Image processing device, operation method of image processing device, program, and medical image processing system

Abstract

To provide an image processing device, an operation method of an image processing device, a program, and a medical image processing system that can achieve denoise processing matched to a user's preference.SOLUTION: An image processing device acquires user identification information (110), acquires a first imaging condition (112), acquires a first image to which the first imaging condition is applied (100), executes denoise processing for the first image applying first denoise strength, and generates a first denoised image (130), acquires a second signal noise ratio corresponding to a second imaging condition the same as or close to the first imaging condition out of second signal noise ratios corresponding to a plurality of second imaging conditions as a first signal noise ratio (120), and sets the denoise strength based on the first signal noise ratio as first denoise strength (122).SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to an image processing device, an operating method for an image processing device, a program, and a medical image processing system. [Background technology]

[0002] Patent Document 1 describes a medical imaging device equipped with an MRI device as an imaging device. In the medical imaging device described in the document, when the magnitude of noise differs from region to region of an image, a denoising filter with a relatively high strength is applied to a region with a relatively high noise level, and a denoising filter with a relatively low strength is applied to a region with a relatively low noise level. Note that MRI is an abbreviation of Magnetic Resonance Imaging, which is the English abbreviation of magnetic resonance imaging.

[0003] Furthermore, the medical imaging device described in Patent Document 1 stores the relationship between the region selection pattern selected by the user and the imaging conditions. This makes it possible to select the region selection pattern selected in the past when imaging conditions similar to those applied in the past are to be applied.

[0004] Furthermore, the medical imaging device described in Patent Document 1 includes a slide bar in which a region selection pattern is associated with each slide position. A user can select a region selection pattern that provides a desired high-quality image by sliding the slide bar while viewing the high-quality image displayed on the display device and stopping the slide position. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2022-6869 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, in image processing devices according to the related art, the denoising intensity is set before imaging, and denoising is applied to an image generated after imaging is completed. When a user checks an image to which denoising has been applied and wishes to change the denoising intensity, the denoising intensity is reset and reimaging is performed. In an actual clinical setting, performing reimaging is not practical.

[0007] Furthermore, the denoising strength recommended by the manufacturer of the image processing device does not necessarily match the preferences of users such as doctors and technicians, and it is desirable to optimize the denoising strength to suit the preferences of users.

[0008] The present disclosure has been made in consideration of the above circumstances, and aims to provide an image processing device, an operating method for an image processing device, a program, and a medical image processing system that realizes denoising processing that applies denoising strength tailored to the user's preferences. [Means for solving the problem]

[0009] An image processing device according to a first aspect of the present disclosure includes a processor that acquires user identification information, acquires first imaging conditions to be applied to an imaging device, acquires a first image captured and generated using the imaging device to which the first imaging conditions are applied, sets a first denoising intensity to be applied to denoising processing on the first image, and applies the first denoising intensity to perform denoising processing on the first image to generate a first denoising image, wherein the processor acquires, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition close to the first imaging condition from among a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions were applied for each user, and sets the denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

[0010] According to the image processing device of the first aspect, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition close to the first imaging condition is acquired as a first signal-to-noise ratio from a plurality of second signal-to-noise ratios derived from a plurality of second images generated in a plurality of past imaging for each user and associated with the second imaging condition in the past shooting. This realizes denoising processing in which a first denoising strength according to a first signal-to-noise ratio preferred by the user is applied.

[0011] In the image processing device of the second aspect, in the image processing device of the first aspect, the processor may refer to a signal-to-noise ratio database in which a plurality of second imaging conditions are associated with and stored for a plurality of second signal-to-noise ratios, and obtain a first signal-to-noise ratio corresponding to a second imaging condition that is identical to the first imaging condition, or a second imaging condition that is close to the first imaging condition.

[0012] An image processing device according to a third aspect is the image processing device of the second aspect, further comprising: When the first denoising strength is changed, the signal-to-noise ratio database may be updated in accordance with the second denoising strength after the change.

[0013] An image processing device according to a fourth aspect is an image processing device of any one of the first to third aspects, wherein a processor performs denoising processing multiple times with different denoising intensities on a first image to generate multiple third denoising images, derives a third signal-to-noise ratio from each of the multiple third denoising images, and sets the third denoising intensity applied to the third denoising image from which the third signal-to-noise ratio that has the smallest difference from the first signal-to-noise ratio is derived as the first denoising intensity.

[0014] An image processing device according to a fifth aspect is an image processing device of any one of the first to third aspects, wherein the processor calculates a fourth signal-to-noise ratio from the first image, and sets a fourth denoising intensity corresponding to the fourth signal-to-noise ratio as the first denoising intensity using a function representing the relationship between the signal-to-noise ratio and the denoising intensity.

[0015] An image processing device according to a sixth aspect is an image processing device of any one of the first to fifth aspects, wherein the processor causes a first denoising image to be displayed on a display device, causes options representing an evaluation of a first denoising strength to be applied to denoising processing on the first denoising image to be displayed on the display device, obtains a first evaluation signal representing the evaluation of the first denoising strength selected from the options, and feeds back the evaluation of the first denoising strength represented by the first evaluation signal to a setting of the first denoising strength.

[0016] An image processing device according to a seventh aspect is an image processing device of any one of the first to sixth aspects, wherein a processor applies a fifth imaging condition that is identical to the first imaging condition, or a fifth imaging condition that is close to the first imaging condition, applies a first denoising intensity to a past image that was previously captured and generated, performs denoising processing, displays the generated fifth denoised image on a display device, acquires a fifth evaluation signal representing an evaluation of the fifth denoising intensity to be applied to the denoising processing on the fifth denoised image, and decides whether to maintain or change the first denoising intensity depending on the evaluation of the fifth denoising intensity represented by the fifth evaluation signal.

[0017] An image processing device according to an eighth aspect is the image processing device of the seventh aspect, wherein as the previous image, a normal image generated by previously capturing an image of a normal imaging target may be applied.

[0018] An image processing device according to a ninth aspect is the image processing device of the seventh aspect, in which a sixth imaging condition that is the same as the first imaging condition or a sixth imaging condition close to the first imaging condition is applied to a previous image, and an image of the same imaging target generated by previously imaging the same imaging target is applied, and the processor may display on the display device a sixth denoised image obtained by performing denoising on the image of the same imaging target, and a sixth denoising intensity applied to the denoising performed on the image of the same imaging target.

[0019] In an image processing device of a 10th aspect, in the image processing device of the 9th aspect, the processor may acquire a sixth evaluation signal representing an evaluation of a sixth denoising intensity, and decide whether to maintain or change the first denoising intensity depending on the evaluation of the sixth denoising intensity represented by the sixth evaluation signal.

[0020] An operation method of an image processing device according to an eleventh aspect of the present disclosure is an operation method of an image processing device to which a computer having a processor is applied, in which the processor acquires user identification information, acquires first imaging conditions to be applied to an imaging device, acquires a first image captured and generated using the imaging device to which the first imaging conditions are applied, sets a first denoising intensity to be applied to denoising processing on the first image, applies the first denoising intensity to perform denoising processing on the first image to generate a first denoising image, and, when setting the first denoising intensity, acquires, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition identical to the first imaging condition or a second imaging condition close to the first imaging condition from among a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions were applied for each user, and sets a denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

[0021] A program according to a twelfth aspect of the present disclosure is a program that causes a computer to realize a function of acquiring user identification information, a function of acquiring first imaging conditions to be applied to an imaging device, a function of acquiring a first image captured and generated using an imaging device to which the first imaging conditions are applied, a function of setting a first denoising intensity to be applied to denoising processing on the first image, and a function of applying the first denoising intensity to perform denoising processing on the first image to generate a first denoising image, wherein the function of setting the first denoising intensity is a program that acquires, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition, from a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions were applied, for each user, and sets the denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

[0022] The present disclosure also includes a tangible, non-transitory, computer-readable recording medium that stores the program according to the twelfth aspect.

[0023] A medical image processing system according to a thirteenth aspect of the present disclosure is a medical image processing system including an image processing device that performs specified image processing on a medical image generated by imaging a subject using an imaging device, wherein the image processing device includes a processor that acquires user identification information, acquires first imaging conditions to be applied to the imaging device, acquires a first image that is captured and generated using the imaging device to which the first imaging conditions are applied, sets a first denoising intensity to be applied to denoising the first image, and applies the first denoising intensity to perform denoising on the first image to generate a first denoising image, wherein the processor acquires, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition from among a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging sessions to which a plurality of second imaging conditions were applied for each user, and sets the denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity. It is. Effect of the Invention

[0024] According to the present disclosure, the second signal-to-noise ratio is derived from a plurality of second images generated in a plurality of past imaging operations for each user, and a second signal-to-noise ratio corresponding to a second imaging condition identical to the first imaging condition or a second imaging condition close to the first imaging condition is acquired as a first signal-to-noise ratio from a plurality of second signal-to-noise ratios associated with the second imaging conditions in the past imaging operations. This realizes denoising processing in which a first denoising strength according to a first signal-to-noise ratio preferred by the user is applied. [Brief description of the drawings]

[0025] [Figure 1] FIG. 1 is a diagram showing the overall configuration of an MRI apparatus according to an embodiment. [Diagram 2] FIG. 2 is a functional block diagram showing the electrical configuration of the computer shown in FIG. [Diagram 3] FIG. 3 is a block diagram showing the hardware configuration of the electrical configuration of the computer shown in FIG. [Figure 4] FIG. 4 is a flowchart showing the procedure for constructing the S / N ratio database according to the first example. [Diagram 5] FIG. 5 is a flowchart showing the procedure for constructing the SNR database according to the second example. [Figure 6] FIG. 6 is a flowchart showing the procedure of the image processing method according to the embodiment. [Figure 7] FIG. 7 is a functional block diagram showing a first configuration example of the denoising intensity acquisition unit shown in FIG. [Figure 8] FIG. 8 is an explanatory diagram of the background of the region of interest. [Figure 9] FIG. 9 is an explanatory diagram of pixel values ​​of background noise. [Figure 10] FIG. 10 is a functional block diagram showing a second configuration example of the denoising intensity acquisition unit shown in FIG. [Figure 11]FIG. 11 is a table showing the results of calculating the S / N ratio of a denoised image generated by applying a plurality of denoising intensities to a numerical phantom having a different S / N ratio. [Figure 12] FIG. 12 is an explanatory diagram of a function representing the denoising intensity. [Figure 13] FIG. 13 is a functional block diagram showing the electrical configuration of a computer according to the first modified example. [Figure 14] FIG. 14 is an explanatory diagram showing an example of a user interface applied to feedback of the denoising strength. [Figure 15] FIG. 15 is an explanatory diagram showing a first example of a confirmation screen before automatic adjustment of the denoising intensity. [Figure 16] FIG. 16 is an explanatory diagram showing a second example of a confirmation screen before automatic adjustment of the denoising intensity. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, the same components are denoted by the same reference numerals, and duplicated explanations will be omitted. In addition, when multiple components are exemplified in the following embodiments, it can be interpreted that at least one of the multiple components is included.

[0027] [First embodiment] [Example of MRI device configuration] FIG. 1 is a diagram showing the overall configuration of an MRI apparatus according to an embodiment. Hereinafter, an MRI apparatus will be described as one embodiment of the present disclosure, but the technology of the present disclosure is also applicable to modalities such as an X-ray CT apparatus and a positron emission tomography apparatus. Note that CT is an abbreviation for Computed Tomography. A positron emission tomography apparatus may be referred to as PET, which is an abbreviation for Positron Emission Tomography in English.

[0028] The MRI apparatus 1 shown in Fig. 1 includes a measurement device 10, a calculator 20, an output device 30, an input device 40, and an external storage device 50. The measurement device 10 includes a static magnetic field coil 11, a gradient magnetic field coil 12, and a gradient magnetic field power supply 15. The static magnetic field coil 11 generates a static magnetic field in a space in which a subject is placed. The gradient magnetic field coil 12 applies a magnetic field gradient to the static magnetic field generated by the static magnetic field coil 11. The gradient magnetic field power supply 15 is a drive power supply for the gradient magnetic field coil 12.

[0029] The measurement device 10 includes a transmitting coil 13, a receiving coil 14, a transmitter 16, and a receiver 17. The transmitting coil 13 generates a high-frequency magnetic field for a measurement region of the subject. The transmitter 16 supplies a pulse current, which is an excitation current, to the transmitting coil 13. The receiving coil 14 receives a nuclear magnetic resonance signal generated from the subject. The receiver 17 transmits the nuclear magnetic resonance signal received using the receiving coil 14 to a computer 20. The nuclear magnetic resonance signal may be referred to as an echo signal.

[0030] The MRI apparatus 1 is classified into a vertical magnetic field type and a horizontal magnetic field type according to the direction of the static magnetic field to be generated. Various types of static magnetic field coil 11 are adopted according to the magnetic field type. The gradient magnetic field coil 12 includes a plurality of coils that generate a gradient magnetic field in each of three mutually orthogonal axial directions. Each of the plurality of coils included in the gradient magnetic field coil 12 is driven by a gradient magnetic field power supply 15. Due to the application of the gradient magnetic field, position information is added to the nuclear magnetic resonance signal generated from the subject.

[0031] Although FIG. 1 illustrates an example in which the transmitting coil 13 and the receiving coil 14 are provided separately, there may be an embodiment in which a single coil having the functions of the transmitting coil 13 and the receiving coil 14 is provided.

[0032] The measurement device 10 includes a sequence control device 18. The sequence control device 18 controls the operation of the gradient magnetic field power supply 15 and the transmitter 16 to control the timing of generation of the gradient magnetic field and the radio frequency magnetic field. The sequence control device 18 controls the operation of the receiver 17 to control the timing of reception of the nuclear magnetic resonance signal and executes the measurement. A control time chart applied to the sequence control device 18 is called an imaging sequence, which is set in advance according to the measurement and stored in a storage device or the like provided in the computer 20.

[0033] The calculator 20 is a computer including a processor such as a CUP, a memory, a storage device, and the like. The calculator 20 functions as a control device that controls the operation of each part of the measurement device 10 via the sequence control device 18. The calculator 20 also functions as a calculation device that executes calculation processing on the nuclear magnetic resonance signal received via the receiver 17 and the sequence control device 18 and acquires an image signal of a predetermined imaging region. The calculator 20 may be a device that constitutes the MRI device 1, or may be an external device independent of the MRI device 1.

[0034] The computer 20 is electrically connected to an output device 30, an input device 40, and an external storage device 50 so as to be able to communicate data with each other. The output device 30 may be a display that displays the results of arithmetic processing executed by the computer 20.

[0035] A liquid crystal display, an organic electroluminescence (EL) display, a projector, or the like may be applied to the output device 30. The output device 30 may be an appropriate combination of these. Note that the organic EL may be referred to as OEL, which is an abbreviation of Organic Electro-Luminescence.

[0036] The input device 40 is an interface through which an operator inputs conditions and parameters to be applied to measurement, calculation processing, etc. Examples of the input device 40 include a keyboard and a mouse. A display device functioning as the output device 30 may be integrated with the input device by using a touch panel type display. The operator can use the input device 40 to input parameters such as the number of echoes to be measured, the echo time, and the echo interval.

[0037] The external storage device 50 stores data used in various arithmetic processing executed by the computer 20, data derived as a result of the arithmetic processing, conditions and parameters applied to the arithmetic processing, and programs for executing the arithmetic processing. The external storage device 50 may realize a part of the functions of the internal storage device of the computer 20, or a part of the functions of the external storage device 50 may be realized by the internal storage device of the computer 20.

[0038] The nuclear magnetic resonance signal acquired using the measurement device 10 has noise superimposed on the signal from the subject due to the characteristics of the MRI device 1 and the imaging conditions. The noise superimposed on the nuclear magnetic resonance signal deteriorates the image quality of the captured image of the subject generated from the nuclear magnetic resonance signal. The computer 20 has a function of executing a noise reduction process, which is a process for reducing the noise superimposed on the nuclear magnetic resonance signal, as a signal processing for the nuclear magnetic resonance signal. The noise reduction process may be called a denoising process.

[0039] The denoising process applies a trained learning model such as CNN. CNN is an abbreviation for Convolution Neural Network. The external storage device 50 stores an S / N ratio database used when deriving a filter strength applied to the denoising process. The filter strength may be referred to as a denoising strength.

[0040] The SN ratio database stores the preferred SN ratio of an image for each user, such as a doctor or technician, in association with the imaging conditions applied when imaging a subject. The SN ratio database allows the preferred SN ratio of an image to be read out using the imaging conditions as an index.

[0041] The SN ratio may be referred to as a signal-to-noise ratio. An SN ratio database is omitted in Fig. 1. The SN ratio database is indicated by reference numeral 52 and is shown in Fig. 2. The measuring device 10 shown in Fig. 1 is an example of an imaging device, and the computer 20 is an example of an image processing device, both of which are components of a medical image processing system.

[0042] [Example of computer configuration] Fig. 2 is a functional block diagram showing the electrical configuration of the computer shown in Fig. 1. The computer 20 includes an image acquisition unit 100. The image acquisition unit 100 acquires an image represented by a magnetic resonance signal transmitted from the receiver 17 shown in Fig. 1. Note that the image acquired using the image acquisition unit 100 described in the embodiment shown in Fig. 2 is an example of a first image.

[0043] The computer 20 includes an image reconstruction unit 102. The image reconstruction unit 102 generates a reconstructed image based on an image acquired via the image acquisition unit 100. An example of a reconstructed image generated based on an image is a two-dimensional tomographic image. Here, the term "image" includes the meaning of image data representing an image. The term "data" may include the meaning of an electric circuit as a signal.

[0044] The computer 20 includes a normalization unit 104. The normalization unit 104 applies a prescribed normalization constant to the pixel values ​​of the pixels that form the reconstructed image, and the normalization process is described in detail below.

[0045] The computer 20 includes a user ID acquisition unit 110. The user ID acquisition unit 110 acquires a user ID used to identify a user who performs denoising on a standardized reconstructed image. The user ID is information for identifying a user, such as the user's name and an identification number for each user. The user ID acquisition unit 110 may adopt a mode in which a user ID represented using a barcode, a two-dimensional code, or the like is read using a reading device. The user ID described in the embodiment is an example of user identification information. ID is an abbreviation of Identification.

[0046] The calculator 20 includes an imaging condition acquisition unit 112. The imaging condition acquisition unit 112 acquires imaging conditions to be applied to imaging of a subject. Imaging refers to a function of the measurement device 10 in which an electromagnetic wave is irradiated from the static magnetic field coil 11 or the like to the subject, and a nuclear magnetic resonance signal is acquired using the receiver 17. Examples of imaging conditions include the number of imaging matrices and a sequence. The imaging conditions acquired using the imaging condition acquisition unit 112 described in the embodiment are an example of a first imaging condition.

[0047] The calculator 20 includes an S / N ratio acquisition unit 120. The S / N ratio acquisition unit 120 reads out an S / N ratio corresponding to an imaging target part, a user ID, and imaging conditions from an S / N ratio database 52 in which S / N ratios associated with imaging conditions for each user are stored. The S / N ratios read out from the S / N ratio database 52 are used in arithmetic processing as the user's preferred S / N ratio according to the imaging conditions. The imaging target part can be interpreted as the anatomical structure of the imaging target.

[0048] That is, in the SNR database 52, SNRs measured for MR images generated in multiple past imaging sessions are associated with the imaging sites and imaging conditions associated with the MR images, and stored for each user. The imaging sites and imaging conditions may be in DICOM format. The MR images captured and generated in the past described in the embodiment are an example of a second image, and the imaging conditions of the past images are an example of the second imaging conditions.

[0049] The SN ratio acquisition unit 120 searches the SN ratio database 52 using the imaging target part, the user ID, and the imaging conditions as indexes, and reads out the SN ratio preferred by each user according to these. The SN ratio database 52 described in the embodiment is an example of a signal-to-noise ratio database. The multiple SN ratios stored in the SN ratio database 52 described in the embodiment are an example of multiple second signal-to-noise ratios, and the SN ratio SN0 read out from the SN ratio database 52 is an example of a first signal-to-noise ratio acquired from the multiple second signal-to-noise ratios.

[0050] The computer 20 includes a denoising intensity acquisition unit 122. The denoising intensity acquisition unit 122 acquires a denoising intensity to be applied to the denoising process based on the S / N ratio SN0 read from the S / N ratio database 52. The acquisition of the denoising intensity will be described in detail later. Note that the denoising intensity to be applied to the denoising process described in the embodiment is an example of a first denoising intensity.

[0051] The normalization unit 104 performs normalization processing on the reconstructed image using the SNR acquired from the SNR acquisition unit 120 and the denoising intensity acquired from the denoising intensity acquisition unit 122 .

[0052] The computer 20 includes a denoising processor 130. The denoising processor 130 performs denoising processing on the normalized reconstructed image to generate a denoised image. The denoised image is generated by removing at least a portion of noise from a noisy reconstructed image.

[0053] The denoising processing unit 130 includes a trained learning model 132. A CNN is applied to the learning model 132. The learning model 132 is trained and generated using training data in which an image including noise is used as input data and an image from which noise is removed is used as correct data, so that when an image including noise is input, an image from which noise is removed is output. The generation of the learning model 132 may be executed using a computer independent of the computer 20, or may be executed using a learning model generation unit included in the computer 20. When the computer 20 is equipped with a learning model generation unit, the computer 20 may execute re-learning of the learning model 132. Note that the denoising process applied to the denoising processing unit 130 shown in FIG. 2 is performed, and the denoising image generated is an example of a first denoising image.

[0054] The computer 20 includes a denoised image output unit 140. The denoised image output unit 140 converts the denoised image generated using the denoising processing unit 130 into a signal format applicable to the output device 30, and outputs the converted image to the output device 30. For example, when the output device 30 is a display, the denoised image output unit 140 converts the denoised image into a video signal and outputs the video signal to the output device 30.

[0055] Fig. 3 is a block diagram showing a hardware configuration of the electrical configuration of the calculator shown in Fig. 2. The computer applied to calculator 20 may be a personal computer or a workstation. The computer may be a virtual machine. Calculator 20 may be configured using a plurality of computers.

[0056] The computer 20 includes a processor 202, a memory 204 which is a main storage device, a storage 206 which is an auxiliary storage device, an input / output interface 208, and a bus 210. The processor 202 includes a CPU. The processor 202 may include a GPU. Note that CPU is an abbreviation for Central Processing Unit. GPU is an abbreviation for Graphics Processing Unit.

[0057] The processor 202 is connected to a memory 204, a storage 206, and an input / output interface 208 via a bus 210. The input device 40 and the output device 30 are connected to the computer 20 via the input / output interface 208.

[0058] The processor 202 executes programs stored in the memory 204 to realize various functions. The input / output interface 208 includes a communication interface connectable to a network, a connection interface connectable to an external device, and the like. An example of a network is a local area network in a medical facility. As a connection interface connectable to an external device, for example, a universal serial bus and an HDMI can be applied. The universal serial bus can be referred to as USB, which is an abbreviation of Universal Serial Bus. HDMI is an abbreviation of High-Definition Multimedia Interface. USB and HDMI are registered trademarks.

[0059] The processor 202 communicates with various external devices of the MRI apparatus 1 via the input / output interface 208, and transmits and receives necessary information. Note that illustration of the various external devices is omitted.

[0060] The output device 30 may include a display that displays various information in addition to medical images captured using the MRI apparatus 1. The output device 30 is used as a part of a GUI when receiving input from the input device 40. The output device 30 is not limited to one, and may be in the form of a multi-display having multiple displays. GUI is an abbreviation for Graphical User Interface.

[0061] [Construction of SNR database] Fig. 4 is a flow chart showing the procedure for constructing the S / N ratio database according to the first example. The S / N ratio database 52 shown in Fig. 2 is constructed using an information processing device to which a computer is applied, and the following procedure is applied. The procedure for constructing the S / N ratio database according to the first example is applied when past images are prepared for a plurality of subjects. The past images here refer to MRI images captured and generated in the past.

[0062] In the first information input step S10, previous images are input to the information processing device. In the first information input step S10, a user ID and imaging conditions are input to the information processing device for each of the multiple previous images.

[0063] In the past image selection step S12, selection information of past images that have been subjected to a denoising process preferred by the user is input to the information processing device. For example, in the past image selection step S12, past images that have been subjected to one or more user-preferred denoising processes may be selected from past images that have already been subjected to denoising processes. In the past image selection step S12, past images that have been subjected to one or more user-preferred denoising processes may be selected from a plurality of past images that have been generated by performing denoising processes with different denoising intensities on any one past image.

[0064] In the S / N ratio calculation step S14, the S / N ratio of the past image selected in the past image selection step S12 is calculated. The calculation of the S / N ratio will be described in detail later. The S / N ratio may be calculated by a known method.

[0065] In the S / N ratio storage step S16, the S / N ratio calculated in the S / N ratio calculation step S14 is associated with the imaging region and imaging conditions, and stored for each user. In this manner, the S / N ratio database 52 is constructed and stored in the external storage device 50 shown in Fig. 1. The procedure for constructing the S / N ratio database 52 shown in Fig. 4 may also be applied as a procedure for updating the S / N ratio database 52.

[0066] 5 is a flowchart showing the procedure for constructing the SNR database according to the second example. The procedure for constructing the SNR database according to the second example is applied when multiple past images with different denoising intensities are prepared for the same subject.

[0067] In the procedure for constructing the S / N ratio database shown in Fig. 5, a second information input step S11 is executed instead of the first information input step S10 shown in Fig. 4. In the second information input step S11, an imaging region, a user ID, and imaging conditions are input for each of a plurality of past images of the same subject.

[0068] After the second information input step S11, a past image selection step S12, an S / N ratio calculation step S14, and an S / N ratio storage step S16 are executed to construct the S / N ratio database 52. The processing of the past image selection step S12, the S / N ratio calculation step S14, and the S / N ratio storage step S16 is similar to the procedure for constructing the S / N ratio database according to the second example shown in Fig. 4, and therefore a description thereof will be omitted here.

[0069] [Detailed explanation of denoising process] 6 is a flowchart showing the procedure of the image processing method according to the embodiment. In the image processing method applied to the MRI apparatus 1 according to the embodiment, the denoising intensity applied to the denoising process is automatically set by referring to the S / N ratio of MR images captured in the past by the same user. The image processing method according to the embodiment can be understood as a method of operating the image processing device.

[0070] In the third information input step S100, the image acquisition unit 100 shown in Fig. 2 acquires an image to be processed. Information on the imaging site can be acquired when acquiring the image to be processed. Also, in the third information input step S100, the user ID acquisition unit 110 acquires the user ID of the user who performed the imaging to acquire the image to be processed. Furthermore, in the third information input step S100, the imaging condition acquisition unit 112 acquires the imaging conditions applied to the imaging to acquire the image to be processed.

[0071] In the S / N ratio acquisition step S102, the S / N ratio acquisition unit 120 uses the user ID and the imaging conditions acquired in the third information input step S100 as indexes, and refers to the S / N ratio database 52 to acquire the user's preferred S / N ratio SN0 from the S / N ratio database 52. The user's preferred S / N ratio SN0 is an example of the first signal-to-noise ratio.

[0072] If the SN ratio according to the user ID and the imaging conditions is not stored in the SN ratio database 52, the SN ratio of a past image of the same user with similar imaging conditions may be acquired. Examples of similar imaging conditions include imaging conditions in which a part of a plurality of components is the same, and imaging conditions in which the difference in index value is within a specified range. In other words, similar imaging conditions can be understood as those having differences in the SN ratio setting and the denoising intensity setting to such an extent that similar results can be obtained as with the same imaging conditions.

[0073] In the denoising process step S104, the normalization unit 104 applies the S / N ratio SN0 acquired in the S / N ratio acquisition step S102 and the denoising intensity corresponding to the S / N ratio SN0 to perform normalization processing on the image to be denoised. Also, in the denoising process step S104, the denoising processor 130 performs denoising processing on the image that has been subjected to normalization processing using the learning model 132. AI denoising shown in FIG. 6 represents denoising processing using the learning model 132. Note that AI is an abbreviation for artificial intelligence.

[0074] In a denoised image output step S106, the denoised image output unit 140 converts the denoised image that has been subjected to the denoising process into the output format of the output device 30, and outputs it to the output device 30. When the denoised image is output in the denoised image output step S106, the procedure of the image processing method is terminated.

[0075] [Example of denoising strength adjustment] [Multi-stage type] Fig. 7 is a functional block diagram showing a first configuration example of the denoising intensity acquisition unit shown in Fig. 2. The denoising intensity acquisition unit 122 includes a denoising image generation unit 212, an S / N ratio calculation unit 213, an S / N ratio comparison unit 214, and a denoising intensity output unit 215.

[0076] The denoising image generating unit 212 applies different denoising intensities to the reconstructed image Ir generated by the image reconstructing unit 102 shown in FIG. 2 to generate a plurality of provisional denoising images. a , the denoising intensity D b and denoising intensity D c is applied to generate three types of temporary denoised images.

[0077] The SN ratio calculation unit 213 calculates the SN ratio of each of the multiple denoised images generated by the denoised image generation unit 212. a , the denoising intensity D b and denoising intensity D c The signal-to-noise ratio of the denoised image to which each of a , S.N. b and S.N. c An example in which is calculated is illustrated.

[0078] The SN ratio comparison unit 214 compares the SN ratio SN0 acquired as the user's preferred SN ratio corresponding to the imaging conditions with the SN ratio SN calculated by the SN ratio calculation unit 213. a , SNR b and SN ratio c are compared. a , SNR b and SN ratio c If any of the SNRs is equal to the SNR SN0, the SNR that is equal to the SNR SN0 is determined as the comparison result.

[0079] Signal to noise ratio a , SNR b and SN ratio c If the SN ratio SN0 does not match any of the above, a , SNRb and SN ratio c Among these, the S / N ratio closest to the S / N ratio SN0 is used as the comparison result. The S / N ratio closest to the S / N ratio SN0 is the S / N ratio whose absolute value of the difference from the S / N ratio SN0 is the smallest.

[0080] The denoising intensity output unit 215 applies the SN ratio, which is the comparison result of the SN ratio comparison unit 214, to the calculated provisional denoising image, and outputs the denoising intensity. a If the SNR is equal to or closest to the SNR SN0, the denoise intensity output unit 215 outputs the denoise intensity D a Output.

[0081] [Example of SN ratio calculation method] A description will be given of a specific example of an S / N ratio calculation method applied to the S / N ratio calculation unit 213. In an MR image, the S / N ratio can be calculated by utilizing the characteristic that the background noise level is proportional to the noise level of the entire MR image.

[0082] FIG. 8 is an explanatory diagram of the background of a region of interest. FIG. 8 illustrates a region of interest (Roi) and background (Bg) in a denoised image Id, which is a reconstructed image to which AI denoising has been applied. FIG. 9 is an explanatory diagram of pixel values ​​of background noise. FIG. 9 illustrates a histogram of pixel values ​​of the denoised image Id. The horizontal axis of a graph showing a histogram of pixel values ​​of the denoised image Id represents pixel value, and the vertical axis represents frequency.

[0083] The SN ratio calculation unit 213 shown in Fig. 7 extracts a region of interest Roi from a denoised image Id shown in Fig. 8, extracts a background Bg of the region of interest Roi, and estimates a noise pixel value of the background Bg. The background Bg of the region of interest Roi is understood as an air region surrounding an organ or the like which is the region of interest Roi. The pixel value Nv of the noise of the background Bg shown in Fig. 9 is applied as a pixel value having a peak frequency in the histogram of pixel values ​​of the denoised image Id. The noise pixel value Nv is understood as a noise signal value.

[0084] The SN ratio calculation unit 213 calculates the median of pixel values ​​that exceed 0 in the denoised image Id as the signal value Sv when calculating the SN ratio. The signal value Sv may be the average value of pixel values ​​that exceed 0 in the denoised image Id. The SN ratio calculation unit 213 calculates Sv / Nv as the SN ratio of the denoised image Id.

[0085] The multiple provisional denoised images generated by the denoised image generating unit 212 shown in Fig. 7 are an example of a third denoised image. a , SNR b and SN ratio c is an example of a third signal-to-noise ratio, and the denoise intensity D a is an example of a third denoising intensity.

[0086] [Adjustment of denoising strength in normalization processing] The denoising intensity is adjusted in the normalization process in the normalization unit 104 shown in FIG. 2 (Patent Application No. 2022-34774). 0 The output image output from the normalization unit 104 is I n Let us assume that.

[0087] Input image I 0 The median of the pixel values ​​that are greater than 0 in 0 ) and the pixel value of the background noise is Nv, the signal-to-noise ratio (SNR) is SNR=Minput(I 0 ) / Nv.

[0088] If the denoising intensity is D, the output image I n I n ={1 / Minput(I 0 )}×{a×(SNR / D)+b}×I 0 Here, a and b are prescribed constants derived by carrying out experiments, etc.

[0089] For example, if you adjust the denoising strength to three levels, weak, medium, and strong, the output image I nThe value of the denoising strength D, which functions as a coefficient for determining the denoising strength in, is set to 0.5, 1.0, and 1.5.

[0090] [Stepless type] Fig. 10 is a functional block diagram showing a second configuration example of the denoising intensity acquisition unit shown in Fig. 2. The denoising intensity acquisition unit 122 shown in Fig. 2 includes an S / N ratio calculation unit 220 and a denoising intensity calculation unit 222. The S / N ratio calculation unit 220 calculates the S / N ratio SNIr of the reconstructed image Ir generated using the image reconstruction unit 102 shown in Fig. 2.

[0091] The denoising intensity calculation unit 222 calculates a denoising intensity D expressed as a function f(SNIr, SN0) having the S / N ratio SNIr of the reconstructed image Ir and the user's preferred S / N ratio SN0 as parameters. The denoising intensity D calculated by the denoising intensity calculation unit 222 is applied to the normalization process of the reconstructed image Ir performed by the normalization unit 104.

[0092] The reconstructed image Ir that has been subjected to the standardization process is subjected to denoising process in the denoising processor 130 using the learning model 132 shown in FIG.

[0093] 10 is an example of a fourth signal-to-noise ratio. Also, the denoise intensity D shown in the figure is an example of a fourth denoise intensity.

[0094] [Example of calculation of denoising intensity] Fig. 11 is a table showing the results of calculating the S / N ratio of a denoised image generated by applying a plurality of denoising intensities to a numerical phantom with a different S / N ratio. The table shown in Fig. 11 illustrates a case where the number of S / N ratios of the numerical phantom is M and the number of denoising intensities is N. Each of M and N is an integer of 1 or more.

[0095] The signal-to-noise ratio (SN) of the numerical phantom shown in Fig. 11 1 From SN ratio MFor each of the above, a value within the range of the S / N ratio of the reconstructed image Ir to be subjected to denoising is applied. 1 to denoise intensity D N For each of the above, values ​​within the range used in the denoising process for the reconstructed image Ir are applied.

[0096] FIG. 12 is an explanatory diagram of a function that represents the denoising intensity. In FIG. 12, the SN ratio of the numerical phantom is SN 1 When the denoising strength is D 1 From D 5 For the range up to, the denoising strength D = Σ(a i ×SN i ) is shown as a graph. Note that i is an index from 11 to MN, and a i is a predefined constant.

[0097] The signal-to-noise ratio (SN) shown in Fig. 11 1 From SN ratio M 12 is stored in advance for each of the above. The function representing the denoise intensity D may be stored in a storage device included in the computer 20, or may be stored in an external device such as the external storage device 50. The function representing the denoise intensity D shown in the figure is an example of a function representing the relationship between the signal-to-noise ratio and the denoise intensity.

[0098] The denoising intensity calculation unit 222 shown in FIG. 10 calculates the signal-to-noise ratio (SN) shown in FIG. 1 From SN ratio M The denoising intensity calculation unit 222 searches for an SN ratio that is equal to the SN ratio SNIr of the reconstructed image Ir calculated by the SN ratio calculation unit 220 or that has the smallest absolute value of the difference from SNIr. The denoising intensity calculation unit 222 calculates a denoising intensity D=Σ(a i ×SN i ) is used to determine the value of the denoising intensity D that results in the signal-to-noise ratio SN0 desired by the user.

[0099] [First Modification of the Embodiment] Fig. 13 is a functional block diagram showing the electrical configuration of a computer according to the first modified example. A computer 20A shown in Fig. 13 has a function of updating an SN database stored in an SN ratio database 52. That is, the computer 20A includes a denoising intensity update unit 150 that updates the SN database.

[0100] The denoising intensity update unit 150 updates the user's preferred S / N ratio stored in the S / N ratio database 52 when it receives an input signal that is input using the input device 40 and indicates that the denoising intensity is to be updated.

[0101] 14 is an explanatory diagram showing an example of a user interface applied to feedback of denoising intensity. A viewer screen 300 shown in FIG. 14 is displayed using a display which is the output device 30 shown in FIG.

[0102] The viewer screen 300 includes an image display area 302 in which a reconstructed image Ir, which is a denoised image Id that has been subjected to denoising processing, is displayed, and an evaluation display area 304 in which an evaluation of the denoising strength is displayed. In Fig. 14, the evaluations of the denoising strength are displayed as "good," "weak," and "strong," and a check box 306 is displayed for each of the three types of evaluation.

[0103] A user who views the denoised image Id displayed on the viewer screen 300 can use the mouse constituting the input device 40 to check any of a plurality of check boxes 306 displayed in the evaluation display area 304. The user's check is fed back for the next and subsequent imaging.

[0104] In the next or subsequent imaging, if the denoising intensity is changed according to the feedback result and an appropriate denoising intensity is selected for the changed denoising intensity, the user's preferred S / N ratio stored in the S / N ratio database 52 is automatically updated. That is, the S / N ratio of the denoising image evaluated as appropriate for the denoising intensity is stored in the S / N ratio database 52 as the new user's preferred S / N ratio.

[0105] Note that the denoised image evaluated as being just right for the denoising intensity described in the embodiment is an example of a second denoising image. The denoising intensity applied to the second denoising image is an example of a second denoising intensity. The output device 30 that displays the viewer screen 300 shown in FIG. 14 is an example of a display device, and the check box 306 shown in the same figure is an example of a choice that represents the evaluation of the first denoising intensity. The signal that represents the evaluation of the first denoising intensity described in the embodiment is an example of a first evaluation signal.

[0106] [Second Modification of the Embodiment] The computer according to the second modified example has a function of displaying a preview of the denoising image Id when automatically adjusting the denoising intensity. If there is no problem with the denoising intensity of the denoising image Id displayed as the preview, imaging is started. On the other hand, if there is a problem with the denoising intensity, the user changes the denoising intensity. When the denoising intensity is changed, the user's preferred SN ratio stored in the SN ratio database 52 is automatically updated.

[0107] 15 is an explanatory diagram showing a first example of a confirmation screen before automatic adjustment of the denoising strength. A confirmation screen 330 according to the first example is configured to include an image display area 332 in which a preview of the denoising image Id is displayed, a denoising strength evaluation input area 334 in which an evaluation of the denoising strength is input, and an OK button 336 that is clicked to confirm the denoising strength evaluation input. The denoising strength evaluation input area 334 includes five check boxes 338 that are checked when one of five levels of denoising strength is selected.

[0108] 15 is a past MR image obtained by imaging an arbitrary healthy subject, and is a past MR image obtained by applying imaging conditions that are the same as or similar to the imaging conditions of the current imaging to an MR image of the same part as the current imaging target. The denoised image shown in the figure has been subjected to denoising processing in which an automatically adjusted denoising intensity is applied.

[0109] A user viewing the preview of the denoised image Id displayed in the image display area 332 can change the denoising intensity by checking any of the five check boxes 338 and then clicking the OK button 336. On the other hand, if there is no problem with the denoising intensity of the preview of the denoised image Id, the user can click the OK button 336 without changing the checked check box 338, maintain the automatically set denoising intensity, and start imaging.

[0110] 15 is an example of a fifth denoise image, and is an example of a normal image obtained by previously imaging a normal imaging target. The imaging conditions that are the same as or close to the imaging conditions for the current imaging described in the embodiment are an example of a fifth imaging condition that is the same as the first imaging condition, or a fifth imaging condition that is close to the first imaging condition. The evaluation signal when the user changes the check described in the embodiment is an example of a fifth evaluation signal, and the denoise intensity when the user changes the check is an example of a fifth denoise intensity.

[0111] Fig. 16 is an explanatory diagram showing a second example of a confirmation screen before automatic adjustment of the denoising intensity. The image display area 332 shown in Fig. 16 displays a denoising image Id that has been subjected to denoising processing in which a denoising intensity that is automatically adjusted according to the user's preferred S / N ratio is applied to MR images that have been previously captured for the same subject as the subject to be imaged and that have been captured under the same or similar imaging conditions.

[0112] The denoising strength evaluation input area 334 displays a check mark for the denoising strength to be applied to the denoising image Id displayed in the image display area 332. The user can look at the denoising image Id displayed in the image display area 332 and determine whether the denoising strength is appropriate.

[0113] When the user determines that the denoising intensity is appropriate and clicks the OK button 336, the automatically set denoising intensity is maintained and imaging is started. On the other hand, when the user determines that a change in the denoising intensity is necessary and an unchecked checkbox is checked, the denoising image Id to which the checked denoising intensity is applied is redisplayed.

[0114] If the denoising intensity applied to the redisplayed denoised image Id is determined to be appropriate, imaging is started and the user preferred SNR stored in the SNR database 52 is automatically updated.

[0115] In addition, a subject that is the same as the subject of the imaging target described in the embodiment is an example of the same imaging target. The MR image obtained by imaging in the past shown in Fig. 16 is an example of the same imaging target image, and the denoised image obtained by performing denoising on the MR image obtained by imaging in the past is an example of a sixth denoised image.

[0116] The denoising intensity applied to the denoising process of the MR image captured and generated in the past shown in Fig. 16 is an example of a sixth denoising intensity. The imaging condition applied to the MR image captured and generated in the past shown in the same figure is an example of a sixth imaging condition that is the same as the first imaging condition or a sixth imaging condition that is close to the first imaging condition. The evaluation signal when the user changes the check described in the embodiment is an example of a sixth evaluation signal.

[0117] [Effects of the embodiment] The MRI apparatus 1 according to the embodiment can provide the following advantageous effects.

[0118] [1] For each user, such as a doctor, the user's preferred S / N ratio SN0 is obtained by referring to the S / N ratio of a reconstructed image Ir that was previously captured and generated, and a denoising intensity D to be applied to the denoising process on a newly captured reconstructed image Ir is automatically set in accordance with the user's preferred S / N ratio SN0.

[0119] As a result, denoising processing is performed in which a denoising strength D is applied according to a pre-stored user-preferred SNR SN0, and a desirable denoised image Id is acquired.

[0120] [2] When the denoising intensity D is automatically set, the user's preferred SN ratio SN0 is acquired, and from the multiple denoising intensities, a denoising intensity D applied to a denoising image Id having an SN ratio identical to the user's preferred SN ratio SN0, or a denoising intensity D applied to a denoising image Id having an SN ratio close to the user's preferred SN ratio SN0, is determined. As a result, denoising processing is performed in which a denoising intensity D according to the user's preferred SN ratio SN0 is applied from a plurality of denoising intensities D prepared in advance, and a preferred denoising image Id is acquired.

[0121] [3] When the denoising intensity D is automatically set, the user's preferred S / N ratio is acquired, and a function f(SNIr, SN0) that represents the denoising intensity with the S / N ratio as a parameter is applied to determine the denoising intensity D according to the user's preferred S / N ratio SN0. This executes denoising processing in which the denoising intensity D according to the user's preferred S / N ratio is applied from among the denoising intensity D represented using the function f(SNIr, SN0) with the S / N ratio as a parameter, and a desirable denoising image Id is acquired.

[0122] [4] A preview of the denoised image Id is displayed using a display functioning as the output device 30. The viewer screen 300 on which the preview is displayed has a checkbox 306 for inputting an evaluation of the denoising strength D of the denoised image Id.

[0123] This allows a user who has viewed a preview of the denoised image Id to click on the check box 306 and input an evaluation of the denoising strength D of the denoised image Id.

[0124] [5] When an evaluation that the denoising intensity is just right is input for a denoising image Id whose denoising intensity D has been changed, the user's preferred SN ratio stored in the SN ratio database 52 is updated. This causes the user's preferred SN ratio stored in the SN ratio database 52 to be updated in accordance with the user's evaluation of the denoising intensity.

[0125] [6] A denoised image Id, which has been subjected to denoising processing in which the denoising intensity D is automatically set for a past image of the same part obtained by imaging a healthy subject in the past, and the denoising intensity D are displayed on a confirmation screen 330, which is a setting screen for imaging conditions before imaging. If there is no problem with the denoising intensity D, imaging is started. On the other hand, if there is a problem with the denoising intensity D, the denoising intensity D is changed. If the denoising intensity D is changed, the user's preferred SN ratio stored in the SN ratio database 52 is updated.

[0126] This allows the user to check the denoising intensity in advance. In addition, the user's preferred S / N ratio stored in the S / N ratio database 52 is updated in response to a change in the denoising intensity.

[0127] [7] A confirmation screen 330, which is a setting screen for the imaging conditions before imaging, displays denoised images Id and denoising intensity D of past images of the subject to be imaged that are applied with the same imaging conditions as the current imaging conditions or that are applied with imaging conditions similar to the current imaging conditions.

[0128] This allows the user to check the denoising strength applied to the denoising process of the previous image, and allows the user to determine whether the denoising strength D applied to the current image capture is appropriate.

[0129] [8] The confirmation screen 330 displays a check box for each of a plurality of denoising intensity values ​​D. This allows the user to input a command to change the denoising intensity D by checking the check box using the input device 40.

[0130] [9] When the denoising intensity D is changed, the user's preferred SN ratio stored in the SN ratio database 52 is updated. As a result, the user's preferred SN ratio stored in the SN ratio database 52 is updated in response to the change in the denoising intensity.

[0131] [Hardware configuration of each processing unit] The hardware structure of the processing units that execute various processes, such as the image acquisition unit 100, image reconstruction unit 102, normalization unit 104, user ID acquisition unit 110, imaging condition acquisition unit 112, S / N ratio acquisition unit 120, denoising intensity acquisition unit 122, denoising processing unit 130, and denoising image output unit 140 in the computer 20 shown in FIG. 2 and the computer 20A shown in FIG. 13, is, for example, various processors as shown below.

[0132] Various types of processors include CPUs, which are general-purpose processors that execute programs and function as various processing units, GPUs, programmable logic devices (PLDs), such as FPGAs (Field Programmable Gate Arrays), which are processors whose circuit configuration can be changed after manufacture, and dedicated electrical circuits, such as ASICs (Application Specific Integrated Circuits), which are processors with a circuit configuration designed specifically to execute specific processes.

[0133] One processing unit may be composed of one of these various processors, or may be composed of two or more processors of the same type or different types. For example, one processing unit may be composed of multiple FPGAs, or a combination of a CPU and an FPGA, or a combination of a CPU and a GPU. Also, multiple processing units may be composed of one processor. As an example of multiple processing units being composed of one processor, first, as represented by a computer such as a client or a server, one processor is composed of a combination of one or more CPUs and software, and this processor functions as multiple processing units. Second, as represented by a system on chip (SoC), there is a form in which a processor is used that realizes the functions of the entire system including multiple processing units in one IC (Integrated Circuit) chip. In this way, the various processing units are composed of one or more of the above various processors as a hardware structure.

[0134] Furthermore, the hardware structure of these various processors is, more specifically, an electric circuit that combines circuit elements such as semiconductor elements.

[0135] [About the programs that run the computer] A program that causes a computer to realize some or all of the processing functions of the computer 20 in the embodiment and the computer 20A in the modified example of the embodiment can be recorded on a computer-readable medium such as an optical disk, a magnetic disk, a semiconductor memory, or other tangible, non-transitory information storage medium, and the program can be provided through this information storage medium.

[0136] In addition, instead of providing the program by storing it on such a tangible, non-transitory computer-readable medium, it is also possible to provide the program signal as a download service using a telecommunications line such as the Internet.

[0137] Furthermore, a part or all of the processing functions in the computer 20 etc. may be realized by cloud computing, and may also be provided as SaaS (Software as a Service).

[0138] The present disclosure is not limited to the above-described embodiment, and various modifications are possible without departing from the spirit and scope of the technical idea of ​​the present disclosure. [Explanation of symbols]

[0139] 20...Calculator 30...Output device 50...External storage device 52…Signal-to-noise ratio database 100: Image acquisition unit 102...Image reconstruction unit 104…Standardization Department 110...User ID acquisition section 112...imaging condition acquisition unit 120...SN ratio acquisition section 122…Denoise intensity acquisition unit 130... Denoising processing unit 132…Learning Model 140... Denoise image output unit

Claims

1. Obtaining user identification information; Obtaining a first imaging condition to be applied to the imaging device; acquiring a first image captured and generated using the imaging device to which the first imaging condition is applied; setting a first denoising intensity to be applied to the first image; a processor for applying the first denoising intensity to perform a denoising process on the first image to generate a first denoised image; The processor, obtain, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition from a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions are applied for each user; The image processing device sets a denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

2. The processor, 2. The image processing device according to claim 1, wherein a signal-to-noise ratio database in which a plurality of the second imaging conditions are associated with and stored for each of the plurality of second signal-to-noise ratios is referenced, and the first signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition is acquired.

3. The processor, The image processing apparatus according to claim 2 , wherein when the first denoising intensity is changed, the signal-to-noise ratio database is updated in accordance with the second denoising intensity after the change.

4. The processor, performing a denoising process with different denoising intensities on the first image multiple times to generate a plurality of third denoising images; deriving a third signal to noise ratio from each of the third denoised images; The image processing device according to claim 1 , wherein a third denoising intensity applied to the third denoising image from which the third signal-to-noise ratio that has a minimum difference from the first signal-to-noise ratio is set as the first denoising intensity.

5. The processor, calculating a fourth signal to noise ratio from the first image; The image processing apparatus according to claim 1 , wherein a fourth denoising intensity corresponding to a fourth signal-to-noise ratio is set as the first denoising intensity using a function representing a relationship between a signal-to-noise ratio and a denoising intensity.

6. The processor, Displaying the first denoised image on a display device; displaying on the display device an option representing an evaluation of the first denoising strength applied to the denoising process for the first denoising image; obtaining a first evaluation signal representative of an evaluation of the first denoising strength selected from the options; The image processing apparatus according to claim 1 , further comprising: feeding back an evaluation of the first denoising intensity represented by the first evaluation signal to a setting of the first denoising intensity.

7. The processor, a fifth imaging condition that is the same as the first imaging condition or a fifth imaging condition that is close to the first imaging condition is applied, and a denoising process is performed by applying the first denoising intensity to a past image that has been captured and generated in the past, and the generated fifth denoised image is displayed on a display device; obtaining a fifth evaluation signal representing an evaluation of a fifth denoising strength applied to the fifth denoising image; The image processing apparatus according to claim 1 , further comprising: determining whether to maintain or change the first denoising intensity in response to an evaluation of the fifth denoising intensity represented by the fifth evaluation signal.

8. The image processing device according to claim 7 , wherein the previous image is a normal image generated by previously capturing an image of a normal imaging target.

9. A sixth imaging condition that is the same as the first imaging condition or a sixth imaging condition that is close to the first imaging condition is applied to the past image, and an identical imaging target image generated by imaging an identical imaging target in the past is applied; The processor, The image processing device according to claim 7 , further comprising a display device that displays a sixth denoised image obtained by performing denoising on the same captured image, and a sixth denoising intensity applied to the denoising performed on the same captured image.

10. The processor, obtaining a sixth evaluation signal representative of an evaluation of the sixth denoising strength; The image processing apparatus according to claim 9 , further comprising: determining whether to maintain or change the first denoising strength in response to an evaluation of the sixth denoising strength represented by the sixth evaluation signal.

11. A method for operating an image processing device using a computer including a processor, comprising: The processor, Obtaining user identification information; Obtaining a first imaging condition to be applied to the imaging device; acquiring a first image captured and generated using the imaging device to which the first imaging condition is applied; setting a first denoising intensity to be applied to the first image; applying the first denoising intensity to perform a denoising process on the first image to generate a first denoised image; When setting the first denoising intensity, obtain, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition from a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions are applied for each user; A method for operating an image processing apparatus, comprising: setting a denoising strength derived based on the first signal-to-noise ratio as the first denoising strength.

12. On the computer, Ability to obtain user identification information; A function of acquiring a first imaging condition to be applied to the imaging device; a function of acquiring a first image captured and generated using the imaging device to which the first imaging condition is applied; A function of setting a first denoising strength applied to the denoising process for the first image; and A program for implementing a function of applying the first denoising intensity to perform denoising on the first image to generate a first denoised image, The function for setting the first denoising strength is obtain, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition from a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions are applied for each user; A program that sets a denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

13. A medical image processing system including an image processing device that performs prescribed image processing on medical images generated by imaging a subject using an imaging device, The image processing device includes: Obtaining user identification information; Obtaining a first imaging condition to be applied to the imaging device; acquiring a first image captured and generated using the imaging device to which the first imaging condition is applied; setting a first denoising intensity to be applied to the first image; a processor for applying the first denoising intensity to perform a denoising process on the first image to generate a first denoised image; The processor, obtain, as a first signal-to-noise ratio, a second signal-to-noise ratio corresponding to a second imaging condition that is the same as the first imaging condition or a second imaging condition that is close to the first imaging condition from a plurality of second signal-to-noise ratios derived from each of a plurality of second images generated in a plurality of past imaging operations to which a plurality of second imaging conditions are applied for each user; A medical image processing system that sets a denoising intensity derived based on the first signal-to-noise ratio as the first denoising intensity.

Images