WO2022071181A1 - 情報処理装置、情報処理方法、プログラム及びモデル生成方法 - Google Patents
情報処理装置、情報処理方法、プログラム及びモデル生成方法 Download PDFInfo
- Publication number
- WO2022071181A1 WO2022071181A1 PCT/JP2021/035259 JP2021035259W WO2022071181A1 WO 2022071181 A1 WO2022071181 A1 WO 2022071181A1 JP 2021035259 W JP2021035259 W JP 2021035259W WO 2022071181 A1 WO2022071181 A1 WO 2022071181A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- region
- plaque
- baden
- blood vessel
- information processing
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 title claims description 47
- 238000003672 processing method Methods 0.000 title claims description 3
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 82
- 238000004364 calculation method Methods 0.000 claims abstract description 74
- 238000003384 imaging method Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 40
- 238000002059 diagnostic imaging Methods 0.000 description 97
- 230000003902 lesion Effects 0.000 description 40
- 238000012545 processing Methods 0.000 description 30
- 230000002792 vascular Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 16
- 238000004891 communication Methods 0.000 description 10
- 238000010801 machine learning Methods 0.000 description 10
- 238000012986 modification Methods 0.000 description 10
- 230000004048 modification Effects 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 230000001225 therapeutic effect Effects 0.000 description 6
- 238000002591 computed tomography Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 208000031481 Pathologic Constriction Diseases 0.000 description 4
- 238000002608 intravascular ultrasound Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000036262 stenosis Effects 0.000 description 4
- 208000037804 stenosis Diseases 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 230000012447 hatching Effects 0.000 description 3
- 238000012014 optical coherence tomography Methods 0.000 description 3
- 238000013146 percutaneous coronary intervention Methods 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 244000208734 Pisonia aculeata Species 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000012528 membrane Substances 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002966 stenotic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/50—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications
- A61B6/504—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment specially adapted for specific body parts; specially adapted for specific clinical applications for diagnosis of blood vessels, e.g. by angiography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
- A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0891—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of blood vessels
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/12—Diagnosis using ultrasonic, sonic or infrasonic waves in body cavities or body tracts, e.g. by using catheters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/174—Segmentation; Edge detection involving the use of two or more images
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/24—Indexing scheme for image data processing or generation, in general involving graphical user interfaces [GUIs]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10068—Endoscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Definitions
- the present invention relates to an information processing apparatus, an information processing method, a program, and a model generation method.
- Patent Document 1 There is a technology to support diagnostic imaging related to vascular treatment such as PCI (Percutaneous Coronary Intervention).
- PCI Percutaneous Coronary Intervention
- Patent Document 1 it is a medical image processing device that converts 3D CT (Computer Tomography) data into a 2D projected image, extracts a plaque region of a blood vessel from the 3D CT data, and calculates a stenosis rate of the blood vessel. Then, a medical image processing apparatus that displays a projected image in which a color value of a plaque region is determined according to a calculated stenosis rate or the like is disclosed.
- 3D CT Computer Tomography
- the invention according to Patent Document 1 extracts the plaque region from the three-dimensional CT value on a rule basis, and cannot always accurately extract the plaque region and calculate the stenosis rate. Further, the invention according to Patent Document 1 requires three-dimensional CT data, and does not calculate the squeezing rate or the like from the two-dimensional data (image).
- One aspect is to provide an information processing device or the like that can suitably support image diagnosis related to vascular treatment.
- the information processing apparatus includes an acquisition unit that acquires a multi-frame cross-layer image of a patient's blood vessel, and a blood vessel along the axial direction of the blood vessel based on the multi-frame cross-layer image.
- Each region is provided with a determination unit that determines the region where both ends of the treatment range of the blood vessel by a predetermined treatment device should be located.
- diagnostic imaging related to vascular treatment can be suitably supported.
- FIG. It is a flowchart which shows the procedure of plaque Baden calculation processing. It is explanatory drawing about the identification process of the lesion area which concerns on modification 1.
- FIG. It is a block diagram which shows the structural example of the diagnostic imaging apparatus which concerns on Embodiment 2.
- FIG. It is explanatory drawing about the calculation process of plaque Baden which concerns on Embodiment 2.
- FIG. It is a flowchart which shows the procedure of the generation process of a discriminative model. It is a flowchart which shows the procedure of the calculation process of plaque Baden which concerns on Embodiment 2.
- FIG. It is explanatory drawing about the generation processing of the calculation model which concerns on modification 2.
- FIG. It is a flowchart which shows the procedure of the generation process of the calculation model which concerns on modification 2.
- FIG. 1 is an explanatory diagram showing a configuration example of an diagnostic imaging system.
- a diagnostic imaging system that calculates Plaque Burden from a cross-sectional layer image of a patient's blood vessel and supports vascular treatment using a predetermined treatment device will be described.
- the diagnostic imaging device 2 includes a server 1 and a diagnostic imaging device (information processing device) 2. Each device is communicated and connected via a network N such as the Internet.
- Server 1 is a server computer capable of various information processing and transmission / reception of information.
- the device corresponding to the server 1 may be a device such as a personal computer.
- the server 1 functions as a generation device that performs machine learning to learn predetermined training data and uses a patient's vascular tomographic image as an input to generate a calculation model 51 (see FIG. 4) for calculating plaque baden.
- the plaque baden is the area ratio of the plaque to the cross section of the blood vessel, and the cross-sectional area of the plaque and the media is divided by the blood vessel cross-sectional area (the area of the region in the EEM (External Elastic Membrane)). It is a value (see FIG. 4).
- the diagnostic imaging device 2 is an imaging device that captures a medical image of a patient's blood vessel, and is, for example, an IVUS (Intravascular Ultrasound) device that performs an ultrasonic examination using a catheter 201.
- the catheter 201 is a medical instrument inserted into a patient's blood vessel, and an ultrasonic signal is transmitted from an ultrasonic probe attached to the tip of the catheter 201 and a reflected wave is received.
- the diagnostic imaging apparatus 2 generates and displays an ultrasonic tomographic image based on the reflected wave received by the catheter 201.
- the diagnostic imaging apparatus 2 is described as an IVUS apparatus in the present embodiment, it may be an optical imaging apparatus or the like by OCT (Optical Coherence Tomography). Further, the vascular tomographic image may be an image taken by a method other than IVUS and OCT.
- OCT Optical Coherence Tomography
- the data of the calculation model 51 generated by the server 1 is installed in the diagnostic imaging apparatus 2, and the diagnostic imaging apparatus 2 inputs the vascular tomographic image captured by the catheter 201 into the calculation model 51 to calculate the plaque baden. do. Then, the diagnostic imaging apparatus 2 presents the calculated plaque baden to the user (medical worker). Further, the diagnostic imaging apparatus 2 identifies the lesion region of the blood vessel based on the calculated plaque baden, and supports the determination of the indwelling position in which the stent (therapeutic device) to be indwelled in the blood vessel is indwelled.
- the diagnostic imaging apparatus 2 calculates the plaque baden using the calculation model 51, but the server 1 on the cloud may execute the process using the calculation model 51. Further, for example, a general-purpose computer connected to the diagnostic imaging apparatus 2 may perform the processing. As described above, the processing entity that executes a series of processing is not particularly limited.
- a stent is given as an example of a therapeutic device, but as will be described later, a balloon that dilates a blood vessel, a rotablator that cuts a stenotic lesion of a blood vessel, or the like may be used.
- FIG. 2 is a block diagram showing a configuration example of the server 1.
- the server 1 includes a control unit 11, a main storage unit 12, a communication unit 13, and an auxiliary storage unit 14.
- the control unit 11 has an arithmetic processing unit such as one or a plurality of CPUs (Central Processing Units), MPUs (Micro-Processing Units), GPUs (Graphics Processing Units), and stores the program P1 stored in the auxiliary storage unit 14. By reading and executing, various information processing, control processing, etc. are performed.
- the main storage unit 12 is a temporary storage area for SRAM (Static Random Access Memory), DRAM (Dynamic Random Access Memory), flash memory, etc., and temporarily stores data necessary for the control unit 11 to execute arithmetic processing.
- SRAM Static Random Access Memory
- DRAM Dynamic Random Access Memory
- flash memory etc.
- the communication unit 13 is a communication module for performing processing related to communication, and transmits / receives information to / from the outside.
- the auxiliary storage unit 14 is a non-volatile storage area such as a large-capacity memory or a hard disk, and stores the program P1 and other data necessary for the control unit 11 to execute processing.
- the server 1 may be a multi-computer composed of a plurality of computers, or may be a virtual machine virtually constructed by software.
- the server 1 is not limited to the above configuration, and may include, for example, an input unit that accepts operation input, a display unit that displays an image, and the like. Further, the server 1 includes a reading unit for reading a portable storage medium 1a such as a CD (CompactDisk) -ROM, a DVD (DigitalVersatileDisc) -ROM, and reads and executes a program P1 from the portable storage medium 1a. You can do it. Alternatively, the server 1 may read the program P1 from the semiconductor memory 1b.
- a portable storage medium 1a such as a CD (CompactDisk) -ROM, a DVD (DigitalVersatileDisc) -ROM
- FIG. 3 is a block diagram showing a configuration example of the diagnostic imaging apparatus 2.
- the image diagnosis device 2 includes a control unit 21, a main storage unit 22, a communication unit 23, a display unit 24, an input unit 25, an image processing unit 26, and an auxiliary storage unit 27.
- the control unit 21 is an arithmetic processing device such as one or a plurality of CPUs, MPUs, GPUs, etc., and performs various information processing, control processing, etc. by reading and executing the program P2 stored in the auxiliary storage unit 27. ..
- the main storage unit 22 is a temporary storage area such as a RAM, and temporarily stores data necessary for the control unit 21 to execute arithmetic processing.
- the communication unit 23 is a communication module for performing processing related to communication, and transmits / receives information to / from the outside.
- the display unit 24 is a display screen such as a liquid crystal display and displays an image.
- the input unit 25 is an operation interface for a keyboard, a mouse, and the like, and receives operation input from the user.
- the image processing unit 26 is an image processing module that processes signals transmitted and received via the catheter 201 to generate a tomographic image.
- the auxiliary storage unit 27 is a non-volatile storage area such as a hard disk and a large-capacity memory, and stores the program P2 and other data necessary for the control unit 21 to execute processing. Further, the auxiliary storage unit 27 stores the calculation model 51.
- the calculation model 51 is a machine learning model generated by learning predetermined training data, and is a trained model that calculates plaque Baden by inputting a cross-sectional layer image of a blood vessel imaged by the diagnostic imaging apparatus 2. ..
- the calculation model 51 is expected to be used as a program module constituting a part of artificial intelligence software.
- the diagnostic imaging apparatus 2 may include a reading unit that reads a portable storage medium 2a such as a CD-ROM, and may read the program P2 from the portable storage medium 2a and execute the program P2. Alternatively, the diagnostic imaging apparatus 2 may read the program P2 from the semiconductor memory 2b.
- a reading unit that reads a portable storage medium 2a such as a CD-ROM, and may read the program P2 from the portable storage medium 2a and execute the program P2.
- the diagnostic imaging apparatus 2 may read the program P2 from the semiconductor memory 2b.
- FIG. 4 is an explanatory diagram regarding the calculation process of plaque baden.
- FIG. 4 conceptually illustrates how the plaque baden is calculated by inputting each frame of the vascular tomographic image captured by the diagnostic imaging apparatus 2 into the calculation model 51.
- the calculation model 51 is a machine learning model in which predetermined training data has been trained, and is, for example, a CNN (Convolution Neural Network).
- the calculation model 51 may be a model that can calculate plaque Baden from a vascular tomographic image, and may be a model based on other learning algorithms such as a neural network other than CNN, a decision tree, and SVM (Support Vector Machine). good.
- the calculation model 51 has an input layer that accepts an input of an image (tomographic image), an intermediate layer that extracts features of the input image, and an output layer that calculates (outputs) plaque baden.
- the input layer accepts the input of the image and passes the data of the input image to the intermediate layer.
- the intermediate layer includes a convolution layer that folds the data of the input image, extracts the feature amount of the input image, and passes it to the output layer. For the output layer, plaque baden is calculated based on the features extracted in the intermediate layer.
- the process executed by the calculation model 51 is treated as a regression problem, and the plaque baden is predicted by a continuous value.
- this process is treated as a classification problem, for example, "0 to X1", “X1 to X2", “X2 to X3" ... (0 ⁇ X1 ⁇ X2 ⁇ X3 ). May be predicted (classified).
- the plaque baden may be classified into a plurality of stages (for example, a red zone, an orange zone, a yellow zone, and so on) without predicting the plaque baden numerically.
- the server 1 generates a calculation model 51 by using the data to which the correct value of the plaque Baden is given to the vascular tomographic image (cross-sectional layer image) for training as training data (second training data).
- the tomographic image for training is a vascular tomographic image taken with an actual patient as a subject, and the correct answer value is the actual value of plaque baden in the tomographic image of the patient.
- the server 1 inputs the tomographic image for training into the calculation model 51, calculates the plaque baden, and compares it with the correct answer value.
- the server 1 optimizes parameters such as weights between neurons so that the two approximate each other, and generates a calculation model 51.
- the diagnostic imaging apparatus 2 uses the calculation model 51 generated by the server 1 to calculate the plaque baden in the blood vessel of the patient who performs the catheter treatment. Specifically, the diagnostic imaging apparatus 2 sequentially inputs tomographic images of a plurality of frames captured according to the pullback operation of the catheter 201 into the calculation model 51, and calculates the plaque baden in each frame. As a result, the diagnostic imaging apparatus 2 divides the area of the image region corresponding to the plaque and the media (referred to as “plaque region” in the following description) by the area of the EEM region (the region obtained by adding the lumen region to the plaque region). Calculate the area ratio, that is, plaque baden (see the lower right of FIG. 4).
- the diagnostic imaging apparatus 2 presents the plaque baden in each frame calculated above, that is, the plaque baden at each position of the blood vessel along the axial direction to the user, and assists in determining the placement position of the stent.
- FIG. 5 is an explanatory diagram showing a display example of plaque baden.
- the diagnostic imaging apparatus 2 corresponds to a vertical tomographic image 5 in which a predetermined object (first object) representing the size of the plaque baden at each position of a blood vessel is reconstructed into a cross-layer image of a plurality of frames. indicate.
- the diagnostic imaging apparatus 2 displays a gradation bar 6 whose display color is gradually changed according to the size of the plaque baden on the upper side of the vertical tomographic image 5.
- the vertical tomographic image 5 is a vertical cross section that reproduces the state of a blood vessel along the axial direction (traveling direction) of the blood vessel, and is generated from a cross-layer image of a plurality of frames taken according to a pullback motion.
- the diagnostic imaging apparatus 2 displays the vertical tomographic image 5 with the axial direction of the blood vessel as the horizontal direction.
- the diagnostic imaging apparatus 2 displays a gradation bar 6 having a length substantially the same as the length of the vertical tomographic image 5 along the axial direction on the upper side of the vertical tomographic image 5.
- the gradation bar 6 is a bar-shaped object representing the size of the plaque barden, and the display color is changed according to the size of the plaque barden in each position of the corresponding vertical tomographic image 5, that is, the cross-layer image of each frame. It's a bar.
- the diagnostic imaging apparatus 2 changes the display color at each position of the gradation bar 6 according to the size of the plaque barden at each position on the vertical tomographic image 5.
- the region where the plaque baden is large (the region above the threshold value) is shown by hatching.
- the gradation of the display color in the gradation bar 6 is finely set according to the numerical value of the plaque barden, but in FIG. 5, for convenience, the area where the plaque barden is large is shown by the same hatching.
- the object representing the size of the plaque barden is described as the gradation bar 6, but the object is not limited to the gradation bar 6.
- the diagnostic imaging apparatus 2 may display a line graph showing the size of the plaque baden on the upper side of the vertical tomographic image 5.
- the gradation bar 6 is an example of the object, and may be an object that appropriately represents the size of the plaque barden.
- the diagnostic imaging apparatus 2 sets the display color of the area with many plaques with reference to a predetermined threshold value (for example, 50%) set by default. Specifically, the diagnostic imaging apparatus 2 sets the display color of the region where the plaque baden is above the threshold value to a warm color (for example, orange), and sets the region below the threshold value to the default color (for example, white). Then, the diagnostic imaging apparatus 2 changes the display color (for example, increases the brightness) so that the larger the plaque barden is, the more emphasized the region where the plaque barden is above the threshold value, and the gradation display is performed.
- a predetermined threshold value for example, 50%
- the gradation display method for example, it is assumed that the brightness is changed, but the hue, saturation, etc. may be changed, and the specific display method is not particularly limited.
- the diagnostic imaging apparatus 2 hides a region where the reliability of the plaque barden calculation result in the calculation model 51 is low (hereinafter, referred to as “low reliability region”). Is. In FIG. 5, the low reliability region is shown in black.
- the diagnostic imaging apparatus 2 calculates the plaque baden in each frame using the calculation model 51, the diagnostic imaging apparatus 2 acquires the reliability of the calculated value (for example, a probability value of 0 to 1) from the calculation model 51 together with the plaque baden in each frame. do.
- the diagnostic imaging apparatus 2 blanks the area on the corresponding gradation bar 6 as a low reliability area. This makes it possible to present to the user the certainty of the calculation result of plaque baden.
- the diagnostic imaging apparatus 2 displays the gradation bar 6 together with the vertical tomographic image 5, and quantitatively presents the state of the blood vessel at each position. Further, as shown in FIG. 5, the diagnostic imaging apparatus 2 displays the maximum value, the minimum value, and the average value of the plaque baden. As a result, the user (medical worker) can grasp the state of the blood vessel and determine the treatment range to be treated with a treatment device such as a stent.
- FIG. 6 is an explanatory diagram showing another example of the display of plaque baden.
- the warm color is displayed in the area where the plaque barden is equal to or higher than the threshold value, but the area where the plaque baden is less than the threshold value, that is, the area where the plaque is small may also be displayed in gradation.
- the diagnostic imaging apparatus 2 sets the display color of the area where the plaque baden is above the threshold value to a warm color, and sets the display color of the area where the plaque baden is less than the threshold value to a cold color.
- the region where the plaque baden is above the threshold value and the region where the plaque baden is below the threshold value are shown by different hatching.
- the diagnostic imaging apparatus 2 changes the display color (for example, increases the brightness) as the plaque barden becomes smaller, and performs a gradation display.
- the diagnostic imaging apparatus 2 may present a region having a large number of plaques and requiring treatment, and a region having few plaques and a low need for treatment.
- the area for displaying the gradation is set to two, but a threshold value may be further set and the area for displaying the gradation may be divided into three or more areas.
- the user when performing the above-mentioned gradation display, the user may be able to set a threshold value as a display reference.
- the diagnostic imaging apparatus 2 receives a setting input for changing the threshold value from the default value to an arbitrary value from the user, and according to the set threshold value, the area where the plaque barden is equal to or more than the threshold value and the area where the plaque Baden is less than the threshold value. Judge and change the display color.
- the user can visualize the area of interest by himself / herself, and can more preferably support the user.
- FIG. 7 is an explanatory diagram relating to the process of determining the treatment range of blood vessels.
- the treatment range to be treated with a predetermined treatment device is further presented to the user based on the calculation result of plaque baden.
- the diagnostic imaging apparatus 2 determines the positions of both ends of the stent to be placed in the blood vessel.
- vascular treatment such as PCI
- the stent is basically placed so that both ends of the stent are located at the normal site where plaque does not exist, but if that is difficult, both ends of the stent are located at the site where the plaque baden is less than a predetermined value (for example, 50%). Place the stent so that it is located. If that is also difficult, place the stent so that both ends of the stent are located at the site with the least plaque baden.
- a predetermined value for example, 50%
- the plaque baden of all frames is not calculated as described above, but the stenosis is visually identified from the image and the plaque baden of several frames is measured as needed. It is a state. Therefore, in the present embodiment, the range in which both ends of the stent should be located is determined based on the plaque baden calculated by the calculation model 51.
- a stent is mentioned as an example of a therapeutic device, but the therapeutic device is not limited to the stent.
- the diagnostic imaging apparatus 2 may determine the positions of both ends of the expansion range to be expanded by the balloon used for expanding the blood vessel. Further, for example, the diagnostic imaging apparatus 2 may determine the positions of both ends of a cutting range to be cut by a rotablator used for cutting a hard lesion (for example, calcified tissue). As described above, the diagnostic imaging apparatus 2 only needs to be able to determine the positions of both ends of the treatment range to be treated by the treatment device inserted into the blood vessel, and the treatment device is not limited to the stent.
- the diagnostic imaging apparatus 2 divides the vertical tomographic image 5 (longitudinal section) of a blood vessel into a first region in which plaque-baden is above a predetermined threshold and a second region in which plaque-baden is below the threshold.
- the threshold value is, for example, the same value as the threshold value used as the reference for the above-mentioned gradation display, but may be a different value.
- the diagnostic imaging apparatus 2 has a first region L1, L2, L3 ... (L: Lesion) in which the plaque baden is equal to or higher than the threshold value and a second region in which the plaque baden is less than the threshold value, based on the plaque baden in each frame calculated by the calculation model 51.
- the vertical tomographic image 5 is divided into regions H1, H2, H3 ... (H: Healthy).
- FIG. 7 illustrates a vertical tomographic image 5 in which one first region L1 and two second regions HP1 and HD1 (P: Proximal, D: Distal) are present.
- the diagnostic imaging apparatus 2 determines the second regions HP1 and HD1 sandwiching the first region L1 as regions where both ends of the stent should be located.
- the diagnostic imaging apparatus 2 when there is a low reliability region in which the reliability of the plaque baden calculation result is low, a plurality of first regions or second regions straddling the low reliability region are regarded as the same region and grouped. Then, it is suitable. For example, in FIG. 7, there are two low reliability regions in the first region L1. In this case, the diagnostic imaging apparatus 2 determines whether or not the length of each low reliability region in the axial direction of the blood vessel is a predetermined length (for example, 1 mm) or less. When the low reliability region is equal to or less than a predetermined length, the diagnostic imaging apparatus 2 divides a plurality of first regions straddling the low reliability region into the same region. In the example of FIG. 7, since both of the two low-reliability regions have a predetermined length or less, the diagnostic imaging apparatus 2 divides the three regions straddling the low-reliability regions into the same first region L1.
- a predetermined length for example, 1 mm
- the diagnostic imaging apparatus 2 divides the plurality of second regions into the same region when there are a plurality of second regions that straddle a low reliability region having a predetermined length or less. As described above, when the length of the low reliability region in the axial direction is equal to or less than the predetermined length, the diagnostic imaging apparatus 2 sets a plurality of first regions or a plurality of second regions straddling the low reliability regions into the same region. Divide as. This makes it possible to suitably determine the placement position of the stent while ensuring reliability.
- the predetermined length may be variable depending on the user.
- the diagnostic imaging apparatus 2 accepts a setting input for changing a default value of a predetermined length to an arbitrary value (for example, a length of 0.6 mm, 0.8 mm, 1.2 mm, ... In a predetermined step).
- the diagnostic imaging apparatus 2 divides each region by determining whether or not the length is equal to or less than the set predetermined length. This allows the user to more preferably determine the placement position of the stent.
- the diagnostic imaging apparatus 2 may make the threshold value as a reference for region division variable by the user, as in the case of displaying the gradation.
- the diagnostic imaging apparatus 2 may be able to set different threshold values on the proximal side (Proximal) and the distal side (Distal) of the blood vessel. For example, when 50% is set as the threshold value on the proximal side and 60% is set as the threshold value on the distal side, the diagnostic imaging apparatus 2 first determines the position where the plaque baden is 50% or more when viewed from the left side of FIG. The starting point of the region and the position where the plaque baden is less than 60% are divided as the ending point of the first region. As a result, the user can finely adjust the first region and the second region by himself / herself.
- FIG. 8 is an explanatory diagram showing an example of displaying the lesion length.
- the diagnostic imaging apparatus 2 divides the longitudinal tomographic image 5 into the first region L1, L2, L3 ... And the second region H1, H2, H3 ..., And identifies the lesion region L1.
- the diagnostic imaging apparatus 2 measures the length of the lesion region L1 in the axial direction of the blood vessel, that is, the lesion length, and displays it corresponding to the lesion region L1 in the longitudinal tomographic image 5.
- the diagnostic imaging apparatus 2 displays the portion corresponding to the lesion region L1 on the gradation bar 6 with an arrow, and also displays the lesion length. This makes it possible to visualize the length of the lesion site.
- FIG. 9 is an explanatory diagram showing an example of superimposing a stent.
- the diagnostic imaging apparatus 2 further presents to the user a treatment range to be treated by the treatment device, that is, a range in which the stent should be placed.
- the diagnostic imaging apparatus 2 superimposes and displays the stent region 7 (second object) representing the size of the stent (therapeutic device) on the longitudinal tomographic image 5.
- the stent region 7 is an object simulating a stent to be placed in a blood vessel, and is a rectangular object expressing the indwelling range of the stent.
- the length of the stent region 7 may at least match the length of the actual stent in the axial direction, and the length in the radial direction (length in the vertical direction in FIG. 8) may not match.
- the shape of the object is rectangular, but for example, when the treatment device is a balloon, the shape of the object may be a punching bag.
- the diagnostic imaging apparatus 2 may be capable of displaying an object having a shape corresponding to the treatment device.
- both ends of the stent region 7 are located in the second regions HP1 and HD1 in which the stent region 7 having a length equal to or longer than the lesion length measured above is determined as the region where both ends should be located. Is displayed.
- the diagnostic imaging apparatus 2 stores information on each stent product in predetermined length increments (for example, 9 mm, 12 mm, 15 mm, etc.) that can be used for the procedure in the auxiliary storage unit 27. The diagnostic imaging apparatus 2 selects from these stents a stent that exceeds the lesion length and has the smallest length.
- the diagnostic imaging apparatus 2 superimposes and displays the stent region 7 having the selected stent length so that both ends are located in the second regions HP1 and HD1.
- the user may manually select the stent, determine the overlapping position of the stent region 7, and the like.
- the diagnostic imaging apparatus 2 superimposes and displays the stent region 7 so that the center position (midpoint) of the lesion region L1 in the axial direction of the blood vessel coincides with the center position of the stent region 7.
- a downward arrow is displayed at a position corresponding to both ends of the stent, and a bidirectional arrow is displayed at a portion exceeding the first region L1.
- the diagnostic imaging apparatus 2 moves the stent region 7 according to an operation input (for example, mouse operation) from the user, and causes the user to determine the optimum stent placement position.
- the diagnostic imaging apparatus 2 displays a transverse layer image at a position corresponding to both ends of the stent region 7, as shown in FIG. This allows the user to search for the optimal stent placement position while moving the stent region 7.
- the stent region 7 is superimposed and displayed on the vertical tomographic image 5, but the diagnostic imaging apparatus 2 displays the stent region 7 at a location other than the vertical tomographic image 5 (for example, above the gradation bar 6 or inside the gradation bar 6). It may be displayed in. As described above, the diagnostic imaging apparatus 2 may display an object simulating the treatment device, and its display position is not limited to the vertical tomographic image 5.
- FIG. 10 is a flowchart showing the procedure of the generation process of the calculation model 51. Based on FIG. 10, the processing contents when the calculation model 51 is generated by machine learning will be described.
- the control unit 11 of the server 1 acquires training data (second training data) for generating the calculation model 51 (step S11).
- the training data is data in which the correct value of plaque Baden is associated with the cross-sectional layer image of the blood vessel for training.
- the control unit 11 Based on the training data, the control unit 11 generates a calculation model 51 that calculates a plaque baden when a cross-sectional layer image of a blood vessel is input (step S12). Specifically, as described above, the control unit 11 generates a neural network such as a CNN as a calculation model 51. The control unit 11 inputs the vascular tomographic image for training into the calculation model 51, calculates the plaque baden, and compares it with the correct answer value. The control unit 11 optimizes parameters such as weights between neurons so that the calculated plaque Baden approximates the correct answer value, and generates a calculation model 51. The control unit 11 ends a series of processes.
- a neural network such as a CNN
- the control unit 11 inputs the vascular tomographic image for training into the calculation model 51, calculates the plaque baden, and compares it with the correct answer value.
- the control unit 11 optimizes parameters such as weights between neurons so that the calculated plaque Baden approximates the correct answer value, and generates a calculation model 51.
- FIG. 11 is a flowchart showing the procedure of the plaque Baden calculation process.
- the processing contents executed by the diagnostic imaging apparatus 2 will be described with reference to FIG.
- the control unit 21 of the diagnostic imaging apparatus 2 acquires a cross-sectional layer image of a plurality of frames in which a patient's blood vessel is imaged (step S31).
- the control unit 21 inputs the transverse layer image of each frame into the calculation model 51, and calculates the plaque baden in each frame (step S32). Specifically, the control unit 21 acquires the plaque baden in each frame and the reliability of the calculated value from the calculation model 51.
- the control unit 21 specifies the blood vessel position (frame) at which the plaque baden takes the maximum value (step S33). Further, the control unit 21 specifies a low reliability region in which the reliability of the plaque baden is equal to or less than a predetermined value in the axial direction of the blood vessel (step S34). Further, the control unit 21 accepts the threshold setting input (step S35). The control unit 21 may skip step S35 and set a default threshold value.
- the control unit 21 generates (reconstructs) a vertical tomographic image of a blood vessel based on the cross-layer image of a plurality of frames acquired in step S31 (step S36).
- the control unit 21 displays the generated vertical tomographic image and an object (first object) representing the size of the plaque baden at each position of the vertical tomographic image along the axial direction of the blood vessel (step S37).
- the control unit 21 displays a gradation bar whose display color is gradually changed according to the size of the plaque barden. In this case, the control unit 21 displays a gradation bar in which the low reliability region is hidden.
- the control unit 21 Based on the plaque baden at each position (frame) of the blood vessel, the control unit 21 sets the vertical tomographic image (longitudinal section) of the blood vessel as the first region where the plaque baden is above the threshold value and the second region where the plaque baden is below the threshold value. (Step S38). Specifically, the control unit 21 divides the vertical tomographic image into a plurality of regions according to whether or not the plaque barden is equal to or higher than the threshold value, and when the length of the low reliability region is equal to or less than a predetermined length, the control unit 21 divides the vertical tomographic image into a plurality of regions. , A plurality of first regions or second regions straddling the low reliability region are divided as the same region.
- the control unit 21 identifies the first region including the maximum value of plaque baden as a lesion region (step S39). Then, the control unit 21 determines the region where both ends of the treatment range to be treated by the treatment device should be located according to the identified lesion region (step S40). Specifically, the control unit 21 determines the second region located on both sides of the lesion region as the region where both ends of the stent to be placed in the blood vessel should be located.
- the control unit 21 measures the length of the lesion region specified in step S39 in the axial direction, that is, the lesion length, and displays it corresponding to the lesion region of the longitudinal tomographic image (step S41). Further, the control unit 21 displays an object (second object) expressing the size of the treatment device corresponding to the vertical tomographic image (step S42). For example, as described above, the control unit 21 superimposes and displays the stent region expressing the length of the stent in the axial direction of the blood vessel on the vertical tomographic image so that both ends are located in the region determined in step S40. The control unit 21 ends a series of processes.
- the plaque barden at each position of the blood vessel is calculated from the cross-sectional layer image of a plurality of frames obtained by imaging the blood vessel of the patient, and an object such as a gradation bar representing the size of the calculated plaque barden is obtained. Displayed in correspondence with the vertical tomographic image. This makes it possible to suitably support diagnostic imaging related to vascular treatment, such as determination of a stent placement position.
- the certainty of the plaque baden calculation result can be presented to the user.
- the length of the lesion region (lesion length), the indwelling range of the stent (stent region), and the like, it is possible to more preferably support the diagnostic imaging.
- the vertical cross section (longitudinal tomographic image) of the blood vessel is divided into a plurality of regions according to the plaque baden, and both ends of the treatment range of the blood vessel by a therapeutic device such as a stent should be located. To determine. Thereby, it is possible to more preferably support the image diagnosis related to the vascular treatment such as the determination of the placement position of the stent.
- a plurality of regions (first region or second region) straddling the low reliability region are regarded as the same region. To divide. This makes it possible to suitably determine the placement position of the stent and the like while ensuring reliability.
- the first region where the plaque baden has the maximum value is specified as the lesion region.
- the lesion area does not have to be the area where the plaque baden has the maximum value.
- FIG. 12 is an explanatory diagram relating to the specific processing of the lesion region according to the modified example 1.
- the lesion area is specified first, and then the area where both ends of the stent should be located is determined. However, the area where both ends of the stent should be located is determined first, and then the lesion area is specified. May be good.
- the diagnostic imaging apparatus 2 divides the vertical tomographic image 5 into a first region L1, L2, L3 ... And a second region H1, H2, H3 ..., And then sandwiches the first region Lx into two second regions. Search for regions Hy and Hz. When two second regions Hy and Hz sandwiching the first region Lx are searched, the diagnostic imaging apparatus 2 identifies the first region Lx in which the two second regions Hy and Hz are located on both sides as the lesion region Lx. .. Then, the diagnostic imaging apparatus 2 measures the length (lesion length) of the lesion region Lx, and superimposes and displays the stent region 7.
- the diagnostic imaging apparatus 2 may determine the region in which the stent should be placed from the positional relationship between the first region and the second region.
- the determination method described in the first embodiment is an example, and various methods are assumed.
- Embodiment 2 In the first embodiment, a mode in which the calculation model 51 directly calculates the plaque baden has been described. In this embodiment, a mode is described in which a plaque region is identified from a vascular tomographic image (transverse layer image) using another machine learning model and plaque baden is indirectly calculated.
- the contents overlapping with the first embodiment are designated by the same reference numerals and the description thereof will be omitted.
- FIG. 13 is a block diagram showing a configuration example of the diagnostic imaging apparatus 2 according to the second embodiment.
- the auxiliary storage unit 27 of the diagnostic imaging apparatus 2 according to the present embodiment stores the discriminative model 52.
- the discriminative model 52 is a machine learning model generated by learning predetermined training data, like the calculation model 51 according to the first embodiment, and identifies the plaque region in the tomographic image by inputting the vascular tomographic image. It is a trained model to do.
- the discriminative model 52 is expected to be used as a program module constituting a part of artificial intelligence software.
- FIG. 14 is an explanatory diagram relating to the calculation process of plaque baden according to the second embodiment.
- the identification model 52 is used to identify the image region corresponding to the plaque and the media (referred to as “plaque region” in the following description) from the transverse layer image of the blood vessel, and the plaque baden is identified from the area of the identified plaque region.
- plaque region the media
- the state of calculating is conceptually illustrated. An outline of the present embodiment will be described with reference to FIG.
- the discriminative model 52 is a machine learning model in which predetermined training data has been trained, and is, for example, a neural network generated by deep learning.
- the discriminative model 52 will be described as being a semantic segmentation model which is a kind of CNN.
- the semantic segmentation model is a neural network that identifies objects in an image on a pixel-by-pixel basis. And prepare. In the deconvolution layer, which object exists in which position in the image is identified based on the feature amount extracted by the convolution layer, and data obtained by binarizing which object each pixel corresponds to is generated.
- the server 1 generates the discriminative model 52 using the training data (first training data) in which the data indicating the correct plaque region is associated with the tomographic image for training.
- the server 1 inputs a tomographic image for training into the identification model 52 to identify the plaque region, and optimizes parameters such as weights between neurons so that the identified plaque region approximates the correct plaque region.
- the diagnostic imaging apparatus 2 identifies the plaque region using the identification model 52 generated by the server 1.
- the discriminative model 52 is described as a semantic segmentation model in the present embodiment, it may be a model based on a neural network other than the semantic segmentation model, a GAN (Generative Adversarial Network), or another learning algorithm.
- GAN Geneative Adversarial Network
- the server 1 generates a discriminative model 52 that can identify a region other than the plaque region by learning the training data to which the correct answer data of the region other than the plaque region (for example, EEM region, lumen region, etc.) is given. You may. That is, the discriminative model 52 may be able to discriminate at least the plaque region, and may be able to discriminate other image regions as well.
- a discriminative model 52 that can identify a region other than the plaque region by learning the training data to which the correct answer data of the region other than the plaque region (for example, EEM region, lumen region, etc.) is given. You may. That is, the discriminative model 52 may be able to discriminate at least the plaque region, and may be able to discriminate other image regions as well.
- the diagnostic imaging apparatus 2 sequentially inputs a tomographic image of each frame into the identification model 52 and identifies the plaque region in each frame. Then, the diagnostic imaging apparatus 2 calculates the plaque baden based on the identification result of the plaque region. That is, the diagnostic imaging apparatus 2 calculates the area of the plaque region and the area of the entire blood vessel including the plaque region, and calculates the plaque baden.
- the configuration in which the machine learning model directly calculates the plaque baden is not essential, and the plaque baden may be calculated indirectly from the prediction result in the machine learning model.
- FIG. 15 is a flowchart showing the procedure of the generation process of the discriminative model 52. Based on FIG. 15, the processing content when generating the discriminative model 52 by machine learning will be described.
- the control unit 11 of the server 1 acquires training data (first training data) for generating the discriminative model 52 (step S201).
- the training data is data in which data indicating the correct plaque region is associated with the cross-sectional layer image of the blood vessel for training.
- the control unit 11 Based on the training data, the control unit 11 generates an identification model 52 that identifies the plaque region when a cross-layer image of a blood vessel is input (step S202). Specifically, as described above, the control unit 11 generates the CNN related to semantic segmentation as the discriminative model 52. The control unit 11 inputs a cross-sectional layer image for training into the identification model 52, identifies the plaque region, and compares it with the correct plaque region. The control unit 11 optimizes parameters such as weights between neurons so that the two are close to each other, and generates a discriminative model 52. The control unit 11 ends a series of processes.
- FIG. 16 is a flowchart showing a procedure for calculating the plaque baden according to the second embodiment.
- the diagnostic imaging apparatus 2 After acquiring the cross-sectional layer image of a plurality of frames in which the blood vessel of the patient is imaged (step S31), the diagnostic imaging apparatus 2 performs the following processing.
- the control unit 21 of the diagnostic imaging apparatus 2 inputs a transverse layer image of each frame into the identification model 52 and identifies the plaque region in each frame (step S221). Then, the control unit 21 calculates the plaque baden in each frame based on the identification result (step S222).
- the control unit 21 shifts the process to step S31.
- the plaque baden can be calculated by using the discrimination result in the discrimination model 52 that discriminates the plaque region from the cross-layer image of the blood vessel.
- FIG. 17 is an explanatory diagram relating to the generation process of the calculation model 51 according to the modification 2.
- FIG. 17 conceptually illustrates how transfer learning is performed by applying a part of the network structure and parameters of the discriminative model 52 (semantic segmentation model) to the calculation model 51 (CNN).
- the server 1 will be described assuming that the discriminative model 52 has already been generated.
- the server 1 learns by setting parameters such as weights obtained by generating (learning) the discrimination model 52 as initial values in the calculation model 51 having a network structure similar to that of a part of the network structure of the discrimination model 52. conduct.
- the calculation model 51 has a structure in which an input layer similar to the discrimination model 52 and a part of an intermediate layer (convolution layer) corresponding to the decoder portion of the discrimination model 52 are connected to an output layer for plaque Baden calculation. ..
- the server 1 sets the parameters obtained in the training of the discriminative model 52 to the initial values of the parameters of the intermediate layer, and gives training data (first training data) to perform training.
- the server 1 inputs a tomographic image for training, calculates a plaque baden, compares it with the correct answer value, and optimizes the parameters so that the two are close to each other. By performing the transfer learning in this way, the amount of data and the learning time learned by the calculation model 51 can be suppressed.
- FIG. 18 is a flowchart showing the procedure of the generation process of the calculation model 51 according to the modification 2. After acquiring the training data (first training data) for generating the calculation model 51 (step S11), the server 1 executes the following processing. It should be noted that the server 1 will be described assuming that the discriminative model 52 has already been generated.
- the control unit 11 of the server 1 generates the calculation model 51 based on the parameters obtained by generating the discrimination model 52 and the training data (step S301). Specifically, as described above, the control unit 11 performs learning after setting parameters such as weights obtained in the learning of the discriminative model 52 to the initial values of the parameters of the calculation model 51.
- the control unit 11 inputs a tomographic image for training into the calculation model 51, calculates the plaque baden, compares it with the correct answer value, and optimizes the parameters so that the two are close to each other.
- the control unit 11 ends a series of processes.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Vascular Medicine (AREA)
- Quality & Reliability (AREA)
- Optics & Photonics (AREA)
- High Energy & Nuclear Physics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
(実施の形態1)
図1は、画像診断システムの構成例を示す説明図である。本実施の形態では、患者の血管を撮像した横断層像からプラークバーデン(Plaque Burden)を算出し、所定の治療デバイスを用いた血管治療を支援する画像診断システムについて説明する。画像診断装置2は、サーバ1、画像診断装置(情報処理装置)2を含む。各装置は、インターネット等のネットワークNを介して通信接続されている。
制御部11は、一又は複数のCPU(Central Processing Unit)、MPU(Micro-Processing Unit)、GPU(Graphics Processing Unit)等の演算処理装置を有し、補助記憶部14に記憶されたプログラムP1を読み出して実行することにより、種々の情報処理、制御処理等を行う。主記憶部12は、SRAM(Static Random Access Memory)、DRAM(Dynamic Random Access Memory)、フラッシュメモリ等の一時記憶領域であり、制御部11が演算処理を実行するために必要なデータを一時的に記憶する。通信部13は、通信に関する処理を行うための通信モジュールであり、外部と情報の送受信を行う。補助記憶部14は、大容量メモリ、ハードディスク等の不揮発性記憶領域であり、制御部11が処理を実行するために必要なプログラムP1、その他のデータを記憶している。
制御部21は、一又は複数のCPU、MPU、GPU等の演算処理装置であり、補助記憶部27に記憶されたプログラムP2を読み出して実行することにより、種々の情報処理、制御処理等を行う。主記憶部22は、RAM等の一時記憶領域であり、制御部21が演算処理を実行するために必要なデータを一時的に記憶する。通信部23は、通信に関する処理を行うための通信モジュールであり、外部と情報の送受信を行う。表示部24は、液晶ディスプレイ等の表示画面であり、画像を表示する。入力部25は、キーボード、マウス等の操作インターフェイスであり、ユーザから操作入力を受け付ける。画像処理部26は、カテーテル201を介して送受信する信号を処理し、断層像を生成する画像処理モジュールである。
サーバ1の制御部11は、算出モデル51を生成するための訓練データ(第2訓練データ)を取得する(ステップS11)。訓練データは、訓練用の血管の横断層像に対し、プラークバーデンの正解値が対応付けられたデータである。
画像診断装置2の制御部21は、患者の血管を撮像した複数フレームの横断層像を取得する(ステップS31)。制御部21は、各フレームの横断層像を算出モデル51に入力し、各フレームにおけるプラークバーデンを算出する(ステップS32)。具体的には、制御部21は、各フレームにおけるプラークバーデンと、算出値の信頼度とを算出モデル51から取得する。
実施の形態1では、プラークバーデンが最大値を取る第1領域を病変領域として特定した。しかしながら、病変領域はプラークバーデンが最大値を取る領域でなくともよい。
実施の形態1では、算出モデル51がプラークバーデンを直接的に算出する形態について説明した。本実施の形態では、別の機械学習モデルを用いて血管断層像(横断層像)からプラーク領域を識別し、プラークバーデンを間接的に算出する形態について述べる。なお、実施の形態1と重複する内容については同一の符号を付して説明を省略する。
サーバ1の制御部11は、識別モデル52を生成するための訓練データ(第1訓練データ)を取得する(ステップS201)。訓練データは、訓練用の血管の横断層像に対し、正解のプラーク領域を示すデータが対応付けられたデータである。
画像診断装置2の制御部21は、各フレームの横断層像を識別モデル52に入力し、各フレームにおけるプラーク領域を識別する(ステップS221)。そして、制御部21は識別結果に基づき、各フレームにおけるプラークバーデンを算出する(ステップS222)。制御部21は処理をステップS31に移行する。
実施の形態2では、断層像内のプラーク領域を識別可能な識別モデル52について説明した。一方で、識別モデル52の学習で得た重み等のパラメータを算出モデル51に適用し、識別モデル52での学習結果から算出モデル51を生成する転移学習を行ってもよい。
11 制御部
12 主記憶部
13 通信部
14 補助記憶部
P1 プログラム
2 画像診断装置
21 制御部
22 主記憶部
23 通信部
24 表示部
25 入力部
26 画像処理部
27 補助記憶部
P2 プログラム
51 算出モデル
52 識別モデル
5 縦断層像
6 グラデーションバー
7 ステント領域
Claims (15)
- 患者の血管を撮像した複数フレームの横断層像を取得する取得部と、
前記複数フレームの横断層像に基づき、前記血管の軸方向に沿った前記血管の各位置におけるプラークバーデンを算出する算出部と、
前記血管の縦断面を、前記プラークバーデンが閾値以上の第1領域と、前記プラークバーデンが閾値未満の第2領域とに分割する分割部と、
分割した各領域から、所定の治療デバイスによる前記血管の処置範囲の両端が位置すべき領域を決定する決定部と
を備える情報処理装置。 - 前記決定部は、前記血管に留置するステントの両端が位置すべき領域を決定する
請求項1に記載の情報処理装置。 - 前記決定部は、前記第1領域を挟む2つの前記第2領域を前記両端が位置すべき領域に決定する
請求項1に記載の情報処理装置。 - 前記決定部は、
前記縦断面を分割した各領域から、前記プラークバーデンが最大値を取る前記第1領域を特定し、
特定した前記第1領域を挟む2つの前記第2領域を前記両端が位置すべき領域に決定する
請求項3に記載の情報処理装置。 - 前記閾値を設定する設定入力をユーザから受け付ける受付部を備え、
前記分割部は、設定された前記閾値に応じて前記縦断面を分割する
請求項1~4のいずれか1項に記載の情報処理装置。 - 前記算出部は、前記横断層像を入力した場合にプラークバーデンを算出するよう学習済みの算出モデルに、前記複数フレームの横断層像を夫々入力して、各フレームにおける前記プラークバーデンを算出する
請求項1~5のいずれか1項に記載の情報処理装置。 - 前記算出部は、
前記横断層像を入力した場合にプラークに対応する画像領域を識別するよう学習済みの識別モデルに、前記複数フレームの横断層像を夫々入力して、各フレームにおける前記画像領域を識別し、
識別した前記画像領域の面積に基づき、各フレームにおける前記プラークバーデンを算出する
請求項1~5のいずれか1項に記載の情報処理装置。 - 前記算出部は、各フレームにおける前記プラークバーデンと、各フレームにおける該プラークバーデンの算出結果の確からしさを示す信頼度とを算出し、
前記血管の縦断面において、前記信頼度が所定値以下の低信頼度領域を特定する特定部を備え、
前記分割部は、前記軸方向における前記低信頼度領域の長さが所定長以下である場合、該低信頼度領域を跨る複数の前記第1領域又は第2領域を同一領域として分割する
請求項6又は7に記載の情報処理装置。 - 前記所定長を設定する設定入力をユーザから受け付ける第2受付部を備え、
前記分割部は、設定された前記所定長に応じて前記縦断面を分割する
請求項8に記載の情報処理装置。 - 前記縦断面に対応する前記血管の縦断層像と、該縦断面に対応して表示される所定形状のオブジェクトであって、前記両端が位置すべき領域に両端が位置する第2オブジェクトとを表示する表示部を備える
請求項1~9のいずれか1項に記載の情報処理装置。 - 前記表示部はさらに、前記軸方向に沿った前記縦断層像の各位置における前記プラークバーデンの大小を表す第1オブジェクトを表示する
請求項10に記載の情報処理装置。 - 前記第1領域の前記軸方向における長さを計測する測長部を備え、
前記表示部は、前記縦断層像における前記第1領域と対応付けて、計測した前記長さを表示する
請求項10又は11に記載の情報処理装置。 - 患者の血管を撮像した複数フレームの横断層像を取得し、
前記複数フレームの横断層像に基づき、前記血管の軸方向に沿った前記血管の各位置におけるプラークバーデンを算出し、
前記血管の縦断面を、前記プラークバーデンが閾値以上の第1領域と、前記プラークバーデンが閾値未満の第2領域とに分割し、
分割した各領域から、所定の治療デバイスによる前記血管の処置範囲の両端が位置すべき領域を決定する
処理をコンピュータが実行する情報処理方法。 - 患者の血管を撮像した複数フレームの横断層像を取得し、
前記複数フレームの横断層像に基づき、前記血管の軸方向に沿った前記血管の各位置におけるプラークバーデンを算出し、
前記血管の縦断面を、前記プラークバーデンが閾値以上の第1領域と、前記プラークバーデンが閾値未満の第2領域とに分割し、
分割した各領域から、所定の治療デバイスによる前記血管の処置範囲の両端が位置すべき領域を決定する
処理をコンピュータに実行させるプログラム。 - 患者の血管を撮像した横断層像に対し、プラークに対応する画像領域を示すデータが付与された第1訓練データを取得し、
前記第1訓練データに基づき、前記横断層像を入力した場合に前記プラークに対応する画像領域を識別する識別モデルを生成し、
前記横断層像に対し、プラークバーデンの正解値が付与された第2訓練データを取得し、
前記識別モデルを生成して得た学習済みパラメータと、前記第2訓練データとに基づき、前記横断層像を入力した場合に前記プラークバーデンを算出する算出モデルを生成する
処理をコンピュータが実行するモデル生成方法。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2022553930A JPWO2022071181A1 (ja) | 2020-09-29 | 2021-09-27 | |
EP21875483.6A EP4218599A4 (en) | 2020-09-29 | 2021-09-27 | INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, PROGRAM, AND MODEL GENERATION METHOD |
US18/191,974 US20230230252A1 (en) | 2020-09-29 | 2023-03-29 | Information processing device, information processing method, program, and model generation method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2020-163914 | 2020-09-29 | ||
JP2020163914 | 2020-09-29 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US18/191,974 Continuation US20230230252A1 (en) | 2020-09-29 | 2023-03-29 | Information processing device, information processing method, program, and model generation method |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022071181A1 true WO2022071181A1 (ja) | 2022-04-07 |
Family
ID=80951464
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2021/035259 WO2022071181A1 (ja) | 2020-09-29 | 2021-09-27 | 情報処理装置、情報処理方法、プログラム及びモデル生成方法 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20230230252A1 (ja) |
EP (1) | EP4218599A4 (ja) |
JP (1) | JPWO2022071181A1 (ja) |
WO (1) | WO2022071181A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024071251A1 (ja) * | 2022-09-29 | 2024-04-04 | テルモ株式会社 | コンピュータプログラム、情報処理方法、情報処理装置、及び学習モデル |
WO2024071121A1 (ja) * | 2022-09-28 | 2024-04-04 | テルモ株式会社 | コンピュータプログラム、情報処理方法、及び情報処理装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003503141A (ja) * | 1999-07-02 | 2003-01-28 | ザ・クリーブランド・クリニツク・フアンデーシヨン | 有効な輪郭線作成法及びシステムを使用する血管内の超音波分析 |
JP2012176282A (ja) | 2012-06-04 | 2012-09-13 | Toshiba Corp | 医用画像処理装置 |
JP2012200532A (ja) * | 2011-03-28 | 2012-10-22 | Terumo Corp | 画像診断装置及び表示方法 |
WO2017130927A1 (ja) * | 2016-01-26 | 2017-08-03 | テルモ株式会社 | 画像表示装置およびその制御方法 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3520114A1 (en) * | 2016-09-28 | 2019-08-07 | Lightlab Imaging, Inc. | Stent planning systems and methods using vessel representation |
US11819309B2 (en) * | 2019-03-17 | 2023-11-21 | Lightlab Imaging, Inc. | Arterial imaging and assessment systems and methods and related user interface based-workflows |
-
2021
- 2021-09-27 EP EP21875483.6A patent/EP4218599A4/en active Pending
- 2021-09-27 JP JP2022553930A patent/JPWO2022071181A1/ja active Pending
- 2021-09-27 WO PCT/JP2021/035259 patent/WO2022071181A1/ja active Application Filing
-
2023
- 2023-03-29 US US18/191,974 patent/US20230230252A1/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003503141A (ja) * | 1999-07-02 | 2003-01-28 | ザ・クリーブランド・クリニツク・フアンデーシヨン | 有効な輪郭線作成法及びシステムを使用する血管内の超音波分析 |
JP2012200532A (ja) * | 2011-03-28 | 2012-10-22 | Terumo Corp | 画像診断装置及び表示方法 |
JP2012176282A (ja) | 2012-06-04 | 2012-09-13 | Toshiba Corp | 医用画像処理装置 |
WO2017130927A1 (ja) * | 2016-01-26 | 2017-08-03 | テルモ株式会社 | 画像表示装置およびその制御方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP4218599A4 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024071121A1 (ja) * | 2022-09-28 | 2024-04-04 | テルモ株式会社 | コンピュータプログラム、情報処理方法、及び情報処理装置 |
WO2024071251A1 (ja) * | 2022-09-29 | 2024-04-04 | テルモ株式会社 | コンピュータプログラム、情報処理方法、情報処理装置、及び学習モデル |
Also Published As
Publication number | Publication date |
---|---|
US20230230252A1 (en) | 2023-07-20 |
EP4218599A1 (en) | 2023-08-02 |
JPWO2022071181A1 (ja) | 2022-04-07 |
EP4218599A4 (en) | 2024-03-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022071181A1 (ja) | 情報処理装置、情報処理方法、プログラム及びモデル生成方法 | |
WO2022071121A1 (ja) | 情報処理装置、情報処理方法及びプログラム | |
EP3169237B1 (en) | Stenosis assessment | |
CN108962381B (zh) | 用于动脉粥样硬化的个性化评定、长期预测和管理的基于学习的方法 | |
KR20190083185A (ko) | Ivus 영상 분석방법 | |
US20230076868A1 (en) | Systems and methods for utilizing synthetic medical images generated using a neural network | |
CN114126491B (zh) | 血管造影图像中的冠状动脉钙化的评估 | |
CN113241183B (zh) | 治疗方案预测方法及装置 | |
Meier et al. | Automated morphometry of coronary arteries with digital image analysis of intravascular ultrasound | |
CN116630334B (zh) | 用于多分段血管实时自动分割方法、装置、设备及介质 | |
US20220151500A1 (en) | Noninvasive quantitative flow mapping using a virtual catheter volume | |
JP7300392B2 (ja) | 標準化された冠動脈疾患メトリック | |
WO2021193019A1 (ja) | プログラム、情報処理方法、情報処理装置及びモデル生成方法 | |
Van Hamersvelt et al. | Diagnostic performance of on-site coronary CT angiography–derived fractional flow reserve based on patient-specific lumped parameter models | |
CN109310387B (zh) | 估计管腔内设备沿着管腔的管腔内路径 | |
JP7239698B2 (ja) | 血行力学的シミュレーションのための最も関連のあるx線画像の選択 | |
WO2021193024A1 (ja) | プログラム、情報処理方法、情報処理装置及びモデル生成方法 | |
US20230012527A1 (en) | Program, information processing method, information processing apparatus, and model generation method | |
CN116309346A (zh) | 医学图像的检测方法、装置、设备、存储介质及程序产品 | |
WO2022071208A1 (ja) | 情報処理装置、情報処理方法、プログラム、モデル生成方法及び訓練データ生成方法 | |
CN112971979B (zh) | 仿真系统、仿真方法及装置 | |
WO2021193018A1 (ja) | プログラム、情報処理方法、情報処理装置及びモデル生成方法 | |
JP2004133736A (ja) | 医用画像表示方法及びその装置 | |
WO2021193022A1 (ja) | 情報処理装置、情報処理方法及びプログラム | |
US12048492B2 (en) | Estimating the endoluminal path of an endoluminal device along a lumen |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21875483 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2022553930 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2021875483 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2021875483 Country of ref document: EP Effective date: 20230428 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |