WO2023084969A1 - Dispositif d'aide à l'évaluation d'image médicale et procédé de fonctionnement pour dispositif d'aide à l'évaluation d'image médicale - Google Patents

Dispositif d'aide à l'évaluation d'image médicale et procédé de fonctionnement pour dispositif d'aide à l'évaluation d'image médicale Download PDF

Info

Publication number
WO2023084969A1
WO2023084969A1 PCT/JP2022/037775 JP2022037775W WO2023084969A1 WO 2023084969 A1 WO2023084969 A1 WO 2023084969A1 JP 2022037775 W JP2022037775 W JP 2022037775W WO 2023084969 A1 WO2023084969 A1 WO 2023084969A1
Authority
WO
WIPO (PCT)
Prior art keywords
medical image
evaluation index
index value
image
evaluation
Prior art date
Application number
PCT/JP2022/037775
Other languages
English (en)
Japanese (ja)
Inventor
健司 大西
Original Assignee
富士フイルム株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 富士フイルム株式会社 filed Critical 富士フイルム株式会社
Publication of WO2023084969A1 publication Critical patent/WO2023084969A1/fr

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B1/00—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor
    • A61B1/04—Instruments for performing medical examinations of the interior of cavities or tubes of the body by visual or photographical inspection, e.g. endoscopes; Illuminating arrangements therefor combined with photographic or television appliances
    • A61B1/045—Control thereof

Definitions

  • the present invention relates to a medical image evaluation support device and a method of operating the medical image evaluation support device, and more particularly to a technique for assigning evaluation index values to medical images.
  • Scoring is performed by a doctor giving a score (evaluation index value) based on predetermined criteria to the affected area photographed by the endoscope.
  • MES Mayo Endoscopic Subscore
  • UAEIS Ulcerative Colitis Endoscopic Index of Severity
  • the Mayo Score consists of 4 evaluation indices: 1. frequency of defecation, 2. hematochezia, 3. mucosal findings, and 4. doctor's overall evaluation. do.
  • 4 evaluation indices mucosal findings obtained by directly observing the affected area with an endoscope are frequently used alone as MES.
  • MES normal or inactive findings
  • mild redness, decreased vascular transparency, mild fragility
  • moderate marked A score (evaluation index value) of 0 to 3 is given based on findings: redness, loss of vascular transparency, fragility, erosion) (score 2), severe (spontaneous bleeding, ulceration) (score 3).
  • Patent Document 1 has been proposed as a medical image diagnostic apparatus that can easily perform diagnosis based on the index value of the region of interest.
  • the medical image diagnostic apparatus described in Patent Document 1 includes an index value calculation unit and a display characteristics determination unit.
  • the index value calculation unit calculates the index value of the region of interest of the subject based on the data collected by scanning the subject.
  • the display characteristic determination unit determines display characteristics of at least one of the region of interest and numerical information of the region of interest based on the index value. Then, based on the determined display characteristics, at least one of the region of interest and numerical information of the region of interest is displayed. For example, based on the numerical information of the region of interest, the frame line of the region of interest is displayed in a “red” or “blue” display color.
  • the medical image diagnostic apparatus further includes an index value setting unit that sets a threshold value of the index value, and the display characteristics determination unit combines the index value calculated by the index value calculation unit with the threshold value set by the index value setting unit.
  • the display characteristics are determined based on
  • ⁇ value (evaluation agreement rate - probability of coincidence) / (1 - probability of coincidence)
  • evaluation index values are used for prescriptions such as medications that patients receive, if there is variation in the evaluation index values, the prescriptions that patients receive will vary, making it impossible to receive consistent treatment based on unified standards. There was a problem that it leads to deterioration of the patient's Quality of Life (QOL).
  • QOL Quality of Life
  • Patent Document 1 describes that an operator sets a threshold set by an index value setting unit by inputting, but the determination of the display characteristics corresponding to the evaluation index value is performed by a medical image diagnostic apparatus (display characteristics determination Department), and there is a problem that the doctor who is the operator cannot decide.
  • the present invention has been made in view of such circumstances, and is a medical image evaluation support apparatus capable of reducing variations in evaluation index values assigned to a plurality of medical images and efficiently assigning evaluation index values. And it aims at providing the operating method of a medical image evaluation support apparatus.
  • the invention according to a first aspect is a medical image evaluation support device comprising a processor, wherein the processor acquires a first image group consisting of a plurality of medical images and configures the first image group. displaying at least one first medical image on a display device; receiving a first evaluation index value for the first medical image from a user; applying the first evaluation index value to the first medical image; and calculating a second evaluation index value for a second medical image different from the first medical image in the first image group, and assigning the second evaluation index value to the second medical image.
  • the received first 1 evaluation index value is given to the first medical image. Further, based on the first evaluation index value for the first medical image, a second evaluation index value for the second medical image (image different from the first medical image in the first image group) is calculated, and the calculated second evaluation index A value is applied to the second medical image.
  • the processor associates a first evaluation index value with a first medical image of the first image group and displays the first evaluation index value on the display device, and displays the first image group on the display device. It is preferable to associate the second evaluation index value with the second medical image among them and display it on the display device.
  • the first evaluation index value and the second evaluation index value are evaluation index values representing the degree of disease.
  • the processor performs the processing for the second medical image based on the first evaluation index value, the feature amount of the first medical image, and the feature amount of the second medical image It is preferable to calculate the second evaluation index value. Since the user gives the first evaluation index value to the first medical image based on the feature amount of the first medical image, the relationship between the feature amount of the first medical image and the first evaluation index value is the same as , a second evaluation index value can be calculated for the second medical image based on the feature amount of the second medical image.
  • the feature quantity is at least redness of mucous membrane, bleeding, disappearance of vascular fluoroscopic image, erosion, fragility, ulcer, granulation, edema, and intestinal dilatation. It is a value calculated based on one feature.
  • these feature amounts serve as criteria for endoscopic evaluation indices of inflammatory bowel disease such as ulcerative colitis.
  • a medical image evaluation support device comprising a machine learning model obtained by learning using learning data consisting of a pair of a first medical image and a first evaluation index value corresponding to the first medical image
  • the feature amount is preferably an output value of a machine learning model obtained from a machine learning model input with each medical image of the first image group, or an intermediate feature amount.
  • the first image group includes two or more types of endoscopic images captured under illumination light with different light source types, and the feature amount is at least endoscopic It is preferable to include the light source information of the illumination light for the mirror image. This is because the endoscopic image or how the disease appears in the endoscopic image differs depending on the type of light source. Therefore, the second evaluation index value can be assigned only to the second medical image captured with the same type of light source as the type of light source used when capturing the first medical image.
  • the feature quantity preferably includes at least image processing information for each medical image in the first image group. This is because the appearance of an image differs depending on texture enhancement such as TXI (Texture and color enhancement Imaging) and other types of image processing.
  • texture enhancement such as TXI (Texture and color enhancement Imaging) and other types of image processing.
  • the processor receives the first evaluation index values for the two or more first medical images in the first image group, and assigns the first evaluation index values to the two or more first medical images. It is preferable to calculate the second evaluation index value for the second medical image based on a regression model obtained from the first evaluation index value and the feature amount of the two or more first medical images.
  • a medical image evaluation support device comprising an evaluation index value prediction model for predicting a third evaluation index value from a feature value associated with each medical image in the first image group
  • the processor Based on the third evaluation index value obtained by inputting the feature value associated with each medical image in the first image group into the evaluation index value prediction model and the first evaluation index value assigned to the first medical image to calculate the second evaluation index value for the second medical image. For example, a shift amount or magnification is calculated from the difference between the first evaluation index value input by the user and the third evaluation index value predicted from the feature amount by the evaluation index value prediction model, and the third evaluation index value is adjusted.
  • a second evaluation index value can be calculated.
  • the processor sets the third evaluation index value to the second It is preferable to use the second evaluation index value for the medical image. This is because if the reliability of the third evaluation index value is high, there is no need to adjust the third evaluation index value to calculate the second evaluation index value.
  • the processor displays the first image group and the second image group different from the first image group on the display device so that they can be compared.
  • the second image group may be a group of images of the same patient taken at different dates, a group of images of different patients, or the like.
  • the processor acquires part information indicating the imaging parts of the plurality of medical images forming the first image group, and the received part information are preferably displayed on the display device in association with each other.
  • Each medical image can be displayed in association with the part name (text information) of the imaging part of the medical image, or each medical image can be associated with a schematic diagram of an organ and displayed.
  • the processor estimates the imaging region where each medical image was captured based on each medical image constituting the first image group, It is preferable to display on a display device in association with each constituent medical image and part information indicating an estimated imaging part.
  • the processor receives an input of an imaging region of a medical image to be displayed on the display device, and selects a medical image corresponding to the received imaging region from the first image group.
  • the selected medical image and region information indicating the imaging region are associated with each other and displayed on the display device.
  • the processor receives an initial evaluation index value corresponding to each medical image included in the first image group, and the first evaluation given to the first image group
  • the index value, the second evaluation index value, and the initial evaluation index value are associated with each other and displayed on the display device.
  • the processor selects a medical image having an initial evaluation index value different from the first evaluation index value and the second evaluation index value, and It is preferable to display the second evaluation index value and the initial evaluation index value in association with each other on the display device.
  • An invention according to an eighteenth aspect is a method of operating a medical image evaluation support apparatus comprising a processor and a display device, wherein the processor acquires a first image group consisting of a plurality of medical images and constructs the first image group. displaying at least one first medical image on a display device, a processor accepting a first evaluation index value for the first medical image from a user, applying the first evaluation index value to the first medical image, the processor: Medical image evaluation, which calculates a second evaluation index value for a second medical image different from the first medical image in the first image group based on the first evaluation index value, and assigns the second evaluation index value to the second medical image A method of operating an assist device.
  • the present invention while prioritizing the evaluation index value assigned by the user, it is possible to reduce variations in evaluation index values assigned to a plurality of medical images, and to efficiently assign evaluation index values.
  • FIG. 1 is a block diagram showing an embodiment of a hardware configuration of a medical image evaluation support device according to the present invention.
  • FIG. 2 is a functional block diagram showing the first embodiment of the medical image evaluation support device according to the present invention.
  • FIG. 3 is a diagram for explaining the outline of the present invention.
  • FIG. 4 is a functional block diagram showing a second embodiment of the medical image evaluation support device according to the present invention.
  • FIG. 5 is a chart showing the UCEIS representing an endoscopic index for ulcerative colitis.
  • FIG. 6 is a diagram showing a first display example of a display device that displays a first image group to which evaluation index values are assigned.
  • FIG. 7 is a functional block diagram showing a third embodiment of a medical image evaluation support device according to the present invention.
  • FIG. 1 is a block diagram showing an embodiment of a hardware configuration of a medical image evaluation support device according to the present invention.
  • FIG. 2 is a functional block diagram showing the first embodiment of the medical image evaluation support device according to the present invention
  • FIG. 8 is a functional block diagram showing a fourth embodiment of a medical image evaluation support device according to the present invention.
  • FIG. 9 is a diagram showing a second display example of the display device displaying the first image group to which the evaluation index values are assigned.
  • FIG. 10 is a diagram showing a third display example of the display device displaying the first image group to which the evaluation index values are assigned.
  • FIG. 11 is a functional block diagram showing a fifth embodiment of the medical image evaluation support device according to the present invention.
  • FIG. 12 is a diagram showing a first display example of a display device that displays a first image group and a second image group to which evaluation index values are assigned.
  • FIG. 13 is a diagram showing a second display example of a display device that displays a first image group and a second image group to which evaluation index values are assigned.
  • FIG. 14 is a diagram showing a third display example of the display device displaying the first image group to which the initial evaluation index values are assigned and the first image group to which the evaluation index values are assigned by the medical image evaluation support device.
  • FIG. 15 is a diagram showing a fourth display example of the display device displaying the first image group to which the initial evaluation index value is assigned and the first image group to which the evaluation index value is assigned by the medical image evaluation support device.
  • FIG. 16 is a diagram showing a first display example of a display device that displays a first image group to which evaluation index values are assigned and part information.
  • FIG. 17 is a diagram showing a second display example of the display device that displays the first image group to which the evaluation index values are assigned and the part information.
  • FIG. 18 is a flow chart showing an embodiment of a method for operating a medical image evaluation support apparatus according to the present invention.
  • FIG. 1 is a block diagram showing an embodiment of a hardware configuration of a medical image evaluation support device according to the present invention.
  • a medical image evaluation support apparatus 1 shown in FIG. 1 is an apparatus for assigning evaluation index values (scores) representing disease activity states to medical images. It includes a device 16, an input/output interface 18, an operation unit 20, and the like.
  • the display device 16 may be included in the medical image evaluation support device 1 or may be provided outside the medical image evaluation support device 1 .
  • the processor 10 is composed of a CPU (Central Processing Unit) and the like, and performs integrated control of each part of the medical image evaluation support apparatus 1, and calculates and calculates evaluation index values for medical images by executing a medical image evaluation support program. It functions as a processing unit (an image acquisition unit, a feature amount acquisition unit, a second evaluation index value calculation unit, etc., which will be described later) that performs various types of processing for assigning the obtained evaluation index value to the medical image.
  • a processing unit an image acquisition unit, a feature amount acquisition unit, a second evaluation index value calculation unit, etc., which will be described later
  • the memory 12 includes flash memory, ROM (Read-only Memory), RAM (Random Access Memory), hard disk device, and the like.
  • the flash memory, ROM, or hard disk device is a non-volatile memory that stores an operating system, various programs including a medical image evaluation support program, and the like.
  • the RAM functions as a work area for processing by the processor 10 and temporarily stores programs and the like stored in a flash memory or the like. Note that the processor 10 may incorporate part of the memory 12 (RAM).
  • the memory 12 functions as an image storage unit that stores a plurality of medical images, and stores and manages medical images captured for each patient and examination by the modality of capturing medical images.
  • an endoscopic image captured by an endoscope is used as a medical image to which an evaluation index value is assigned. It's okay.
  • the processor 10 acquires a first image group consisting of a plurality of medical images from the modality or from the memory 12 functioning as an image storage unit, while using the RAM as a work area, according to the medical image evaluation support program.
  • the first image group is, for example, a plurality of medical images taken in one examination, and is chronologically ordered still images or moving images arranged in photographing order. Not limited to things.
  • the first image group in this example is a plurality of endoscopic images captured during one endoscopy of a certain patient based on a still image capturing instruction from a user (doctor).
  • the processor 10 displays on the display device 16 at least one medical image (first medical image) that constitutes the first image group. This is because the user observes the first medical image displayed on the display device 16 and receives the first evaluation index value for the first medical image from the user.
  • the first evaluation index value is an evaluation index value representing the extent (severity) of a disease.
  • the processor 10 Upon receiving the first evaluation index value for the first medical image, the processor 10 assigns the received first evaluation index value to the first medical image. For example, if the first medical image is saved in a DICOM (Digital Imaging and Communications in Medicine) file format, the first evaluation index value can be given as tag information of the DICOM file.
  • DICOM Digital Imaging and Communications in Medicine
  • the processor 10 calculates a second evaluation index value for a medical image (second medical image) different from the first medical image in the first image group based on the first evaluation index value, and calculates the calculated second evaluation index value to the second medical image.
  • the second evaluation index value is an evaluation index value representing the degree of disease, like the first evaluation index value.
  • the display device 16 displays at least one medical image that constitutes the first image group, and also displays attached information attached to the medical image (patient name, examination date, light source type, evaluation index value, etc.). The user can confirm the medical image and attached information displayed on the display device 16 .
  • the display device 16 is also used as part of a GUI (Graphical User Interface) when receiving various types of information from the operation unit 20 .
  • the input/output interface 18 includes a connection section that can be connected to an external device, a communication section that can be connected to a network, and the like.
  • USB Universal Serial Bus
  • HDMI High-Definition Multimedia Interface
  • HDMI High-Definition Multimedia Interface
  • the processor 10 can acquire medical images from modalities or output them externally via an input/output interface 18 .
  • the operation unit 20 includes a pointing device such as a mouse, a keyboard, etc., and functions as a part of the GUI that receives input of a first evaluation index value for the first medical image from the user and various information and instructions from the user.
  • a pointing device such as a mouse, a keyboard, etc.
  • FIG. 2 is a functional block diagram showing the first embodiment of the medical image evaluation support device according to the present invention.
  • the medical image evaluation support apparatus 1-1 shown in FIG. 2 includes an image acquisition unit 50, an image storage unit 54, a display device 16, a first evaluation index value input unit 56, and a second evaluation index value calculation unit 58. .
  • the image acquisition unit 50 and the second evaluation index value calculation unit 58 function when the processor 10 of the medical image evaluation support apparatus 1 having the hardware configuration shown in FIG. 1 executes a medical image evaluation support program.
  • the memory 12 and the operation unit 20 shown in FIG. 1 correspond to the image storage unit 54 and the first evaluation index value input unit 56, respectively.
  • an image acquisition unit 50 uses an endoscope (not shown) to capture a plurality of endoscopic images based on a still image capturing instruction from a user during one endoscopy of a certain patient.
  • a mirror image is acquired as a first group of images.
  • the first image group acquired by the image acquisition section 50 is stored in the image storage section 54 .
  • the processor 10 displays on the display device 16 at least one first medical image forming the first image group stored in the image storage unit 54 .
  • the user observes the first medical image displayed on the display device 16 (the first medical image specified by the user when a plurality of medical images are displayed), and the first evaluation index value input unit 56 1 Enter the first evaluation index value for the medical image. That is, the user inputs the metric value of an endoscopic metric, such as MES, UCEIS, etc., that represents disease activity.
  • the processor 10 receives a first evaluation index value for the first medical image via the first evaluation index value input unit 56, and assigns the received first evaluation index value to the first medical image.
  • FIG. 3 is a diagram for explaining the outline of the present invention.
  • 3-1 in FIG. 3 is a diagram showing a first image group photographed in time series, in which one first medical image designated by the user in the first image group is given a first evaluation by the user. It shows the case of assigning an index value.
  • the user assigns "2" as the first evaluation index value to one first medical image specified by the user.
  • the second evaluation index value calculation unit 58 shown in FIG. 2 receives the first image group from the image storage unit 54 and assigns the first image group from the first evaluation index value input unit 56 to the first medical image in the first image group.
  • the calculated first evaluation index value is input, and a second evaluation index value for a second medical image different from the first medical image in the first image group is calculated based on the first evaluation index value.
  • the second evaluation index value calculation unit 58 calculates the second evaluation index value for the second medical image by reflecting the relationship between the first medical image given to the first medical image by the user and the first evaluation index value. calculate. Calculation of the second evaluation index value by the second evaluation index value calculation unit 58 is performed by, for example, using an image processing method to calculate image feature amounts such as brightness, color, and frequency characteristics of each image of the first medical image and the second medical image.
  • the second evaluation index may be calculated based on the image feature amount of the first medical image extracted and given the first evaluation index value and the image feature amount of the second medical image.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58 are assigned to the corresponding second medical images.
  • the second medical image to which the second evaluation index value has been assigned is stored in the image storage unit 54 in the same manner as the first medical image to which the first evaluation index value has been assigned.
  • each medical image (second medical image) other than the first medical image to which the first evaluation index value is assigned by the user in the first image group is calculated as the second evaluation index value.
  • a state in which the second evaluation index value calculated by the unit 58 is given is shown.
  • the processor 10 associates the first evaluation index value given by the user and the second evaluation index value given by the medical image evaluation support device 1-1 with each medical image. and display it on the display device 16 to notify the user.
  • the first medical image in the first image group By calculating and assigning the second evaluation index value to the second medical image different from the second medical image, the evaluation index value for each medical image in the first image group can be assigned based on a unified standard, and the result is displayed on the display device 16 can be confirmed by In addition, it is possible to reduce variations in the evaluation index values assigned to the medical images in the first image group.
  • evaluation index values can be assigned based on a unified standard to a group of images of the same patient captured in the past (second image group) and to a group of images captured by different users (second image group). , can be compared with the evaluation index values assigned to the first image group.
  • FIG. 4 is a functional block diagram showing a second embodiment of the medical image evaluation support device according to the present invention.
  • the medical image evaluation support apparatus 1-2 of the second embodiment shown in FIG. 58-1 which is different from the medical image evaluation support apparatus 1-1 of the first embodiment.
  • the image acquisition unit 50 acquires each medical image 4-1 constituting the first image group
  • the feature amount acquisition unit 52-1 acquires the feature amount for each medical image 4-1. is.
  • This feature quantity acquisition unit 52-1 acquires the feature quantity 4-2 input by the user using the operation unit 20 corresponding to each medical image 4-1.
  • FIG. 5 is a diagram showing UCEIS, which represents an endoscopic evaluation index for ulcerative colitis.
  • the feature amount acquisition unit 52-1 acquires the UCEIS evaluation item (defect fluoroscopy image score , bleeding score, erosion and ulcer score), e.g. do.
  • the feature amount 4-2 is not limited to the defect fluoroscopic image score, bleeding score, and erosion and ulcer score, redness of mucous membranes, bleeding, disappearance of vascular fluoroscopic image, erosion, fragility, ulcer, granulation, It may be a value calculated based on at least one characteristic of edema and intestinal dilatation. Using such a value makes it easier for the doctor to understand the criterion for the index.
  • the feature quantity 4-2 acquired by the feature quantity acquisition unit 52-1 is recorded as tag information of the DICOM file in which each medical image 4-1 is stored, or is associated with each medical image 4-1 and is stored as an image in text format. It can be stored in the storage unit 54 .
  • the second evaluation index value calculation unit 58-1 corresponds to the first evaluation index value input by the first evaluation index value input unit 56 and each medical image 4-1 acquired by the feature amount acquisition unit 52-1.
  • a second evaluation index value for the second medical image is calculated based on the feature amount 4-2 (the feature amount of the first medical image and the feature amount of the second medical image).
  • the user designates a certain medical image (third medical image in chronological order) among the medical images 4-1 constituting the first image group as the first medical image, and the first medical image A case where the first evaluation index value (“2”) is input via the first evaluation index value input unit 56 for is described.
  • UCEIS score of each medical image 4-1 is (1,1,0) (1,1,2) (2,2,0) (3,2,2)...(2,2,0)
  • the second evaluation index value of the second medical image with the same total score of "4" is set to "2"
  • the second evaluation index value of the second medical image with the total score of "2” is set to "1”
  • the second evaluation index value is "3”.
  • the feature quantity corresponding to each medical image constituting the first image group the total score of the UCEIS defect fluoroscopic image score, bleeding score, and erosion and ulcer score is used, and the second evaluation for the second medical image
  • the feature amount corresponding to the medical image is not limited to the UCEIS defect fluoroscopic image score, bleeding score, and erosion and ulcer score, as well as scores of other evaluation criteria such as MES, ,
  • the feature amount corresponding to the medical image may be calculated by a classical image processing method. Learning may be performed, and the output of the discriminator may be used as the feature quantity.
  • image processing information such as edge enhancement and texture enhancement for medical images performed by an endoscope apparatus may be used as feature amounts.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58-1 are assigned to the corresponding second medical images. Further, the processor 10 causes the display device 16 to display the first evaluation index value given by the user and the second evaluation index value given by the medical image evaluation support device 1-2 in association with each medical image, Notify the user.
  • FIG. 6 is a diagram showing a first display example of a display device that displays a first image group to which evaluation index values are assigned.
  • the medical images constituting the first image group are displayed on the display device 16 shown in 6-1 of FIG. , and the user assigns the first evaluation index value “2” to the first medical image.
  • the second evaluation index value calculation unit 58-1 calculates a second A second evaluation index value is calculated for the medical image.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58-1 are assigned to the corresponding second medical images.
  • the third first medical image and the first evaluation index value "2" are displayed in order, and the first, second, fourth, and last second medical images and the second evaluation index value ( “ 1 ", " 2 ", “ 3 “, “ 2 ”) are displayed in association with each other.
  • the second evaluation index values calculated and assigned by the second evaluation index value calculation unit 58-1 are " 1 ", " 2 ", “ 3 ", and "
  • the second evaluation index value calculation unit 58-1 is " 1 ", " 2 ", " 3 ", and "
  • FIG. 7 is a functional block diagram showing a third embodiment of a medical image evaluation support device according to the present invention.
  • a machine learning model 60 is added. is different from the medical image evaluation support apparatus 1-1 of the first embodiment.
  • the image acquisition unit 50 acquires each medical image 4-1 that constitutes the first image group. to get
  • the machine learning model 60 when the machine learning model 60 inputs a medical image, it outputs the feature amount of the medical image.
  • the machine learning model 60 of this example can apply a neural network, SVM, or the like, and for example, learns with learning data consisting of a first medical image and a first evaluation index value pair corresponding to the first medical image. It can be a trained model obtained.
  • the machine learning model 60 which is a trained model, extracts the feature amount of the medical image, and uses the output value of the fully connected layer as the feature amount corresponding to the evaluation index value of the input medical image. Output.
  • the output value of the machine learning model 60 can be recorded as tag information of the DICOM file in which each medical image 4-1 is stored, or can be stored in the image storage unit 54 in text format in association with each medical image 4-1. can.
  • the second evaluation index value calculation unit 58-2 uses the first evaluation index value input by the first evaluation index value input unit 56 and the feature amount corresponding to each medical image 4-1 acquired by the machine learning model 60.
  • a second evaluation index value for the second medical image is calculated based on the output value of a certain machine learning model 60 .
  • the second evaluation index value calculation unit 58-2 calculates the first evaluation index value input by the first evaluation index value input unit 56 and the machine learning model 60 when the first medical image is input to the machine learning model 60. If the output value matches, the output value of the machine learning model 60 when the second medical image is input to the machine learning model 60 is applied as the second evaluation index value for the second medical image. , when the first evaluation index value input by the first evaluation index value input unit 56 does not match the output value of the machine learning model 60 when the first medical image is input to the machine learning model 60 calculates the second evaluation index value for the second medical image by correcting the output value of the machine learning model 60 based on the difference or ratio between the two.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58-2 are assigned to the corresponding second medical images. Further, the processor 10 causes the display device 16 to display the first evaluation index value given by the user and the second evaluation index value given by the medical image evaluation support device 1-3 in association with each medical image, Notify the user.
  • the machine learning model 60 may output, for example, an intermediate feature amount of the medical image calculated in the intermediate layer of the neural network as the feature amount of the medical image instead of the output value of the fully connected layer.
  • the second evaluation index value calculation unit 58-2 calculates the machine learning model when the first evaluation index value and the first medical image input by the first evaluation index value input unit 56 are input to the machine learning model 60.
  • a second evaluation index value for the second medical image is calculated based on the intermediate feature amount of 60 and the intermediate feature amount of the machine learning model 60 when the second medical image is input to the machine learning model 60 .
  • FIG. 8 is a functional block diagram showing a fourth embodiment of a medical image evaluation support device according to the present invention.
  • the medical image in this example is an endoscopic image, as described above.
  • the endoscopic image includes images captured by a light source called special light.
  • special light There is a special light image that
  • Normal light is white light (WL: White Light) in which light in all wavelength bands of visible light is almost evenly mixed, and normal light (WLI: White Light Imaging) images are used for normal light observation.
  • WL White Light
  • WLI White Light Imaging
  • Special light is light in one specific wavelength band, or light in various wavelength bands that is a combination of light in a plurality of specific wavelength bands, depending on the purpose of observation.
  • Special light images include, for example, LCI (Linked Color Imaging) images and BLI (Blue Light Imaging or Blue LASER Imaging) images.
  • Endoscope light source devices include, for example, V-LED (Violet Light Emitting Diode), B-LED (Blue Light Emitting Diode), G-LED (Green Light Emitting Diode), and R-LED (Red Light Emitting Diode) and emits normal light or special light by adjusting the emission intensity of the four color LEDs.
  • V-LED Volt Light Emitting Diode
  • B-LED Blue Light Emitting Diode
  • G-LED Green Light Emitting Diode
  • R-LED Red Light Emitting Diode
  • the special light for LCI has a high ratio of V light, which has a high absorption rate in superficial blood vessels, compared to normal light, and is suitable for capturing minute changes in color tone.
  • This is an image that has undergone color enhancement processing such that reddish colors are made redder and whitish colors are made whiter, centering on colors near the mucous membrane.
  • the special light for BLI is light with a high ratio of V light, which has a high absorption rate in superficial blood vessels, and a low ratio of G light, which has a high absorption rate in middle-layer blood vessels.
  • V light which has a high absorption rate in superficial blood vessels
  • G light which has a high absorption rate in middle-layer blood vessels.
  • WLI images, LCI images, and BLI images can be captured by the user switching the observation mode of the endoscope.
  • the image acquisition unit 50 acquires each medical image 4-1 that constitutes the first image group.
  • endoscopic images WLI images and LCI images in this example are mixed.
  • the feature amount acquisition unit 52-2 like the feature amount acquisition unit 52-1 of the medical image evaluation support device 1-2 of the second embodiment shown in FIG.
  • UCEIS evaluation items defective fluoroscopic image score, bleeding score, erosion and ulcer score
  • (1,1,0) (1,1,2) (2,2,0) (3 , 2, 2) for example, (1,1,0) (1,1,2) (2,2,0) (3 , 2, 2) .
  • the light source information of illumination light is, for example, information indicating normal light (WLI), special light (LCI, BLI), and the like.
  • the light source information of the illumination light can be obtained by image processing from the characteristics according to the type of light source of each image of the WLI image, the LCI image, and the BLI image, or by being discriminated by a learning model.
  • the feature quantity acquisition unit 52-2 obtains the feature quantity 4-2 ((1,1,0) (1,1,2) (2, 2,0) (3,2,2)...(2,2,0)) and the light source information 8 (WLI, WLI, LCI, LCI... LCI) of the illumination light of each medical image 4-1 is acquired.
  • the feature quantity 4-2 acquired by the feature quantity acquisition unit 52-2 and the light source information 8 are recorded as tag information of the DICOM file in which each medical image 4-1 is stored, or are associated with each medical image 4-1. can be stored in the image storage unit 54 in text form.
  • the feature amount acquisition unit 52-2 does not need to acquire the light source information of the illumination light.
  • the second evaluation index value calculation unit 58-3 calculates the first evaluation index value input by the first evaluation index value input unit 56, the feature amount 4-2 acquired by the feature amount acquisition unit 52-2, and the light source information 8.
  • a second evaluation index value for the second medical image is calculated based on.
  • the second evaluation index value calculation unit 58-3 uses the type of light source such as WLI, LCI, and BLI in the endoscope apparatus as a feature amount to distinguish how the disease looks different depending on the light source, and calculates the second evaluation index. value can be calculated.
  • image processing information such as edge enhancement and texture enhancement performed by the endoscope apparatus may be used as the feature amount. By using the image processing information, it is possible to select a medical image to be processed for the second evaluation index value according to the type of image processing.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58-3 are assigned to the corresponding second medical images. Further, the processor 10 causes the display device 16 to display the first evaluation index value given by the user and the second evaluation index value given by the medical image evaluation support device 1-4 in association with each medical image, Notify the user.
  • FIG. 9 is a diagram showing a second display example of the display device displaying the first image group to which the evaluation index values are assigned.
  • the medical images constituting the first image group are displayed on the display device 16 shown in 9-1 of FIG. is specified by the user, and the user assigns the first evaluation index value (“2”) to the first medical image. Also, light source information (WLI, LCI) is displayed in association with each medical image.
  • WLI, LCI light source information
  • the second evaluation index value calculation unit 58-3 calculates the light source information of the first medical image to which the first evaluation index value has been assigned by the user, from the remaining second medical images that have not been specified among the medical images 4-1. It can be used to select a medical image to process the second evaluation index value of the image.
  • the second evaluation index value calculation unit 58-3 calculates the remaining second evaluation index value that has not been specified.
  • the medical images only the medical images whose light source information is LCI are processed to be given a second evaluation index value, and the first evaluation index value of the first medical image of LCI and the feature amount acquisition unit 52-2 acquire.
  • the second LCI medical image is 2 Calculate the evaluation index value.
  • the second evaluation index values calculated by the second evaluation index value calculation unit 58-3 are given to the corresponding second medical images of the LCI. , displays the first medical image and the first evaluation index value (“2”) of the third LCI in chronological order, displays the second medical image and the first evaluation index value (“2”) of the fourth and last LCI in chronological order, 2 evaluation index values (“3”, “2”) are displayed in association with each other.
  • the second evaluation index value is not assigned to the first and second WLI second medical images in chronological order.
  • the second evaluation index value is also assigned to the first and second WLI second medical images in chronological order.
  • the user selects two or more of the medical images forming the first image group displayed on the display device 16. It is possible to specify a medical image and input first evaluation index values for two or more medical images (first medical images) via the first evaluation index value input unit 56, and the processor 10 can input two or more first evaluation index values. A first evaluation index value for each medical image can be received, and the received first evaluation index value can be applied to the corresponding first medical image.
  • FIG. 10 is a diagram showing a third display example of the display device displaying the first image group to which the evaluation index values are assigned.
  • the display device 16 shown in 10-1 in FIG. 10 displays the medical images that make up the first image group, and the first and third medical images in chronological order among the medical images are the first medical images. It shows how the user assigns the first evaluation index values (“1” and “2”) to the first and third first medical images specified by the user as images.
  • the second evaluation index value calculators 58, 58-1, 58-2, and 58-3 of the medical image evaluation support apparatuses 1-1 to 1-4 of the first to fourth embodiments have two or more The regression model obtained from the first evaluation index values (“1”, “2”) assigned to the (two in this example) first medical images and the feature values of the two first medical images, and the unspecified It is preferable to calculate the second evaluation index value for the second medical image based on the remaining feature amount of the second medical image.
  • the display device 16 shown in 10-2 of FIG. 10 displays the first and third first medical images and first evaluation index values (“1”, “2”) in chronological order.
  • the second, fourth, and last second medical images are displayed in association with the second evaluation index values (“2”, “3”, “2”) in this order.
  • the first evaluation index values are received for the two or more first medical images in the two or more first image group, and given to the two or more first medical images. Since the second evaluation index value for the second medical image not specified by the user is calculated based on the regression model obtained from the first evaluation index value and the feature amount of two or more first medical images, a better A second evaluation index value can be calculated.
  • Regression models include linear regression models, machine learning models such as regression and classification deep learning, neural networks, and SVM.
  • FIG. 11 is a functional block diagram showing a fifth embodiment of the medical image evaluation support device according to the present invention.
  • the medical image evaluation support apparatus 1-5 of the fifth embodiment shown in FIG. It differs from the medical image evaluation support device 1-2 of the second embodiment in that it includes a unit 58-4.
  • the evaluation index value prediction model 62 is the feature quantity associated with each medical image in the first image group, and the third evaluation index value It is a prediction model that predicts
  • the evaluation index value prediction model 62 receives the feature amount associated with each medical image of the first image group from the feature amount acquisition unit 52-1, and obtains an evaluation index value (third evaluation index value ), respectively.
  • the evaluation index value prediction model 62 includes feature amounts of a plurality of medical images acquired by the feature amount acquisition unit 52-1, and a plurality of evaluation index values given by the user to each of the plurality of medical images.
  • the evaluation index value prediction model 62 can apply a machine learning model such as regression type/classification type deep learning, neural network, SVM, etc. obtained by inputting a larger number of pairs.
  • the second evaluation index value calculator 58-4 calculates the third evaluation index value predicted by the evaluation index value prediction model 62 (the third evaluation index value predicted from the feature value associated with each medical image in the first image group ) and the first evaluation index value input by the first evaluation index value input unit 56, the second evaluation index value for the second medical image is calculated.
  • the second evaluation index value calculation unit 58-4 calculates, for example, the third evaluation index value predicted by the evaluation index value prediction model 62 from the feature quantity of the first medical image, and the user input via the first evaluation index value input unit 56. If the first evaluation index value for the first medical image that was input in the second medical image matches, the third evaluation index predicted from the feature amount of the second medical image is used as the second evaluation index value for the second medical image. The values are applied as they are, and if they do not match, the third evaluation index value predicted by the evaluation index value prediction model 62 is corrected based on the difference or ratio between the two, and the second evaluation for the second medical image is performed. Index values can be calculated.
  • the second evaluation index value calculation unit 58-4 calculates the third evaluation index value for the second medical image.
  • a second evaluation index value can be used, whereby fluctuations in the evaluation index value can be suppressed.
  • Processor 10 preferably displays the first group of images and a second group of images, different from the first group of images, on display device 16 for comparison.
  • FIG. 12 is a diagram showing a first display example of a display device that displays a first image group and a second image group to which evaluation index values are assigned.
  • the first group of images captured during the current examination of the same patient and the second group of images captured during the previous examination are displayed in a comparable manner.
  • Each medical image forming the first image group and each medical image forming the second image group are displayed in association with an evaluation index value.
  • the user designates one or a plurality of first medical images from among the medical images constituting the first image group captured during the current examination and inputs the first evaluation index value
  • the user can easily compare the current diagnosis result with the past (previous) diagnosis result.
  • FIG. 13 is a diagram showing a second display example of the display device that displays the first image group and the second image group to which the evaluation index values are assigned.
  • a first group of images captured during examination of patient A and a second group of images captured during examination of patient B, which is different from patient A, are displayed in a comparable manner. .
  • Each medical image forming the first image group and each medical image forming the second image group are displayed in association with an evaluation index value.
  • the examination of patient B It becomes possible to input based on the evaluation index values (first evaluation index value, second evaluation index value) assigned to each medical image constituting the second image group captured at that time. It also facilitates the user to compare diagnostic results of different patients.
  • FIG. 14 is a diagram showing a third display example of the display device displaying the first image group to which the initial evaluation index values are assigned and the first image group to which the evaluation index values are assigned by the medical image evaluation support device. .
  • the initial evaluation index value for each medical image that constitutes the first image group can be input by the user by operating the operation unit 20 and assigned to each medical image.
  • an evaluation index value is assigned by the medical image evaluation support device 1 to the same first image group as the first image group to which the initial evaluation index value has been assigned.
  • the first image group to which the initial evaluation index value is assigned and the first image group to which the evaluation index value is assigned by the medical image evaluation support apparatus 1 are displayed so as to be comparable. It is
  • the user can compare the initial evaluation index value given to the same first image group and the evaluation index value given by the medical image evaluation support device 1 this time.
  • the initial evaluation index value may be an evaluation index value given to the first group images by the medical image evaluation support apparatus 1 in the past.
  • an evaluation index value different from the initial evaluation index value is assigned.
  • FIG. 15 is a diagram showing a fourth display example of the display device displaying the first image group to which the initial evaluation index value is assigned and the first image group to which the evaluation index value is assigned by the medical image evaluation support device. .
  • the medical image evaluation support apparatus 1 selects and displays medical images and evaluation index values that differ from the evaluation index value and the initial evaluation index value. 14 is different from the display form of the display device 16 shown in FIG.
  • FIG. 16 is a diagram showing a first display example of a display device that displays a first image group to which evaluation index values are assigned and part information.
  • each medical image constituting the first image group is associated with an evaluation index value (“1”, “2”, “2”, “3”, . . . “1”). are displayed.
  • a schematic diagram of an organ to be inspected (in this example, the large intestine) is displayed, and an arrow that associates each medical image with the schematic diagram is added.
  • Processor 10 may add an arrow that associates each medical image with the schematic.
  • the user can easily confirm which part of the large intestine each medical image is an image of.
  • the schematic diagram and the arrows that associate each medical image with the schematic diagram correspond to the region information.
  • the processor 10 may display each medical image in association with the part name (text information) of the imaged part of the medical image instead of the part information such as a schematic diagram.
  • the site information By displaying the site information, the user can easily understand the inflammation level in association with the site, and can make a more appropriate diagnosis.
  • the part information is not limited to the case where the user inputs it corresponding to each medical image, and may be obtained by automatically detecting the imaging part. For example, by using image processing to detect landmarks peculiar to the region shown in each medical image, region information is acquired, and the region to be imaged is estimated from the amount of movement of the endoscope from the detected landmarks, etc. Alternatively, the part information may be obtained by determining the part to be imaged using a learning model.
  • an endoscope is inserted into the ileocecal region at the tip of the large intestine, and then each part of the large intestine is observed while pulling the endoscope. can be a landmark.
  • FIG. 17 is a diagram showing a second display example of the display device that displays the first image group to which the evaluation index value is assigned and the part information.
  • the display device 16 shown in FIG. 17 displays a schematic diagram of the large intestine in the same manner as the display device 16 shown in FIG. ) is selected, and only the medical images obtained by photographing the selected region from among the medical images forming the first image group are displayed together with their evaluation index values.
  • selected sites are displayed in gray, and unselected sites are indicated by dashed lines.
  • a specific part of the large intestine can be selected by moving the mouse cursor to the desired position on the schematic diagram and clicking.
  • one of the medical images constituting the first image group can be selected.
  • only the medical image of the selected site may be displayed.
  • FIG. 18 is a flow chart showing an embodiment of the operating method of the medical image evaluation support apparatus according to the present invention, and shows functions of the processor 10 of the medical image evaluation support apparatus 1 shown in FIG.
  • the processor 10 acquires a first image group consisting of a plurality of medical images (step S10), and displays at least one first medical image forming the first image group on the display device 16 (step S12). ).
  • the processor 10 receives the first evaluation index value for the first medical image displayed on the display device 16 from the first evaluation index value input unit operated by the user (step S14), and sets the received first evaluation index value to the first evaluation index value. Add to the medical image (step S16).
  • the second evaluation index value calculation unit of the processor 10 performs a second evaluation on a second medical image different from the first medical image to which the first evaluation index value is assigned by the user, among the medical images forming the first image group.
  • An index value is calculated (step S18).
  • the processor 10 preferably calculates a second evaluation index value for the second medical image based on the first evaluation index value, the feature amount of the first medical image, and the feature amount of the second medical image.
  • the processor 10 gives the second evaluation index value calculated in step S18 to the second medical image (step S20).
  • the processor 10 causes the display device 16 to display the first evaluation index value given by the user and the second evaluation index value calculated by the second evaluation index value calculation unit in association with each medical image, and notifies the user. preferably.
  • the first evaluation index value given by the user is referred to the first medical image designated by the user in the first image group, and the second medical image different from the first medical image in the first image group is obtained.
  • the evaluation index value for each medical image in the first image group can be assigned according to a unified standard, and the result can be confirmed on the display device 16 .
  • Hardware structures for executing various controls of the medical image evaluation support apparatus of the above embodiment are various processors as shown below.
  • the circuit configuration can be changed after manufacturing such as CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), which is a general-purpose processor that executes software (program) and functions as various control units.
  • Programmable Logic Device PLD
  • ASIC Application Specific Integrated Circuit
  • One processing unit may be composed of one of these various processors, or composed of two or more processors of the same type or different types (for example, a plurality of FPGAs, or a combination of a CPU and an FPGA).
  • a plurality of control units may be configured by one processor.
  • a single processor is configured by combining one or more CPUs and software.
  • a processor functions as multiple controllers.
  • SoC System On Chip
  • various control units are configured using one or more of the above various processors as a hardware structure.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Surgery (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Biomedical Technology (AREA)
  • Optics & Photonics (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Biophysics (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

L'invention concerne un dispositif d'aide à l'évaluation d'image médicale et un procédé de fonctionnement pour un dispositif d'aide à l'évaluation d'image médicale, qui sont capables de réduire la variation de valeurs d'indice d'évaluation conférées à de multiples images médicales et de conférer efficacement des valeurs d'indice d'évaluation. Ce dispositif d'aide à l'évaluation d'image médicale comprend un processeur qui acquiert un premier groupe d'images comprenant de multiples images médicales, affiche, sur un dispositif d'affichage, au moins une première image médicale constituant le premier groupe d'images, reçoit, en provenance d'un utilisateur, une première valeur d'indice d'évaluation pour la première image médicale, et transmet la première valeur d'indice d'évaluation à la première image médicale (3-1 sur la figure 3). Le processeur (10) calcule, sur la base de la première valeur d'indice d'évaluation communiquée à la première image médicale par l'utilisateur, une seconde valeur d'indice d'évaluation pour une seconde image médicale, qui diffère de la première image médicale et constitue le reste du premier groupe d'images à distance de la première image médicale ; et le processeur (10) transmet la seconde valeur d'indice d'évaluation à la seconde image médicale (3-2 sur la figure 3).
PCT/JP2022/037775 2021-11-11 2022-10-11 Dispositif d'aide à l'évaluation d'image médicale et procédé de fonctionnement pour dispositif d'aide à l'évaluation d'image médicale WO2023084969A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2021184042 2021-11-11
JP2021-184042 2021-11-11

Publications (1)

Publication Number Publication Date
WO2023084969A1 true WO2023084969A1 (fr) 2023-05-19

Family

ID=86335653

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/037775 WO2023084969A1 (fr) 2021-11-11 2022-10-11 Dispositif d'aide à l'évaluation d'image médicale et procédé de fonctionnement pour dispositif d'aide à l'évaluation d'image médicale

Country Status (1)

Country Link
WO (1) WO2023084969A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7465409B2 (ja) 2022-01-19 2024-04-10 コ，ジファン 人工知能基盤の血管学習による大腸ポリープ検出方法及び装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007275589A (ja) * 2006-03-31 2007-10-25 Given Imaging Ltd 患者の状態を判断するためのシステム及び方法
WO2020066807A1 (fr) * 2018-09-27 2020-04-02 Hoya株式会社 Système d'endoscope électronique
JP2020065685A (ja) * 2018-10-24 2020-04-30 富士フイルム株式会社 内視鏡システム
WO2020100630A1 (fr) * 2018-11-14 2020-05-22 富士フイルム株式会社 Système de traitement d'image médicale
WO2020162275A1 (fr) * 2019-02-08 2020-08-13 富士フイルム株式会社 Dispositif de traitement d'image médicale, système d'endoscope et procédé de traitement d'image médicale

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007275589A (ja) * 2006-03-31 2007-10-25 Given Imaging Ltd 患者の状態を判断するためのシステム及び方法
WO2020066807A1 (fr) * 2018-09-27 2020-04-02 Hoya株式会社 Système d'endoscope électronique
JP2020065685A (ja) * 2018-10-24 2020-04-30 富士フイルム株式会社 内視鏡システム
WO2020100630A1 (fr) * 2018-11-14 2020-05-22 富士フイルム株式会社 Système de traitement d'image médicale
WO2020162275A1 (fr) * 2019-02-08 2020-08-13 富士フイルム株式会社 Dispositif de traitement d'image médicale, système d'endoscope et procédé de traitement d'image médicale

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7465409B2 (ja) 2022-01-19 2024-04-10 コ，ジファン 人工知能基盤の血管学習による大腸ポリープ検出方法及び装置

Similar Documents

Publication Publication Date Title
RU2571387C2 (ru) Преобразование данных пациента в медицинские инструкции
JP6899903B2 (ja) 医療画像処理装置、内視鏡装置、診断支援装置、医療業務支援装置、及び、レポート作成支援装置
JP2024045234A (ja) 腸の病理学のための画像スコアリング
JP7258354B2 (ja) 生体組織内の異常を検出するための方法及びシステム
JP6834001B2 (ja) 医療画像処理装置、診断支援装置、医療業務支援装置、及び、レポート作成支援装置
US11544844B2 (en) Medical image processing method and apparatus
EP4091532A1 (fr) Dispositif de traitement d'image médicale, système endoscopique, méthode d'aide au diagnostic, et programme
US20240105143A1 (en) Medical image processing apparatus
WO2019082741A1 (fr) Dispositif de traitement d'image médicale
WO2023084969A1 (fr) Dispositif d'aide à l'évaluation d'image médicale et procédé de fonctionnement pour dispositif d'aide à l'évaluation d'image médicale
US20220375610A1 (en) Multi-Variable Heatmaps for Computer-Aided Diagnostic Models
WO2020040086A1 (fr) Système de traitement d'image médicale
US20220108801A1 (en) Diagnosis and treatment support system
WO2020039929A1 (fr) Dispositif de traitement d'image médicale, système endoscopique, et procédé de fonctionnement d'un dispositif de traitement d'image médicale
WO2020039685A1 (fr) Dispositif de traitement d'image médicale
US20230245758A1 (en) Medical image processing apparatus and operation method thereof
US20230245312A1 (en) Medical image processing apparatus and operation method thereof
JP6834019B2 (ja) 医療画像処理装置、及び、内視鏡装置
Spaide et al. Geographic atrophy segmentation using multimodal deep learning
US20170140102A1 (en) Examination Result Sheet Creation Apparatus, Examination Result Sheet Creation Method, Non-Transitory Computer Readable Medium, Examination Result Sheet, and Examination Apparatus
US20210272254A1 (en) Image analysis device, analysis function decision method, and analysis function decision program
JP7483326B2 (ja) 医用画像処理装置、方法及びシステム
WO2022054373A1 (fr) Dispositif d'image médicale et son procédé de fonctionnement
JP6866497B2 (ja) 医療画像処理装置、及び、内視鏡装置
US20230143966A1 (en) Image processing apparatus and non-transitory computer readable medium

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: 22892469

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2023559486

Country of ref document: JP

Kind code of ref document: A