to one embodiment.

[0051] Figure 10 This is a diagram illustrating an example of the power spectrum of periodic noise components appearing in an ultrasound image obtained from an electronic endoscope system according to one embodiment.

[0052] Figure 11 This is an example diagram showing the noise of multiple switching power supplies superimposed on the echo signal in an electronic endoscope system according to one embodiment.

[0053] Figure 12 This is an example diagram illustrating the noise factors of the electronic endoscope insertion section used in an electronic endoscope system according to one embodiment.

[0054] Figure 13 This is an example diagram of an echo signal formed by the superposition of periodically occurring noise components in an electronic endoscope system according to one embodiment.

[0055] Figure 14This is an example diagram of the noise suppression section of a band removal filter used in an electronic endoscope system according to one embodiment.

[0056] Figure 15 This is an example diagram showing the frequency characteristics of all filters used to filter echo signals in an electronic endoscope system according to one embodiment.

[0057] Figure 16 The image shows a digital echo signal and brightness example after the echo signal, including periodically generated noise, obtained by an electronic endoscope system according to one embodiment has been filtered by a frequency band removal filter.

[0058] Figure 17 An example diagram of a noise suppression unit in an electronic endoscope system according to one embodiment, which removes noise pixels from an ultrasound image obtained as a B-mode image and performs interpolation processing.

[0059] Figure 18 This is an example diagram of an electronic endoscope system according to one embodiment, which includes a switching power supply, a battery, and a switching device.

[0060] Figure 19A This is an example diagram of an electronic endoscope system according to one embodiment, in which the converter is used as a dedicated power source for charging a parallel-installed battery.

[0061] Figure 19B This is an example diagram of an electronic endoscope system according to one embodiment, in which the converter is used as a dedicated power source for charging a parallel-installed battery.

[0062] Figure 20 This is an example diagram of an electronic endoscope position measuring device used in an electronic endoscope system according to one embodiment.

[0063] Figure 21 This is an example diagram illustrating the process of learning and determining whether or not periodically generating noise components occur in an electronic endoscope system according to one embodiment.

[0064] Figure 22 An example diagram illustrating how AI derives effective noise mitigation strategies in an electronic endoscope system as one implementation.

[0065] Figure 23 This is an example diagram illustrating the workflow for noise correction in an electronic endoscope system according to one embodiment.

[0066] Figure 24 This is an example diagram illustrating the workflow of noise correction using artificial intelligence (AI) in an electronic endoscope system as one embodiment. Detailed Implementation

[0067] (The overall structure of an electronic endoscope system)

[0068] Figure 1 This is a block diagram illustrating an example of the overall configuration of an electronic endoscope system according to one embodiment. The electronic endoscope system 10 for acquiring ultrasound images includes: an electronic endoscope 12, a camera image processor 22, and an ultrasound image processor 30.

[0069] The electronic endoscope 12 includes: an illumination unit 14 for illuminating living tissue, an imaging element 16 for capturing images of the living tissue, a drive signal processing unit 18 for preprocessing signals captured by the imaging element 16, and an ultrasonic probe 20 for imparting ultrasound waves to the living tissue to obtain echo signals. The ultrasonic probe 20 is a phased array probe, in which multiple probe elements arranged in a predetermined direction output ultrasound waves, each of which outputs ultrasound waves at predetermined times to acquire echo signals in various directions.

[0070] The image signal of the living tissue is input from the imaging element 16 to the drive signal processing unit 18 at a predetermined frame period, and output to the system controller 102 and the imaging image processing unit 26 of the imaging image processor 22. The frame period is, for example, 1 / 30 second or 1 / 60 second.

[0071] The drive signal processing unit 18 then accesses the memory 92 and reads the inherent information of the electronic endoscope 12. The inherent information of the electronic endoscope 12 recorded in the memory 92 includes, for example, the number of pixels, sensitivity, operable frame rate, and model of the imaging element 16.

[0072] The camera image processor 22 includes a light source unit 24 that sends a light source to the illumination unit 14, and a camera image processing unit 26 that generates a camera image by processing the camera signal output from the camera element 16.

[0073] The ultrasound image processor 30 includes: a transceiver unit 38 that sends a drive signal to the ultrasound probe 20 and receives echoes; an ultrasound image processing unit 32 that processes the echo signal from the ultrasound probe 20 to generate an ultrasound image; a noise detection unit 34 that detects noise components in the echo signal that appear periodically at a preset threshold level or higher; and a noise suppression unit 36 ​​that suppresses the detected noise components.

[0074] The ultrasound image processing unit 32 generates a one-dimensional B-mode image along one direction by performing prescribed operations, for example, using brightness modulation as density image data, based on digital echo signals. Furthermore, the ultrasound image processing unit 32 creates a two-dimensional B-mode image by arranging multiple one-dimensional B-mode images in different directions, generated from echo signals obtained from the phased array ultrasound probe 20, in accordance with the phased array scanning along a prescribed azimuth direction. Furthermore, known techniques such as gain processing and contrast processing are used to perform image processing on the created image, and simultaneously, grayscale processing is performed corresponding to the display range of the image in the ultrasound image display unit 46.

[0075] The ultrasound image processor 30 includes a display unit 46 that can display generated ultrasound images and has an input function, allowing input via a touch panel. Furthermore, the ultrasound image processor 30 also includes an input unit 42, a power supply unit 44, and an AC power input unit 48 for operating the ultrasound image processor 30.

[0076] The input unit 42 accepts various types of information input using a keyboard, mouse, touch panel, etc. The ultrasound image display unit 46 displays various information, including the created ultrasound images. In addition to the ultrasound image processor 30, the power supply unit 44 provides power to drive the electronic endoscope 12 and the camera image processor 22. For example, as a component device, the power supply unit 44 includes a switching power supply, i.e., a DC / DC converter, and generates DC voltage through the switching frequency of the DC / DC converter. Multiple DC / DC converters are provided, each converting the input DC voltage to the required DC voltage and supplying power to each device.

[0077] The noise detection unit 34 detects whether the echo signal contains periodically generated noise components that exceed a preset threshold level. The method for detecting the noise components is not particularly limited; preferably, for example, it includes a period amplitude detection unit that calculates the noise period of the periodically generated noise components, calculates the period amplitude from the minimum and maximum values ​​of the noise period, and then uses the noise period and period amplitude to detect the noise components.

[0078] The noise detection unit 34 can further perform frequency analysis using a Fast Fourier Transform (FFT) to determine whether there are spectral peaks above a set threshold level. Echo signals reflected at the boundary of living tissue are unlikely to have spectral peaks because they are not periodic; therefore, spectral peaks are likely peaks of periodically occurring noise components.

[0079] The noise suppression unit 36 ​​is used to suppress detected noise components. Noise component suppression includes: when the echo signal is output through the ultrasonic probe 20, a process is implemented to modify the operation of each device in the electronic endoscope system 10 to suppress periodically occurring noise components; and interpolation processing is performed on the two-dimensional B-mode image created by the ultrasonic image processing unit 32. The process of modifying the operation of each device and the interpolation processing of the two-dimensional B-mode image will be described later.

[0080] (Electronic endoscope)

[0081] Figure 2 This is an example diagram of an electronic endoscope with an ultrasonic probe used in an electronic endoscope system according to one embodiment.

[0082] The electronic endoscope 12 has an operating part 52, a tip part 56, an insertion part 54 having a flexible part 58 mainly inside, a flexible cable 60 having an optical fiber inside, a scanner connector cable 62, a connector 64, and a scanner connector 66.

[0083] The tip 56 is a sensor for detecting living tissue, including an imaging element 68, a pop-out end face 70, and an ultrasonic probe 20. The ultrasonic probe 20 has multiple ultrasonic transducers as probe elements, for example, an array of transducers in which piezoelectric elements are arranged in an array. Each transducer transmits ultrasonic waves according to a drive signal, and simultaneously receives reflected waves from the subject and outputs an analog received signal. Each transducer is, for example, constructed using elements in which electrodes are formed at both ends of a piezoelectric body made of piezoelectric ceramic PZT (Lead Zirconate Titanate) or polymeric piezoelectric element PVDF (Poly Vinylidene Di Fluoride).

[0084] An objective lens and an image sensor 16 are provided on the imaging element section 68 at the tip end 56. The objective lens causes the reflected light from the living tissue illuminated by the illumination light to form an image on the light-receiving surface of the image sensor 16. The image sensor 16 is, for example, a single-plate color CCD (charge-coupled device) image sensor with a Bayer-type pixel arrangement. The single-plate color CCD image sensor stores the optical image formed by each pixel on the light-receiving surface as a charge corresponding to the light intensity, and generates and outputs image signals corresponding to the R (red), G (green), and B (blue) color components. The image sensor 16 is not limited to a CCD image sensor, but can also be a CMOS (complementary metal-oxide-semiconductor) image sensor or other types of imaging devices. The image sensor 16 can also be a device equipped with a complementary color filter.

[0085] The incident illumination light is emitted from the ejector end face 70 located at the tip end 56 through the light distribution lens. The illumination light is then incident on the illumination section 14 of the electronic endoscope 12 via the light guide 94.

[0086] The outer portion of the tip 56 is made of rigid resin. The imaging element section 68 is provided with an imaging element 16 and an objective lens, an illumination lens, etc. (not shown).

[0087] Additionally, the tip portion 56 includes a gas / water supply nozzle 72 for discharging or aspirating liquids or gases. The gas / water supply nozzle 72 is used to spray liquids, such as water, to clean the surfaces of the objective lens and illumination lens associated with the imaging element 16, and to spray gases, such as air, to remove residual liquids and foreign matter from the surfaces of the objective lens and illumination lens. The tip portion 56 also includes a balloon (not shown) for ultrasound diagnosis by injecting liquid into it and bringing it into contact with biopsy tissue, and has a balloon inlet 88 and a balloon aspiration outlet 90. Furthermore, the tip portion 56 also includes a forceps lifter 76 for contacting a flexible puncture needle (not shown) with biopsy tissue, and an opening for aspirating liquids or gases from the biopsy tissue through the forceps lifter 76.

[0088] The insertion part 54 is provided with a curved part 78 that bends in the vertical and horizontal directions. The portion from the base end (to the side of the operation part 52) ​​to the distance from the curved part 78 is a flexible part 58 that can bend according to its own weight and the operator's manipulation.

[0089] The flexible section 58 is located between the bending section 78 and the operating section 52, and inside it are sensor signal lines located at the tip 56, as well as multiple independent flow paths through which gas or fluid flows from the opening. These independent flow paths consist of tubes, pipes, or elongated holes.

[0090] On one side of the operating section 52 and the tip 56, a flexible treatment instrument insertion port protrusion 80 for inserting a puncture needle and a forceps lifting guidewire cleaning port 74 are provided. A removable cover is fitted to the end opening of the treatment instrument insertion port protrusion 80. Inside the insertion section 54, a treatment instrument insertion and suction tube extends from the treatment instrument insertion port protrusion 80 toward the tip 56 and is flexible. The treatment instrument insertion and suction tube opens on the forceps lifting platform 76. A puncture needle inserted from the treatment instrument insertion port protrusion 80 into the treatment instrument insertion and suction tube can protrude outward from the tip opening of the treatment instrument insertion and suction tube located on the forceps lifting platform 76, protruding from the tip opening for palpating biopsies.

[0091] The operating unit 52 has multiple operating buttons 84 for flow path switching, and internally provides a common flow path for fluid flow extending from the connector 64 into the flexible cable 60. The bending operating lever 82 is a lever operated by a physician for bending the bending section 78 in the vertical and horizontal directions. The bending section 78 bends vertically and horizontally according to the rotation operation of the bending operating lever 82.

[0092] The flexible cable 60 connects the connector 64, which is connected to the camera image processor 22, to the operation unit 52. The connector 64 is also provided with an open port for a common flow path for supplying or drawing in fluid.

[0093] Connector 64 has a light source insertion portion 86 and is connected to the camera image processor 22. Illumination light generated by the light source unit located in the camera image processor 22 is transmitted from connector 64 to tip 56 via flexible cable 60, operation portion 52, and optical guide cable within insertion portion 54. Furthermore, drive signals are transmitted from connector 64 to image sensor 16 via signal lines from camera image processor 22 to flexible cable 60. Image signals captured by image sensor 16 are transmitted to camera image processor 22 via flexible cable 60, operation portion 52, and signal lines within insertion portion 54.

[0094] The scanner connector 66 connects to the ultrasound image processor 30, transmitting the echo signal scanned by the ultrasound probe 20 to the ultrasound image processing unit 32 via the scanner connector cable 62. The ultrasound image processing unit 32 processes the echo signal, generates a biopsy image of the examined object, and displays the generated image on the display unit 46. Furthermore, the scanner connector cable 62 sends a drive signal from the ultrasound image processor 30 to the piezoelectric element of the ultrasound probe 20. The piezoelectric element converts electrical energy into mechanical energy and generates ultrasound waves through expansion and contraction due to voltage changes.

[0095] Connector 64 is connected to drive signal processing unit 18. Image signals of the living tissue are input from imaging element 16 to drive signal processing unit 18 at a predetermined frame period, and output to system controller of imaging image processor 22 and imaging image processing unit 26. The frame period is, for example, 1 / 30 second or 1 / 60 second.

[0096] (Processor for electronic endoscopes)

[0097] Figure 3 This is a block diagram illustrating the schematic structure of an electronic endoscope and a camera image processor used in an electronic endoscope system according to one embodiment. The camera image processor 22 includes a system controller 102 and a timing controller 100 for control. The system controller 102 executes various programs stored in the memory 104 and centrally controls the entire electronic endoscope system 10. Furthermore, the system controller 102 is connected to an operation panel 97.

[0098] The system controller 102 is used to change various operations of the electronic endoscope system 10 and the parameters for each operation according to instructions from the operator input from the operation panel 97. Operator input instructions include instructions for switching operating modes of the electronic endoscope system 10. According to one embodiment, the operating modes are divided into a normal mode and a special mode. The timing controller 100 outputs clock pulses to various parts within the electronic endoscope system 10 to adjust the operating time of each part.

[0099] The light source 96 transmits illumination light to the illumination section 14 of the electronic endoscope 12 via the light guide 94. For example, a high-brightness lamp such as a xenon lamp, halogen lamp, mercury lamp, or metal halide lamp is used as the light source. The illumination light transmitted from the light source is focused by a condenser lens (not shown) and then limited to an appropriate amount of light by an aperture. The aperture is mechanically connected to an electric motor via a transmission mechanism such as a support arm and gears (not shown). The aperture opening can be adjusted to ensure that the brightness of the image displayed on the screen of the camera image display section 28 is appropriate.

[0100] The illumination light passing through the aperture is incident on the illumination section 14 of the electronic endoscope 12 via the light guide 94. The incident illumination light is emitted from the ejector end face 70 located at the tip end 56 through the light distribution lens.

[0101] The light source of the light source unit 96 can be replaced by a semiconductor light-emitting element that emits light in a predetermined wavelength range, such as a light-emitting diode or a laser diode, instead of a white light source that emits white light.

[0102] The system controller 102 performs various operations based on the inherent information of the electronic endoscope 12 and generates control signals. The system controller 102 uses the generated control signals to control the operation and timing of various circuits in the camera image processor 22, so as to properly process the electronic endoscope 12 connected to the camera image processor 22. The system controller 102 obtains the inherent information from the memory 92 read by the drive signal processing unit 18.

[0103] The timing controller 100 supplies clock pulses to the drive signal processing unit 18 according to the timing control executed by the system controller 102. The drive signal processing unit 18 drives and controls the camera element 16 according to the clock pulses supplied from the timing controller 100 and in a timing sequence synchronized with the image frame rate processed on the camera image processor 22 side.

[0104] The camera image processor 22 includes a camera image processing unit 26. The camera image processing unit 26 performs its function by reading and executing a program recorded in the memory 104 through the system controller 102. Therefore, the camera image processing unit 26 can be installed within the system controller 102 so as to operate together with the system controller 102.

[0105] The camera image processing unit 26 is equipped with a pre-processing circuit that performs mosaic processing on each of the R, G, and B image signals input from the drive signal processing unit 18 at frame periods. Specifically, interpolation processing using the surrounding pixels of G and B is performed on each of the R image signals, interpolation processing using the surrounding pixels of R and B is performed on each of the G image signals, and interpolation processing using the surrounding pixels of R and G is performed on each of the B image signals. As a result, all image signals are converted into image data with information on the three color components of R, G, and B. Furthermore, the pre-processing circuit performs known processing such as color correction, matrix operations, and white balance correction.

[0106] The camera image processing unit 26 may include a post-processing signal circuit. The post-processing signal circuit performs predetermined signal processing on the image data to generate motion image data and converts it into a predetermined video format signal. The converted video format signal is used to display the motion image on the display unit 46. Thus, a motion image of living tissue is displayed on the display screen.

[0107] (Imaging mode of ultrasound image processor)

[0108] The ultrasound image processor 30 has at least two imaging modes: Mode A and Mode B. Mode A is a mode that displays the amplitude information of echo signals in a specified direction within the subject on a time axis. Mode B is a mode that generates and displays a one-dimensional or two-dimensional image representing the shape of tissue in the subject along a given direction.

[0109] Figure 4 This describes the imaging principle of ultrasound images in an electronic endoscope system according to one embodiment. Ultrasound imaging is based on the ultrasonic pulse reflection method. Figure 4 This refers to the state in which an ultrasonic beam is generated from an ultrasonic probe 20 within a living organism. The ultrasonic beam is emitted into the organism in pulse form, for example, at approximately 10 MHz. Due to the difference in acoustic impedance between the organism's tissues and the surrounding tissues, the emitted ultrasonic waves are reflected and received again by the ultrasonic probe 20. This reflected wave is the echo signal.

[0110] The ultrasonic beam scanning of the electronic endoscope 12 is a phased array sector scan, which can acquire echo signals along a predetermined direction. Figure 4 This shows an example of an echo signal displayed in Mode A. In Mode A, the horizontal axis represents time, the vertical axis represents reflection intensity (amplitude), and the echo signal is displayed. Time represents biological depth, i.e., the distance between tissues within the organism.

[0111] In mode A, the waveform of the echo signal is converted to brightness through brightness modulation based on the reflection intensity, such as... Figure 4As shown, the tomographic image is represented by brightness as a shaded image, which is the B mode. The imaging modes of the ultrasound image processor 30 can also include the known M mode and Doppler mode.

[0112] Figure 5 This is an example diagram illustrating signal processing before the echo signal acquired during ultrasound image acquisition is displayed as a B-mode ultrasound image. The echo signal acquired by the ultrasound probe 20 is amplified by the amplifier circuit 120 and integrated by the integrator circuit 122 to remove harmonic noise. The amplifier circuit 120 and the integrator circuit 122 can be integrated using an integrated inverting amplifier-integrator circuit, providing both amplification and low-pass filter functions. Next, the analog signal, i.e., the echo signal, is digitized in the A / D converter 124 within a sampling period dictated by the clock signal 126, becoming a digital echo signal.

[0113] The digital echo signal is modulated by the luminance modulation unit 128 according to the reflection intensity and converted into luminance. The converted digital echo signal undergoes image processing in the ultrasound image generation unit 130 to become a two-dimensional B-mode tomographic image. This digital image signal is converted into an analog signal by the D / A converter 132 and displayed as a B-mode tomographic image in the ultrasound image display unit 46.

[0114] Figure 6 It means Figure 4 The diagram shows an example of the echo signal and luminance signal after the echo signal of mode A passes through amplifier circuit 120 and integrator circuit 122. Figure 6 As shown, Figure 4 The echo signal in Mode A shown has high-frequency noise removed. For example... Figure 6 As shown, the echo signal in mode A becomes a luminance signal with high-frequency noise removed. As mentioned above, existing noise processing of ultrasound images is essentially done by... Figure 5 The amplifier circuit 120 and integrator circuit 122 shown are either integrated by an inverting amplifier circuit. Additionally, the frequency of the clock signal 126 of the A / D inverter 124, which is more than half the sampling frequency, is the undigitized frequency generated by the sampling, and the A / D inverter 124 also functions as a low-pass filter.

[0115] (Periodic noise components)

[0116] Figure 7 This is an example diagram of a mode B ultrasound image acquired by an electronic endoscope system according to one embodiment. The sector-shaped portion of the sector scan is the ultrasound image, and the upper white portion corresponds to the drive signal of the ultrasound probe 20. In this example, as indicated by the arrow, a radially extending white curved curve appears, which is the noise component with a period (strictly speaking, a period that varies within a certain range).

[0117] Figure 8 This diagram illustrates the scanning positions of the ultrasound beam and the noise generation areas in a B-mode ultrasound image obtained by an electronic endoscope system according to one embodiment. Noise generation areas at different scanning positions in scans A, B, and C are indicated by circles. For example, in scan B, there are four noise generation locations; if the time intervals are T1, T2, and T3, then T1, T2, and T3 are generated with a certain amplitude. The time intervals T1, T2, and T3 are not fixed but rather fall within a certain range.

[0118] Furthermore, it can be seen that the time intervals between noise occurrences in scans A and C are not constant, but rather occur within a certain range, just like in scan B. Therefore, the noise component represented by a radially curved curve has a periodic amplitude within a certain range during the scan. In other words, in the frequency domain, noise can be considered as a noise component that occurs periodically within a certain frequency band.

[0119] Figure 9 This is an example diagram showing the frequency of noise components appearing in a B-mode ultrasound image acquired by an electronic endoscope system according to one embodiment. When the interval x representing the high-brightness points of the noise component is measured, it is 5.3 cm. If the speed of sound within a living organism is set as v, the noise frequency fn can be expressed as fn = v / 2x. When the speed of sound v is set, for example, to 1540 m / s, the frequency fn of the noise component is 14.5 kHz. This frequency becomes the passband of the low-pass filter in the integrator circuit 122, or the inverting amplifier integrator circuit, which existing noise processing cannot handle. Furthermore, the speed of sound varies depending on the living tissue, such as blood 1570 m / s, fat 1450 m / s, kidney 1560 m / s, and muscle 1590 m / s. The average is 1540 m / s.

[0120] Figure 10 This is a diagram illustrating an example of the power spectrum of a periodic noise component appearing in an ultrasound image obtained by an electronic endoscope system according to one embodiment. The generation period of the noise component is not fixed but varies within a certain range of periodic amplitude; therefore, the peak position of the power spectrum in the frequency domain also varies within a certain bandwidth. If the minimum period of the measured noise component is set as Tmin and the maximum period as Tmax, then... Figure 10 As shown, the power spectrum varies between the minimum frequency 1 / Tmax and the maximum frequency 1 / Tmin.

[0121] As a periodically generated noise component, Figure 7The noise frequency fn of the ultrasound image shown, calculated as an example, is 14.5 kHz. This frequency is quite low compared to the several MHz signals used in the electronic endoscope 12 or the ultrasound image processor 30, or the switching frequency of the switching power supply above several hundred kHz.

[0122] This periodic noise component is believed to be attributed to the following factors.

[0123] (1) The synchronization of clock signal frequencies used by multiple component devices or frequencies that are integer multiples of the switching signal frequencies of multiple component devices results in the generation of noise components.

[0124] (2) Noise components caused by switching power supply.

[0125] (3) Noise components generated in the electronic endoscope insertion section 54.

[0126] Regarding (1)

[0127] For example, the ultrasound image processor 30 synchronizes the repetition frequency (PRF) of the drive signal of the ultrasound probe 20 with an integer multiple of the switching frequency of the switching power supply to generate a two-dimensional B-mode image. If the switching frequency is set to an approximately integer multiple of the PRF and then the ultrasound is output, artifacts may appear on the B-mode image.

[0128] When the switching frequency is set to an approximately integer multiple of the PRF, and signal transmission and reception occur between the ultrasonic probe 20 and the ultrasonic image processor 30, the switching power supply will switch on and off during the time the echo signal is detected. Therefore, noise components from the switching are generated in each echo signal. When multiple echo signals obtained from the same ultrasonic beam are added together, the noise components caused by the switching in each echo signal are summed. Therefore, depending on the number of echo signals added, the noise components caused by the switching are also summed; this is noise components that are periodically generated in the B-mode image by being modulated into a high-intensity brightness signal.

[0129] Regarding (2)

[0130] Switching power supplies, such as buck DC / DC converters, use MOSFETs as switching elements. Resonance occurs in parasitic inductance and capacitance, which are parasitic elements in the converter circuit when the MOSFETs are switched. This resonance, through the floating capacitance in the output inductor of the buck DC / DC converter, generates high-frequency noise and noise components at the output of the DC / DC converter.

[0131] In the switching power supply of a DC / DC converter, surge absorption components and buffer circuits are included to suppress surge / ringing voltages that cause noise components. However, even with such noise countermeasures, it is difficult to completely eliminate noise components. Changing the switching frequency only changes the occurrence time (period) of the noise components, rather than eliminating them completely. Therefore, changing the switching frequency is better for preventing the synchronous superposition of noise components from multiple switching power supplies.

[0132] Figure 11 This is an example diagram illustrating the superposition of noise from multiple switching power supplies used in an electronic endoscope system according to one embodiment onto an echo signal. The example diagram shows the noise components of multiple switching power supplies superimposed on the echo signal. For example, noise components superimposed on the DC outputs Vda, Vdb, and Vdc of three switching power supplies synchronously become resonant noise components within a certain time period and are superimposed on the echo signal. This resonant noise component sometimes occurs periodically due to the setting of the switching frequency. In this case, the resonant noise component occurs periodically in the B-mode image.

[0133] Regarding (3)

[0134] Figure 12 This is an example diagram illustrating the noise factors of an electronic endoscope insertion section used in an electronic endoscope system according to one embodiment. As described above, an illumination section 14, an imaging element 16, and an ultrasonic probe 20 are provided at the front end of the insertion section 54. Light is transmitted to the illumination section 14 via a light guide 94. The imaging element 16 is connected to the imaging element drive signal line 142 and the imaging element output signal line 144. The ultrasonic probe 20 is connected to the ultrasonic element drive / output signal line 146, which transmits the drive signal and output signal of the ultrasonic element. Furthermore, if necessary, the insertion section 54 is wound together with the transmitting coil 118 of the electronic endoscope position measuring device 110, which will be described later.

[0135] A floating capacitor exists between the signal line and the transmitting coil 118, inducing noise in the ultrasonic element drive / output signal line 146 due to electrostatic coupling. Floating capacitor C1 is the floating capacitor between the camera element drive signal line 142 and the ultrasonic element drive / output signal line 146. Floating capacitor C2 is the floating capacitor between the camera element output signal line 144 and the ultrasonic element drive / output signal line 146. Floating capacitor C3 is the floating capacitor between the ultrasonic element drive / output signal line 146 and the ground line. Floating capacitor C4 is the floating capacitor between the ultrasonic element drive / output signal line and the transmitting coil 118. Further, floating capacitors C5, C6, and C7 are the floating capacitors between the ground line and the camera element drive signal line 142, the camera element output signal line 144, and the transmitting coil 118.

[0136] Due to the floating capacitors, the ultrasonic element drive / output signal line 146 is electrostatically coupled to each line. As the signal of each line, the flowing current is superimposed on the ultrasonic element drive / output signal line 146 as a differential mode noise component. Furthermore, due to electrostatic coupling, the floating capacitors C5 to C7 occurring between each signal line and the ground line become common-mode noise components. As current flows through these components, this noise is superimposed on the ultrasonic element drive / output signal line 146.

[0137] Furthermore, the magnetic field of the transmitting coil 118 and external EM wave noise are also superimposed on the ultrasonic element drive / output signal line 146. Additionally, as... Figure 6 As shown, EM wave noise may also be superimposed on the ultrasonic element drive / output signal line 146 as on the clock signal 126 of the A / D converter 124 that digitizes the echo signal. Generally, these noises are negligible, but if they are superimposed, they can appear as periodic noise components in the ultrasound image.

[0138] The connection line between the imaging element 16 and the ultrasonic probe 20 within the electronic endoscope uses shielding material to suppress noise components. Specifically, noise components are suppressed using the reflection loss, absorption loss, and multiple reflection correction of the shielding material. Reflection loss is the loss caused by reflection from the shielding material. Absorption loss is the loss caused by the induced current flowing through the shielding material when the EM wave is incident on it. In multiple reflection correction, a portion of the EM wave that penetrates the interior of the shielding material is reflected at the boundary, and its leakage to the outside during repeated reflections suppresses noise components. In multiple reflection correction, the thickness of the shielding material, the skin effect, and the wavelength of the EM wave should be considered for correction.

[0139] If this type of shielding material is used to suppress noise in the insertion part 54, the outer diameter of the insertion part 54 will increase, thus making it impossible to make the insertion part 54 sufficiently thin. Furthermore, in order to improve the performance of the electronic endoscope 12, when the drive signal or output signal is set to a small signal or a high-frequency signal, noise components are easily generated, which may adversely affect the high-quality ultrasound image. Therefore, the shielding material is insufficient to suppress noise components.

[0140] The periodic generation of noise components is caused by a single factor, but rather by a combination of various noise components. To suppress this periodically generated noise component and achieve high-quality ultrasound images, it is necessary to detect noise components that periodically generate in the echo signal above a pre-set threshold level and suppress the detected noise components. There are two methods for suppressing noise components: one is to suppress the noise component so that it is not included in the echo signal; the other is to perform image processing on the 2D B-mode image obtained from the echo signal containing noise components to eliminate noise pixels.

[0141] (Suppression of noise components)

[0142] Figure 13 This is an example diagram of an echo signal formed by the superposition of periodically occurring noise components in an electronic endoscope system according to one embodiment. Figure 4 As shown, the noise component corresponds to the high-brightness portion in the B-mode image, and is a waveform of an echo signal reflected in living tissue by a driving signal with an amplitude greater than that of the ultrasound probe 20. Therefore, in the noise detection unit 34, signals above a preset threshold level are identified as noise components and detected.

[0143] As described above, frequency analysis is performed using the FFT of the digital echo signal to determine whether there are spectral peaks above a set threshold level, thereby detecting noise components.

[0144] (Amplification and gain suppression of noise components)

[0145] Figure 14 This diagram illustrates how to change the amplification gain of noise pixel values ​​in an ultrasound image obtained from an electronic endoscope system according to one embodiment. The echo signal is converted into a digital echo signal by an amplification circuit 120, an integrator circuit 122, and an A / D converter 124. The noise detection unit 34 presets a noise threshold and identifies signals above this threshold as noise components. The period (T1, T2, T3, ...) of the digital echo signal detected as a noise component is measured by the period amplitude detection unit 150. In this case, it is sometimes also part of random noise; preferably, short periods below a certain period are excluded.

[0146] The periodic amplitude detection unit 150 is set in Figure 1 In the noise detection unit 34 of the ultrasound image processor 30 shown, for example, the period can be calculated from the waveform of the digital echo signal displayed in A mode, and the peak frequency of the power spectrum obtained by frequency analysis using the FFT of the digital echo signal can be determined. By identifying multiple occurrence points of the noise component and calculating the time intervals (T1, T2, T3, ...) between the occurrence points, the period can be calculated. In this way, the period amplitude detection unit 150 calculates the period amplitude, and the noise component is detected using the noise period and period amplitude. Therefore, the detection accuracy is high.

[0147] The measured periods (T1, T2, T3, ...) vary within a certain range. The period amplitude detection unit 150 calculates the magnitude of the deviation and the maximum and minimum periods, Tmax and Tmin. The digital filter design unit 152 designs a band removal filter 154, which uses the period amplitude detected by the period amplitude detection unit 150 as the bandwidth in the frequency domain to remove the frequency bands of the minimum frequency 1 / Tmax and the maximum frequency 1 / Tmin (setting the removal bands). By passing the digital echo signal through this band removal filter 154, even if periodic noise components appear in the echo signal, the amplitude of these noise components can be suppressed, and periodic noise components in the B-mode image can also be suppressed.

[0148] Preferably, the gain modification process is performed sequentially over time. A frequency band removal filter is used to determine the frequency band corresponding to the noise period from the noise period and period amplitude, and this frequency band is used to remove noise components from the echo signal. It should be noted that, based on the characteristics of the frequency band removal filter, the removal frequency band is set so that the frequency band corresponding to the fluctuation period (T1, T2, T3, ...) becomes the removal bandwidth.

[0149] Figure 15 The frequency characteristics of all filters used to filter the echo signal are shown. The echo signal is filtered by integrator 122 and band-removal filter 154, but in integrator 122, the frequency band B of random harmonic noise components is suppressed, and the periodically generated noise components are suppressed by the lower frequency band A.

[0150] Figure 16 This is an example diagram of a digital echo signal obtained from an electronic endoscope system according to one embodiment, after passing through a band-removal filter, including an echo signal containing periodic noise. Harmonic noise is suppressed by the integrator circuit 122, and the digital echo signal is also suppressed after passing through the band-removal filter 154. Thus, periodically generated noise components are suppressed in the brightness signal obtained by the brightness modulation unit 128.

[0151] Furthermore, the amplification gain adjustment is configured as follows: In the luminance modulation unit 128 that converts the digital echo signal into a luminance signal, instead of the band removal filter 154, the amplification gain value of the signal corresponding to the noise component is reduced based on the determined noise component occurrence time information, thereby reducing the luminance value. Thus, even if periodic noise components are generated in the echo signal, the amplitude of these noise components can be suppressed by adjusting the amplification gain, thereby suppressing periodic noise components in the B-mode image.

[0152] (Interpolation processing of noisy pixel values)

[0153] In one embodiment of the electronic endoscope system, noisy pixels can be removed from an ultrasound image obtained as a B-mode image, and interpolation processing can be performed.

[0154] Figure 17 This is an example diagram of a noise suppression unit in an electronic endoscope system according to one embodiment, which removes noise pixels from an ultrasound image obtained as a B-mode image and performs interpolation processing. In the ultrasound image (B-mode image) generated by the ultrasound image generation unit 130, noise pixels corresponding to periodically occurring noise components detected by the noise detection unit 34 are replaced by interpolated pixel values ​​generated based on the pixel values ​​of surrounding pixels by the image interpolation unit 160. The image interpolation unit 160 is provided with... Figure 1 The noise suppression unit 36 ​​of the ultrasound image processor 30 shown.

[0155] The image interpolation unit 160 performs pixel interpolation, for example, using the four neighboring pixels on the four sides of the noise pixel. The pixel value of the noise pixel is interpolated using the pixel values ​​of the four neighboring pixels. For example, the average of the pixel values ​​of the four neighboring pixels is set as the pixel value of the noise pixel. This interpolation method is called bilinear interpolation, but other interpolation methods can be used, such as bicubic interpolation. If the noise pixel is adjacent, the image interpolation unit 160 searches for pixels adjacent to the noise pixel until it finds a non-noise pixel adjacent to the noise pixel, and performs pixel interpolation using the pixel values ​​of multiple non-noise pixels surrounding the noise pixel region.

[0156] Thus, the noise suppression unit 36 ​​performs correction processing, replacing the pixel value of the noise component detected in the ultrasound image at the noise pixel location with the interpolated pixel value generated based on the pixel values ​​of the surrounding pixels located around the noise pixel location. Therefore, even if the noise component appears periodically in the echo signal, the noise component can be suppressed in the B-mode image.

[0157] (Eliminate power supply noise)

[0158] In many cases, a switching power supply is used to power each device of the electronic endoscope system 10. In this case, as mentioned above, noise components are generated due to the switching operation of the switching power supply. To suppress noise components, it is preferable to change the switching frequency in the DC / DC converter and the switching frequency in the AC / DC converter. Furthermore, to eliminate noise components, it is preferable to use a battery without switching operation.

[0159] Figure 18This diagram illustrates an example of an electronic endoscope system according to one embodiment, including a switching power supply, a battery, and a switching device. The AC / DC converter 202 uses commercial AC power as its input voltage and is switched via a switching signal S to output a DC voltage Vis. A battery S204 is also included in the AC / DC converter 202, configured such that a switching switch S206 switches between the AC / DC converter 202 (serving as a switching power supply) and the battery S204. This ensures that the DC voltage Vis is a voltage free of noise components generated by the switching operation. Furthermore, by switching to the battery S204, the AC / DC converter 202 stops operating, preventing external noise from the commercial power supply from entering.

[0160] The DC voltage Vis output from AC / DC converter 202 is the input voltage of DC / DC converters A206, B214, and C220. A battery A210 is connected in parallel with DC / DC converter A208, configured such that switch A212 switches between DC / DC converter A208 as a switching power supply and battery A210. Similarly, DC / DC converter C220 has batteries B216 and C222 connected in parallel, configured such that switch B218 and switch C224 switch them respectively.

[0161] The battery S104, DC / DC converter A208, DC / DC converter B214, and DC / DC converter C220 not only effectively suppress noise but also serve as emergency power supplies in case the AC outlet is disconnected or a power failure occurs. Furthermore, preferably, each converter also charges the parallel-installed battery. Thus, each battery is always charged and can be switched and used at any time.

[0162] Figure 19A , Figure 19B This is an example diagram of an electronic endoscope system according to one embodiment, in which a converter is used as a dedicated power source for charging a parallel-installed battery. When a battery is installed for each converter, a battery that operates without noise during switching can always be used, and the converter can also be used as a dedicated power source for the battery. In this case, the battery capacity can be less than... Figure 18 The switching method shown has a capacity. When the AC / DC converter 202 is used as a dedicated charging power source to charge the parallel-installed battery S204, the output voltage V of the battery S204 is... iS Feedback is sent to the AC / DC converter 202, and the battery S204 is controlled to have a constant output voltage V. is This control can also be achieved by switching the switch signal S on or off.

[0163] Figure 19BThis example demonstrates how a DC / DC converter A208 is used as a dedicated power source to charge a parallel-connected battery A210. The output voltage Vda of battery A210 is fed back to DC / DC converter A208, controlling battery A210 to have a specific output voltage Vda. DC / DC converters B214 and C220 can also be configured in the same manner. The output voltages Vda, Vdb, and Vdc are different voltage values ​​and correspond to the supply voltage of each component device.

[0164] For this purpose, multiple DC / DC converters are used. The battery consists of individual cells connected in series, with an output voltage of approximately 1-3V, allowing for an arbitrary output voltage value. The voltage value can be obtained depending on the number of individual cells connected in series. Therefore, if the output voltage of AC / DC converter 202 can be used to charge individual cells of the battery, then DC / DC converters A208, B214, and C220 are unnecessary.

[0165] Furthermore, in order to suppress noise components, preferably, the noise suppression unit 36 ​​changes at least one operating frequency of the plurality of constituent devices of the camera image processor 22 and the ultrasound image processor 30 to separate it from the generation period of the noise components that are periodically generated in the echo signal.

[0166] In this case, as described above, preferably, the device includes an electronic endoscope 12, an image processor 22, and a DC / DC converter as a switching power supply for driving the ultrasound image processor 30. The switching frequencies of DC / DC converters A208, B214, and C220 are changed to operating frequencies that are altered within the device to separate them from the generation periods of noise components that periodically occur in the echo signal. This helps suppress periodically generated noise components in the echo signal.

[0167] Preferably, the switching frequency in the AC / DC converter 202 is used as the altered operating frequency in the constituent device, and the altered operating frequency is spaced out from the generation period of the noise components that are periodically generated in the echo signal. This helps to suppress the periodically generated noise components in the echo signal.

[0168] At least one of the ultrasonic frequency of the ultrasonic probe 20 and the camera element operating frequency of the camera element 16 can be used as the operating frequency to be changed in the device. This will help suppress periodically generated noise components in the echo signal.

[0169] Furthermore, the sampling frequency of the echo signal in the A / D inverter 124 can also be the operating frequency that is changed within the component equipment. This helps to suppress periodically generated noise components in the echo signal.

[0170] Thus, the ultrasound image processor 30 can generate high-quality ultrasound images because it can detect noise components above a preset threshold that are included in the echo signal and appear periodically, and process them to suppress the detected noise components.

[0171] More specifically, in the electronic endoscope system 10, the echo signal output from the ultrasonic probe 20 is superimposed with various noises, many of which are random noises with harmonic components. Therefore, high image quality is achieved by removing harmonic components. However, periodically generated noise components may be generated by many signals and their harmonics synchronizing with each other, with the period of harmonic synchronization in the low-frequency range and the amplitude being relatively large. To address this, the ultrasonic image processor 30 is equipped with a noise detection unit 34, which identifies periodically occurring echo signals above a preset threshold as noise and suppresses these noise components through a suppression unit 36.

[0172] Therefore, the electronic endoscope system 10 utilizes the fact that periodically generated noise is a noise component with a large amplitude, detects it by setting a threshold, and the suppression unit 36 ​​suppresses the noise component, thereby reducing the detected noise component and obtaining a high-quality ultrasound image.

[0173] Furthermore, when using the electronic endoscope system 10, an electronic endoscope position measuring device 110 can be used, which can use magnetic force to position the electronic endoscope 12 inserted into the body cavity. Figure 20 An example of an electronic endoscope position measuring device 110.

[0174] Figure 20 This is an example diagram of an electronic endoscope position measuring device used in an electronic endoscope system 10 according to one embodiment. Multiple transmitting coils 118 for the electronic endoscope position measuring device 110 are wound at predetermined intervals around the insertion portion 54 of the electronic endoscope 12, and generate a magnetic field through an electric current. When the electronic endoscope 12 is inserted into the body cavity, the multiple transmitting coils 118-1, 118-2, ..., 118-n generate a magnetic field at each position, which is received by a receiving coil 116 to measure the position of the endoscope. Figure 20 The image shows a case where the transmitting coil 118 is in the insertion part 54 and the receiving coil 116 is outside, but the transmitting coil 118 can be outside and the receiving coil 116 can be inside the endoscope.

[0175] The receiving coil 116 has multiple coil blocks, for example, arranged beside the bottom layer. Each coil block of the receiving coil 116 is wound such that its respective coil surface is orthogonal to each other in three directions. The coil is designed to detect a signal proportional to the magnetic field strength of the axial component orthogonal to its coil surface. The coil block receives the generated magnetic field and converts it into a voltage signal, outputting the voltage signal as the detection result. The drive unit 112 controls the operating state of these transmitting coils 118 and receiving coils 116.

[0176] Each transmitting coil 118-1, 118-2, ..., 118-n is supplied with a high-frequency sine wave by the driving unit 112 via the image processor 22. Each transmitting coil 118-1, 118-2, ..., 118-n emits electromagnetic waves with a magnetic field to the surroundings by applying the sine wave. Furthermore, the driving unit 112 can also specify the time for each transmitting coil 118-1, 118-2, ..., 118-n to generate its magnetic field individually.

[0177] The receiving coil 116 receives the magnetic field generated by the transmitting coil 118, generates a current through the received magnetic field, and converts it into a voltage signal. The signal is transmitted from the receiving coil 116 to the driving unit 112. The driving unit 112 provides the signal from the receiving coil 116 to the positioning unit 114, performs amplification and other prescribed signal processing, and then converts it into digital data through an A / D converter.

[0178] The frequency is extracted by performing a Fast Fourier Transform on the digital data, and the magnetic field detection information corresponding to the frequency components of the sine wave of each transmitting coil 118-1, 118-2, ..., 118-n is separated and extracted for positioning. Based on the digital data of the separated magnetic field detection information, the spatial position coordinates of each transmitting coil 118-1, 118-2, ..., 118-n set on the ultrasonic probe 20 are calculated. The positioning unit 114 further connects the position coordinates of each transmitting coil 118-1, 118-2, ..., 118-n to generate a linearly interpolated image, which serves as the position image of the electronic endoscope.

[0179] In addition, the length from the coil position at the entrance of the insertion part to the coil position at the tip of the insertion part can also be calculated as the insertion length.

[0180] When using this electronic endoscope position measuring device 110, if noise components are periodically generated in the echo signal, preferably, the electronic endoscope position measuring device 110, as one of the components, changes at least one of its operating frequencies to be spaced out from the generation period of the periodically generated noise components in the echo signal. The drive signal of the transmitting coil 118 may cause periodically generated noise components in the echo signal. Therefore, preferably, the driving unit 112 controls the driving unit 112 to change the frequency of the drive signal to the operating frequency. Alternatively, preferably, the noise suppression unit 36 ​​controls the driving unit 112 to stop transmitting the drive signal so as to stop the generation of magnetism in the transmitting coil 118 when noise components are detected. This helps to suppress periodically generated noise components in the echo signal.

[0181] In the described implementation, noise components are detected using pre-set thresholds or frequency analysis, but they can also be detected using predictive models derived from machine learning, i.e., using artificial intelligence (AI).

[0182] In the artificial intelligence detection, feature quantities are extracted through machine learning in the noise detection unit 34. Learning data is used, specifically ultrasound images without periodic noise components (echo signals) and ultrasound images with periodic noise components. This allows for pre-learning of the presence or absence of periodic noise components. Based on this learning result, it is determined whether the input ultrasound image contains periodic noise components.

[0183] Figure 21 This is an example diagram illustrating the process of learning and determining the presence or absence of periodically generated noise components in an electronic endoscope system according to one embodiment. The noise component inference model 230 is created by extracting feature quantities from ultrasound images used for learning about the presence or absence of noise components through machine learning. By inputting the ultrasound image into this noise component inference model 230, the presence or absence of noise components can be determined based on the feature quantities extracted internally by the noise component inference model 230. These feature quantities include the amplitude of the noise component, the periodicity of the noise component, and the expansion of the noise component.

[0184] Figure 22 This is an example diagram illustrating how AI derives effective noise suppression strategies in an electronic endoscope system according to one embodiment. The noise suppression performed by the AI ​​is conducted using a noise countermeasure inference model 232. The noise countermeasure inference model 232 is a model in the noise suppression unit 36 ​​that establishes a relationship between the machine learning of the operating environment information of the electronic endoscope, the camera image processor, or the ultrasound image processor, the feature quantities of the ultrasound image, and information regarding the effectiveness of corresponding strategies for suppressing noise components based on these feature quantities and the operating environment information.

[0185] When the noise detection unit 34 detects noise components in the newly generated ultrasound image, it extracts feature quantities from the ultrasound image and inputs the operating environment information at the time of noise detection and the extracted feature quantities into the created noise countermeasure inference model 232, thereby setting changes to the environmental operating information that is effective in noise suppression. In the suppression unit 36, based on the set environmental operating information, the operating environment of the electronic endoscope 12, the camera image processor 22, the ultrasound image processor 30, or the electronic endoscope position measurement device 110 is changed. Periodic noise is suppressed by changing the operating environment.

[0186] The noise component inference model 230 of the noise detection unit 34 and the noise countermeasure inference model 232 of the noise suppression unit 36 ​​are, for example, made using neural networks. For example, a convolutional neural network (CNN), which is widely used in image recognition, can be used.

[0187] Operating environment information, preferably, includes the operating frequencies of the multiple component devices included in the camera image processor 22 and the ultrasound image processor 30. Preferably, the operating frequencies include the ultrasound frequency of the ultrasound probe 20 and the operating frequency of the camera element 16, the sampling frequency of the echo signal in the A / D converter 124, the switching frequencies of the DC / DC converters 208, 214, and 220, the switching frequency of the AC / DC converter 202, and the operating frequency of the transmitting coil 118 in the electronic endoscope position measuring device 110. Because the operating frequencies are changed via AI, noise can be effectively suppressed.

[0188] Figure 23 The flowchart illustrates the noise correction process. First, in step S1, an ultrasound image is generated. The ultrasound image is generated by the ultrasound image generation unit 130 after the echo signal from the ultrasound probe 20 is converted into pixels of varying density by the brightness modulation unit 128. In step S2, periodic noise is detected. Noise detection identifies signals above a predetermined threshold as noise. If no noise is detected, the process ends.

[0189] When noise is detected, ultrasound image correction is performed. Ultrasound image correction is performed by the band removal filter 154 or the image interpolation unit 160. In step S4, noise detection is performed again to determine whether the noise has been corrected by ultrasound image correction. If no noise is detected, the process ends. When noise is detected, in step S5, the operating environment, including the operating frequencies of multiple component devices in the electronic endoscope 12, the camera image processor 22, the ultrasound image processor 30, and the electronic endoscope position measuring device 110, is changed. The changed operating environment includes the operating frequencies of the multiple component devices in the electronic endoscope 12, the camera image processor 22, the ultrasound image processor 30, and the electronic endoscope position measuring device 110.

[0190] The operating frequencies include the ultrasonic frequency of the ultrasonic probe 20, the operating frequency of the imaging element 16, the sampling frequency of the echo signal in the A / D converter 124, the switching frequencies of the DC / DC converters 208, 214, and 220, the switching frequency of the AC / DC converter 202, and the operating frequency of the transmitting coil 118 in the electronic endoscope position measuring device 110. The operation of the imaging element 16 or the transmitting coil 118 can be stopped. Furthermore, the DC / DC converters 208, 214, and 220, and the AC / DC converter 202 can be replaced with a battery.

[0191] In step S6, noise detection is performed again to confirm the effect of the changed operating environment. If no noise is detected, the process ends. If noise is detected, an alarm is triggered and the process exits in step S7. The alarm will be displayed on the ultrasound image display unit 46. At this time, the surgeon will be informed of the noise to avoid affecting the diagnosis of the ultrasound image.

[0192] Figure 24 This is a flowchart of the AI-based noise correction process. First, in step S11, an ultrasound image is generated. In step S12, periodic noise is detected using the noise component inference model 230 in the noise detection unit 34. Noise detection is performed using the machine learning-based noise component inference model 230. If no noise is detected, the process ends.

[0193] In step S13, the operating environment, including the operating frequencies of multiple component devices in the electronic endoscope 12, camera image processor 22, ultrasound image processor 30, and electronic endoscope position measurement device 110, is changed. Changing the operating environment requires extracting feature quantities from the ultrasound image, matching these feature quantities using a created noise countermeasure inference model 232, and setting changes to the environmental operation information to effectively suppress noise based on the matching results. In the noise suppression unit 36, the operating environment of the electronic endoscope 12, camera image processor 22, ultrasound image processor 30, or electronic endoscope position measurement device 110 is changed based on the set environmental operation information.

[0194] The operational environment information includes the operating frequencies of multiple components in the electronic endoscope 12, camera image processor 22, ultrasound image processor 30, or electronic endoscope position measuring device 110.

[0195] The operating frequencies include the ultrasonic frequency of the ultrasonic probe 20, the operating frequency of the imaging element 16, the sampling frequency of the echo signal in the A / D converter 124, the switching frequencies of the DC / DC converters 208, 214, and 220, the switching frequency of the AC / DC converter 202, and the operating frequency of the transmitting coil 118 in the electronic endoscope position measuring device 110. The operation of the imaging element 16 or the transmitting coil 118 can be stopped. Furthermore, the DC / DC converters 208, 214, and 220, and the AC / DC converter 202 can be replaced with a battery.

[0196] In step S14, to confirm the effect of the changed operating environment, noise detection is performed again by the AI. If no noise is detected, the process ends. If noise is detected, in step S15, the ultrasound image is corrected by interpolation. In step S16, noise detection is performed by the AI ​​to confirm the correction effect of the ultrasound image. If no noise is detected, the process ends. If noise is detected, an alarm is triggered and the process exits in step S17. The alarm is displayed on the ultrasound image display unit 46. At this time, the surgeon is informed of the noise to avoid affecting the diagnosis of the ultrasound image.

[0197] The endoscopic diagnostic data management system provided by the present invention has been described in detail above. However, the endoscopic diagnostic data management system provided by the present invention is not limited to the described embodiments, and various improvements and changes can be made without departing from the spirit of the present invention.

Claims

1. An electronic endoscope system for acquiring ultrasound images, wherein, have: An electronic endoscope has an imaging element at its front end for imaging living tissue and an ultrasonic probe for transmitting ultrasound waves to the living tissue to obtain an echo signal. A camera image processor has an image processing unit that processes camera signals output from the camera element to generate a camera image; The system also includes an ultrasound image processor comprising an ultrasound image processing unit, a noise detection unit, and a noise suppression unit. The ultrasound image processing unit processes the echo signal output from the ultrasound probe to generate an ultrasound image. The noise detection unit detects noise components in the echo signal that are generated periodically at or above a predetermined threshold level. The noise suppression unit suppresses the detected noise components. The frequency of noise components generated at a predetermined threshold level or higher is determined based on the ratio of sound velocity to the interval of high-brightness points in twice the ultrasound image. The brightness of the point in the ultrasound image corresponds to the echo signal obtained at the point of occurrence at different times.

2. The electronic endoscope system as described in claim 1, wherein, The noise detection unit includes a periodic amplitude detection unit, which calculates the noise period generated periodically by the noise component and calculates the periodic amplitude based on the minimum and maximum values ​​of the noise period. The noise detection unit uses the noise period and the periodic amplitude to detect the noise component.

3. The electronic endoscope system as described in claim 1 or 2, wherein, The noise suppression unit performs gain modification processing to change the amplification gain value between the pixel value of the noise pixel position in the ultrasound image of the detected noise component and the pixel other than the noise pixel at the noise pixel position.

4. The electronic endoscope system of claim 3, wherein the noise suppression unit includes a frequency band removal filter, which calculates a frequency band corresponding to the noise period based on the noise period and the period amplitude, and uses the frequency band to remove the noise component from the echo signal.

5. The electronic endoscope system as described in claim 1 or 2, wherein, The noise suppression unit performs a correction process, replacing the pixel value at the noise pixel location in the ultrasound image of the detected noise component with an interpolated pixel value generated based on the pixel values ​​of the surrounding pixels at the noise pixel location.

6. The electronic endoscope system as described in claim 1 or 2, wherein, The noise suppression unit alters at least one of the operating frequencies of the multiple component devices of the camera image processor and the ultrasound image processor to separate them from the generation period of the noise component that is periodically generated by the echo signal.

7. The electronic endoscope system as claimed in claim 6, wherein, The operating frequency includes at least one of the ultrasonic frequency of the ultrasonic probe and the operating frequency of the camera element.

8. The electronic endoscope system as claimed in claim 6, wherein, The ultrasound image processor includes an A / D converter as a component device, which converts the echo signal of the analog signal into a digital signal through sampling; The operating frequency includes the sampling frequency of the echo signal in the A / D conversion unit.

9. The electronic endoscope system as claimed in claim 6, wherein, The ultrasound image processor includes a DC / DC converter as a component device. This DC / DC converter is a switching power supply used to drive the electronic endoscope, the camera image processor, and the ultrasound image processor. The operating frequency includes the switching frequency in the DC / DC converter.

10. The electronic endoscope system as claimed in claim 9, wherein, The ultrasound image processor includes an AC / DC converter as a component device. The AC / DC converter converts AC power received from an external commercial power supply into DC power and provides the DC power to the switching power supply of the DC / DC converter. The operating frequency includes the switching frequency in the AC / DC converter.

11. The electronic endoscope system of claim 10, wherein, The DC / DC converter and the AC / DC converter include a battery that supplies DC power. When the noise component is detected, the noise suppression unit switches from the DC / DC converter and the AC / DC converter to the battery.

12. The electronic endoscope system as claimed in claim 1 or 2, wherein, The electronic endoscope system includes an electronic endoscope position measuring device, which uses magnetism to determine the position of the electronic endoscope when it is inserted into the body cavity. The endoscope position measuring device includes a drive unit that winds a transmitting coil around the electronic endoscope and a positioning unit that determines the position of the electronic endoscope using a signal from the transmitting coil when the electronic endoscope is inserted into a body cavity. When the noise suppression unit detects the noise components, it changes the operating frequency of the transmitting coil or stops the generation of the magnetism.

13. The electronic endoscope system as claimed in claim 1, wherein, The noise detection unit, Using both noise-free and noisy ultrasound images as learning data, a pre-machine learning model for inferring the presence or absence of the noise component is created. The presence or absence of the noise component is determined by inputting the ultrasound image generated in the image processing unit into the noise component inference model.

14. The electronic endoscope system of claim 13, wherein, The noise suppression unit creates a noise countermeasure inference model based on machine learning, which relates the relationship between the action environment information of the electronic endoscope, the camera image processor, or the ultrasound image processor and the feature quantity of the ultrasound image, and the information on whether the measures to suppress the noise components corresponding to the action environment information are effective. When the noise detection unit detects the noise component in the newly generated ultrasound image, it extracts the feature quantity of the ultrasound image, and inputs the operating environment information when detecting the noise component and the extracted feature quantity into the created noise countermeasure inference model to set an effective noise suppression countermeasure.

15. The electronic endoscope system of claim 14, wherein, The motion environment information includes the operating frequencies of the multiple component devices of the camera image processor and the ultrasound image processor. The operating frequencies include the ultrasonic frequency of the ultrasonic probe, the operating frequency of the imaging element, the sampling frequency of the echo signal in the A / D converter, the switching frequencies in the DC / DC converter and AC / DC converter, and the operating frequency of the transmitting coil of the electronic endoscope position measuring device.

Images