WO2019102917A1 - Dispositif, procédé et programme de détermination de radiologue - Google Patents

Dispositif, procédé et programme de détermination de radiologue Download PDF

Info

Publication number
WO2019102917A1
WO2019102917A1 PCT/JP2018/042142 JP2018042142W WO2019102917A1 WO 2019102917 A1 WO2019102917 A1 WO 2019102917A1 JP 2018042142 W JP2018042142 W JP 2018042142W WO 2019102917 A1 WO2019102917 A1 WO 2019102917A1
Authority
WO
WIPO (PCT)
Prior art keywords
image interpretation
image
interpretation doctor
degree
doctor
Prior art date
Application number
PCT/JP2018/042142
Other languages
English (en)
Japanese (ja)
Inventor
晶路 一ノ瀬
佳児 中村
王　彩華
瑞希 武井
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 WO2019102917A1 publication Critical patent/WO2019102917A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Definitions

  • the present invention relates to an image interpretation doctor determination apparatus, method, and program for determining an image interpretation doctor who interprets a medical image.
  • CT Computer Tomography
  • MRI Magnetic Resonance Imaging
  • CAD Computer-Aided Diagnosis
  • CAD Computer-Aided Diagnosis
  • the analysis result generated by the analysis process is associated with examination information such as patient name, gender, age, and modality for acquiring a medical image, stored in a database, and provided for diagnosis.
  • a radiologist or the like who has acquired the medical image determines an image interpretation doctor corresponding to the medical image, and notifies the determined image interpretation doctor that the analysis result by the medical image and CAD exists.
  • the image interpretation doctor interprets the medical image with reference to the delivered medical image and the analysis result in the image interpretation terminal of the image interpretation terminal to create an image interpretation report.
  • the image interpretation doctor can judge the reliability of the analysis result by CAD. It is possible to carefully read and interpret low-quality medical images. However, if the image interpretation doctor does not have the skill to appropriately interpret, even if the reliability of the analysis result by CAD is known, it is not possible to appropriately interpret the medical image. In particular, when the reliability of the analysis result by CAD is low, an image reading doctor with low image reading skills may miss an area having hemorrhage or infarction, or a lesion when the image reading is performed.
  • the present invention has been made in view of the above circumstances, and it is an object of the present invention to appropriately determine a medical image interpretation doctor based on analysis results of medical images by CAD or the like.
  • An image interpretation doctor determination apparatus is an analysis unit that analyzes a medical image and acquires an analysis result. And a determination unit that determines a radiologist or radiologist group that interprets the medical image based on the analysis result.
  • the image interpretation doctor group means a group to which a plurality of image interpretation doctors belong.
  • the image interpretation doctor who belongs to the image interpretation doctor group is not limited to the doctor in the same hospital, and may be spread over a plurality of hospitals.
  • the determination unit acquires at least one of the degree of difficulty of image interpretation and the degree of urgency of image interpretation based on the analysis result, and at least one of the difficulty and degree of emergency.
  • the image interpretation doctor or the image interpretation doctor group may be determined based on
  • the determination unit may determine at least one of the difficulty level and the urgency level by referring to a first database in which various analysis results are associated with at least one of the difficulty level and the urgency level. .
  • the determination unit has a first evaluation unit that outputs a first evaluation value for determining at least one of the difficulty level and the urgency level by the input of the analysis result.
  • a first evaluation unit that outputs a first evaluation value for determining at least one of the difficulty level and the urgency level by the input of the analysis result.
  • at least one of the degree of difficulty and the degree of urgency may be determined based on the first evaluation value.
  • the determination section refers to the second image database in which at least one of various difficulty levels and various emergency levels is associated with the image interpretation doctor or the image interpretation doctor group.
  • the image interpretation doctor group may be determined.
  • the determination unit outputs a second evaluation value for determining an image interpretation doctor or an image interpretation doctor group by inputting at least one of the difficulty level and the urgency level. It may have an evaluation unit and determine a radiologist or radiologist group based on the second evaluation value.
  • the image interpretation doctor determination apparatus further includes a certainty factor calculation unit that calculates the certainty factor of the analysis result, The determination unit may determine at least one of the degree of difficulty and the degree of urgency based on the degree of certainty.
  • the analysis unit identifies the disease and the disease site included in the medical image, and calculates the probability that the identified disease site is the identified disease
  • the certainty factor calculation unit may calculate the certainty factor based on the probability.
  • the determination unit has a third evaluation unit that outputs a third evaluation value for determining an image interpretation doctor or an image interpretation doctor group by input of the analysis result.
  • the image interpretation doctor or the image interpretation doctor group may be determined based on the third evaluation value.
  • the medical image may include the brain or heart of the patient.
  • the analysis unit may identify a bleeding area, an infarct area or an ischemic area included in a medical image.
  • the determination section may display the determined image interpretation doctor or image interpretation doctor group on the display section.
  • the determination unit may notify the determined image interpretation doctor or the image interpretation doctor group of the determination.
  • the image interpretation doctor determination method analyzes a medical image to obtain an analysis result, Based on the analysis result, an image interpretation doctor or an image interpretation doctor group that interprets the medical image is determined.
  • the image interpretation doctor determination method according to the present invention may be provided as a program for causing a computer to execute.
  • Another image interpretation doctor determination apparatus is a memory for storing instructions for causing a computer to execute;
  • a processor configured to execute the stored instructions, the processor Analyze medical images and obtain analysis results, Based on the analysis result, a process of determining a radiologist or radiologist group that interprets the medical image is executed.
  • a medical image is analyzed, and a radiologist or radiologist group that interprets the medical image is determined based on the analysis result. For this reason, it is possible to appropriately determine a radiologist or radiologist group who interprets a medical image from the analysis result.
  • Diagram for explaining the process performed by the analysis unit Diagram showing the first database Diagram showing the second database Diagram showing an interpretation list displayed on an interpretation workstation Flow chart showing processing performed in the first embodiment
  • FIG. 3 a diagram showing an example in which the maximum value of the certainty factor of each pixel is different Flow chart showing processing performed in the second embodiment Diagram for explaining other processing performed by the analysis unit
  • FIG. 1 is a view showing a schematic configuration of a medical information system to which an image interpretation doctor determination apparatus according to an embodiment of the present invention is applied.
  • the medical information system 1 shown in FIG. 1 is based on an examination order from a doctor of a medical department using a well-known ordering system, imaging of a region to be examined of a subject, storage of medical images acquired by imaging, This is a system for performing interpretation of medical images and creation of an interpretation report, viewing of interpretation reports by a doctor of a medical department requesting the examination, and detailed observation of medical images to be interpreted. As shown in FIG.
  • the medical information system 1 includes a plurality of modalities (imaging devices) 2, a plurality of image interpretation workstations (WS) 3 which are image interpretation terminals, a medical department workstation (WS) 4, an image server 5, and images.
  • the database 6, the interpretation report server 7, the interpretation report database 8, and the interpretation doctor determination apparatus 10 according to the present embodiment are connected and configured to be able to communicate with each other via the network 9.
  • Each device is a computer in which an application program for functioning as a component of the medical information system 1 is installed.
  • the application program is distributed by being recorded in a recording medium such as a digital versatile disc (DVD) or a compact disc read only memory (CD-ROM), and is installed in the computer from the recording medium.
  • DVD digital versatile disc
  • CD-ROM compact disc read only memory
  • it is stored in a storage device of a server computer connected to a network or in a network storage in an accessible state from the outside, downloaded to the computer in response to a request, and installed.
  • the modality 2 is a device that generates a medical image representing a region to be diagnosed by imaging the region to be diagnosed of a subject. Specifically, a CT apparatus, an MRI apparatus, a PET (Positron Emission Tomography) apparatus, and the like.
  • the medical image generated by the modality 2 is transmitted to the image server 5 and stored.
  • the image interpretation WS3 is a computer used by a medical image interpreting doctor to create an image interpretation report and an image interpretation report of a medical image, and is configured by a processing device, a high definition display, and an input device such as a keyboard and a mouse.
  • a request for browsing medical images to the image server 5 various image processing on medical images received from the image server 5
  • Each process such as a browsing request and display of an interpretation report received from the interpretation report server 7 is performed by executing a software program for each process.
  • the image processing server may be separately connected to the network 9 without performing various image processing in the image reading WS3, and may be performed in response to a processing request from the image reading WS3.
  • the medical department WS4 is a computer used by doctors in the medical department for detailed observation of an image, viewing of an interpretation report, creation of an electronic medical record, etc., and includes a processing device, a high definition display, and an input device such as a keyboard and a mouse. Be done.
  • a request for image browsing to the image server 5 display of an image received from the image server 5, automatic detection or highlighting of a lesion or the like in the image, a request for viewing an interpretation report for the interpretation report server 7, and interpretation
  • Each process such as display of an interpretation report received from the report server 7 is performed by executing a software program for each process.
  • the image server 5 is a general-purpose relatively high-performance computer on which a software program for providing the function of a database management system (DBMS) is installed. Further, the image server 5 is provided with a large capacity storage in which the image database 6 is configured. This storage may be a large-capacity hard disk drive connected to the image server 5 by a data bus, or connected to a NAS (Network Attached Storage) and a SAN (Storage Area Network) connected to the network 9. It may be a disk device. Further, when receiving the medical image registration request from the modality 2, the image server 5 arranges the medical image into a database format and registers the medical image in the image database 6.
  • DBMS database management system
  • the incidental information includes, for example, an image ID for identifying an individual medical image, a patient ID (identification) for identifying an object, an examination ID for identifying an examination, and a unique ID assigned to each medical image.
  • UID unique identification
  • date of examination when the medical image was generated is generated
  • examination time is associated with type of modality used in examination to acquire the medical image
  • patient name is associated with patient information
  • patient information such as age, gender, examination site ( Includes imaging site), imaging information (imaging protocol, imaging sequence, imaging method, imaging conditions, use of contrast agent, etc.), and information such as the serial number or acquisition number when multiple medical images were acquired in a single examination It is possible.
  • the image server 5 receives the browsing request from the image interpretation WS 3 via the network 9, the image server 5 searches for medical images registered in the image database 6 and transmits the extracted medical image to the image interpretation WS 3 of the request source.
  • the diagnostic reading report server 7 incorporates a software program for providing a database management system (DBMS) function to a general-purpose computer.
  • DBMS database management system
  • an image ID for identifying a medical image to be interpreted for example, an image ID for identifying a medical image to be interpreted, an interpretation doctor ID for identifying an imaging diagnostician who has performed an interpretation, a lesion name, location information of a lesion, a finding, a certainty factor of finding
  • An interpretation report in which information such as is recorded is registered.
  • the network 9 is a wired or wireless local area network that connects various devices in the hospital.
  • the network 9 may be configured by connecting local area networks of each hospital by the Internet or a dedicated line. In any case, it is preferable that the network 9 be configured to realize high-speed transfer of medical images, such as an optical network.
  • the image interpretation doctor determination apparatus 10 is obtained by installing the image interpretation doctor determination program of the present embodiment in one computer.
  • the computer may be a workstation or a personal computer directly operated by a doctor performing diagnosis, or a server computer connected with them via a network.
  • the image interpretation doctor determination program is recorded on a recording medium such as a DVD or a CD-ROM, distributed, and installed in the computer from the recording medium.
  • it is stored in a storage device of a server computer connected to a network or in a network storage in an accessible state from the outside, downloaded to the computer in response to a request, and installed.
  • FIG. 2 is a diagram showing a schematic configuration of an image interpretation doctor determination apparatus according to the first embodiment of the present invention, which is realized by installing an image interpretation doctor determination program in a computer.
  • the image interpretation doctor determination apparatus 10 includes a central processing unit (CPU) 11, a memory 12, and a storage 13 as a standard workstation configuration. Further, a display 14 such as a high definition liquid crystal display, and an input unit 15 such as a keyboard and a mouse are connected to the image interpretation doctor determination device 10.
  • the storage 13 is formed of a storage device such as a hard disk or a solid state drive (SSD).
  • the storage 13 stores various information including medical images and information necessary for processing acquired from the image server 5 via the network 9.
  • the medical image for determining the image interpretation doctor may be directly transmitted from the modality 2 to the image interpretation doctor determination apparatus 10, but may be transmitted to the image interpretation doctor determination apparatus 10 after being temporarily stored in the image server 5. Good.
  • an image interpretation doctor determination program is stored.
  • the image interpretation doctor determination program analyzes the medical image and acquires an analysis result as processing to be executed by the CPU 11, and determines an image interpretation doctor or an image interpretation doctor group that interprets the medical image based on the analysis result. To define. These processes are performed when the image interpretation doctor determination apparatus 10 receives, from the image server 5, a notification that a new medical image to be interpreted is stored in the image server 5.
  • the computer functions as the analysis unit 21 and the determination unit 22 as the CPU 11 executes these processes in accordance with the program.
  • the CPU 11 executes the function of each unit by the image interpretation doctor determination program, but as a general purpose processor that executes software and functions as various processing units, in addition to the CPU 11,
  • a programmable logic device which is a processor whose circuit configuration can be changed after manufacturing an FPGA (field programmable gate array) or the like can be used.
  • the processing of each part may be executed by a dedicated electric circuit or the like which is a processor having a circuit configuration designed exclusively for executing a specific processing such as an application specific integrated circuit (ASIC).
  • ASIC application specific integrated circuit
  • One processing unit may be configured of one of these various types of processors, or a combination of two or more processors of the same or different types (for example, a plurality of FPGAs or a combination of a CPU and an FPGA) It may be configured.
  • a plurality of processing units may be configured by one processor.
  • one processor is configured by a combination of one or more CPUs and software as represented by computers such as clients and servers.
  • a processor functions as a plurality of processing units.
  • SoC system on chip
  • IC integrated circuit
  • the hardware-like structure of these various processors is more specifically an electric circuit (circuitry) combining circuit elements such as semiconductor elements.
  • the analysis unit 21 analyzes a medical image by CAD or the like to acquire an analysis result. Specifically, the disease and disease site included in the medical image are specified, the probability that the specified disease site is the identified disease is calculated, and the disease and disease site included in the medical image is specified based on the probability. To obtain the analysis result. For this purpose, the analysis unit 21 performs learning using a deep learning method in order to calculate the probability of being a certain disease for a pixel or pixel region of interest in a medical image. It has a convolutional neural network (CNN: Convolutional Neural Network). The CNN takes a feature amount in a medical image as an input, and outputs a probability indicating which one of a plurality of diseases each disease of the medical image is.
  • CNN convolutional neural network
  • a three-dimensional image of the brain is input, and the diseases are hemorrhage and infarction.
  • the analysis unit 21 outputs the probability that each pixel of the three-dimensional image is a bleeding area, the probability that it is an infarct area, and the probability that it is a background area that does not belong to either the bleeding area or the infarct area. Do.
  • the probability takes a value between 0 and 1, and the larger the value, the higher the probability of the disease. Also, the sum of the three probabilities is one. Then, the analysis unit 21 classifies the pixel into the disease having the largest probability among the probabilities for the three diseases calculated for each pixel.
  • the disease which the analysis part 21 specifies as an analysis result is not limited to a hemorrhage area
  • FIG. 3 is a diagram for explaining the process performed by the analysis unit 21.
  • CNN outputs a probability indicating whether each pixel of the medical image is a bleeding area, an infarct area, or a background area.
  • the background area is indicated as class 1
  • the bleeding area is indicated as class 2
  • the infarct area is indicated as class 3 for the sake of explanation.
  • the medical image is assumed to be composed of 3 ⁇ 4 12 pixels.
  • the probability of class 1 the probability of the pixel surrounded by a thick line is a large value compared to the probability of the same pixel of class 2 and class 3.
  • the analysis unit 21 classifies each pixel into any one of three classes 1 to 3 as shown in the right side of FIG.
  • the analysis unit 21 specifies the bleeding site for the pixel classified into the bleeding area according to the position in the brain where the pixel is present.
  • the bleeding area is specified to three bleeding sites: putamen hemorrhage, subcortical hemorrhage and cerebellar hemorrhage.
  • the analysis unit 21 also calculates the bleeding volume and the infarct volume by multiplying the number of pixels by the volume per voxel of the medical image for the pixels classified into the bleeding area and the infarct area.
  • the analysis unit 21 acquires the disease included in the medical image, the bleeding site when the disease is bleeding, and the volume of the bleeding area as an analysis result.
  • the determination unit 22 determines an image interpretation doctor or an image interpretation doctor group that interprets the medical image. Specifically, based on the analysis result, the determination unit 22 acquires at least one of the degree of difficulty of image interpretation and the degree of urgency of image interpretation, and the image interpretation doctor based on at least one of the degree of difficulty and the degree of urgency. Or decide on a group of interpreting physicians. In the present embodiment, an image interpretation doctor or an image interpretation doctor group is determined based on both the degree of difficulty and the level of urgency. For this purpose, the determination unit 22 determines the degree of difficulty and the degree of urgency with reference to the first database DB1 in which various analysis results are associated with the degree of difficulty and the degree of urgency. The first database DB1 is stored in the storage 13.
  • FIG. 4 is a diagram showing a first database.
  • the level of urgency and difficulty according to the bleeding volume is associated with the bleeding site (putamen hemorrhage, subcortical hemorrhage and cerebellar hemorrhage) in the hemorrhage area There is.
  • the degree of urgency is higher in the order of C, B and A
  • the degree of difficulty is higher in the order of C, B and A.
  • the degree of urgency is determined to be A
  • the degree of difficulty is determined to be C.
  • the determination unit 22 refers to the second database DB2 in which various difficulty levels and various urgency levels are associated with the image reading doctor or the image reading doctor group based on the determined difficulty level and the urgency level. Or decide on a group of interpreting physicians.
  • the second database DB2 is stored in the storage 13.
  • FIG. 5 is a diagram showing a second database.
  • the name of the image interpretation doctor and the name of the image interpretation doctor group, affiliation, specialty, career, attendance status, available difficulty, and availability urgency are associated.
  • the degree of urgency is A
  • the degree of difficulty is C
  • Doctor yamada, and Group 2 (Group 2) will be determined as the radiologist and radiologist group.
  • only one of the image interpretation doctor and the image interpretation doctor group may be determined in the determination unit 22.
  • priority is set in advance as to whether to select the image interpretation doctor preferentially or to select the image interpretation doctor group in advance, and the image interpretation is performed by referring to the second database DB2.
  • only the image interpretation doctor may be selected if the image interpretation doctor has been set as a priority.
  • the radiologist group is a group to which a plurality of radiologists belong, and some radiologists have highly experienced radiography skills, and some radiologists have less experience and poor radiology skills. For this reason, the column of career is blank (-).
  • the determined image interpretation doctor or image interpretation doctor group information is sent to the image interpretation report server 7 and sent to the determined image interpretation doctor or image interpretation doctor group WS3 as an image interpretation list representing medical images to be interpreted, and image interpretation It is displayed on the display of WS3.
  • FIG. 6 is a diagram showing an image interpretation list displayed on the image interpretation WS3.
  • the image interpretation list L1 shows the patient name, gender, age, examinations performed and urgency.
  • the image interpretation doctor or the image interpretation doctor who belongs to the image interpretation doctor group selects the patient name from the image interpretation list L1 to display the medical image on his / her image interpretation WS3 to perform the image interpretation.
  • the information may be notified to the image interpretation doctor determined or the image interpretation doctor belonging to the image interpretation doctor group. Specifically, notification may be made to the portable terminal possessed by the image interpretation doctor or the image interpretation doctor group belonging to the image interpretation doctor group that the image interpretation doctor has been determined. Thus, the image interpretation doctor or the image interpretation doctor belonging to the image interpretation doctor group can confirm that there is a medical image to be read by the image interpretation doctor.
  • the determined name of the image interpretation doctor or the name of the image interpretation doctor group may be displayed on the display 14 of the image interpretation doctor determination apparatus 10.
  • FIG. 7 is a flowchart showing the process performed in the first embodiment.
  • the analysis unit 21 acquires a medical image from the image server 5 (step ST1), and the medical image is received. The analysis is performed to obtain an analysis result (step ST2).
  • the determination unit 22 refers to the first database DB1 to acquire the difficulty level and the urgency level (step ST3), and further refers to the second database DB2 to interpret based on the difficulty level and the urgency level.
  • a doctor or radiologist group is determined (step ST4). Then, the information on the determined image interpretation doctor or image interpretation doctor group is transmitted to the image interpretation report server 7 (step ST5), and the process is ended.
  • a medical image is analyzed, and an image interpretation doctor or an image interpretation doctor group that interprets the medical image is determined based on the analysis result. For this reason, it is possible to appropriately determine a radiologist or radiologist group who interprets a medical image from the analysis result.
  • At least one of the degree of difficulty of image reading and the degree of urgency of image reading is acquired based on the analysis result, and the image reading doctor or image reading doctor group is determined based on at least one of the degree of difficulty and urgency I made it.
  • an image reading doctor or image reading doctor group can be determined according to at least one of the degree of difficulty and the degree of urgency of diagnosis. For example, if the degree of difficulty of interpretation is high or the degree of urgency of interpretation is high, if the degree of difficulty is high by determining an interpretation doctor or an interpretation doctor group having a high skill of interpretation, or the degree of urgency Medical image can be read appropriately.
  • the image interpretation doctors belonging to the image interpretation doctor group may span not only within the same hospital but also across hospitals.
  • the affiliation of the image interpretation doctor registered in the second database DB2 also specifies the hospital name.
  • FIG. 8 is a diagram showing a schematic configuration of an image interpretation doctor determination apparatus according to a second embodiment of the present invention.
  • the same reference numerals in FIG. 8 denote the same parts as in FIG. 2 and a detailed description will be omitted.
  • the image interpretation doctor determination apparatus 10 according to the second embodiment includes a certainty factor calculation unit 23 that calculates the certainty factor of the analysis result, and the determination unit 22 determines the difficulty level based on the certainty factor.
  • the second embodiment differs from the first embodiment in that the second embodiment is configured.
  • the analysis unit 21 outputs the probability of which one of a plurality of diseases each of the pixels of the medical image is a disease by CNN, and classifies the pixel into a disease having the highest probability. That is, the larger the probability of the CNN output, the higher the degree of certainty of the disease likeness of each pixel. That is, when the class of diseases to be classified is three classes, if the probability of one class for a certain pixel, for example, the probability of class 2 is 0.9, that pixel is assigned to class 2. On the other hand, even if the probability of class 2 is 0.34 for a certain pixel, that pixel is assigned to class 2. However, there is a large difference in the reliability of the analysis result between the probability of 0.9 and 0.34.
  • the image reading doctor can determine as class 2 by looking at the pixel regardless of the image reading skill.
  • the probability is 0.34, in order to determine the pixel as class 2, it may be difficult to determine that it is not a highly skilled reading doctor.
  • the certainty factor calculation unit 23 calculates the average value of the maximum value of the probability of each pixel calculated by the analysis unit 21 as the certainty factor.
  • FIG. 9 is a diagram in which only the maximum values of classes 1 to 3 are assigned to each pixel in the output of the CNN shown in FIG.
  • FIG. 10 is a diagram showing an example in which the class of each pixel can be assigned as shown on the left side of FIG. 3 but the maximum value of the probability of each pixel is different.
  • the average value of the maximum value of probability that is, the certainty factor is 0.72.
  • the average value of the maximum value of the probability that is, the certainty factor is 0.39.
  • the determination unit 22 determines the difficulty level with reference to the certainty factor calculated by the certainty factor calculation unit 23.
  • the degree of certainty is determined so that the degree of difficulty C is 0 or more and less than 0.33, the degree of difficulty B is 0.33 or more and less than 0.66, and the degree of difficulty A is 0.66 or more and 1.0 or less.
  • the determination unit 22 refers only to the difficulty level for the first database DB1.
  • FIG. 11 is a flowchart showing the process performed in the second embodiment.
  • the analysis unit 21 acquires a medical image from the image server 5 (step ST11), and the medical image is received. Analysis is performed to obtain an analysis result (step ST12).
  • the certainty factor calculation unit 23 calculates the certainty factor of the analysis result (step ST13). Then, the determination unit 22 refers to the first database DB1 to acquire the difficulty level and the urgency level (step ST14), and further refers to the second database DB2 to interpret based on the difficulty level and the urgency level. A doctor or radiologist group is determined (step ST15). Then, the information on the determined image interpretation doctor or image interpretation doctor group is transmitted to the image interpretation report server 7 (step ST16), and the process is ended.
  • the degree of certainty of the analysis result is calculated, and the degree of difficulty is determined based on the degree of certainty, thereby appropriately determining the degree of difficulty of interpretation of the medical image according to the analysis result. can do.
  • the degree of difficulty is determined based on the degree of certainty of the analysis result, but both the degree of difficulty and the degree of urgency or only the degree of urgency may be determined.
  • the degree of difficulty is determined in accordance with the value of the probability output by the CNN in the analysis unit 21.
  • the analysis unit 21 may use, for example, a classifier such as a support vector machine that classifies each pixel of a medical image into two classes using linear input elements.
  • a classifier such as a support vector machine that classifies each pixel of a medical image into two classes using linear input elements.
  • FIG. 12 the output value that is the classification result is classified into a tumor and a non-tumor by separation plane A0. .
  • the pixels whose output values are plotted above the separation plane A0 are classified as tumors, and the pixels whose output values are plotted below are classified as non-tumors.
  • the further the output value is from the classification plane A0 the higher the degree of certainty that it is tumor and non-tumor.
  • the certainty factor calculation unit 23 outputs the output of each pixel.
  • the distance from the separation plane A0 of the value may be calculated as the certainty factor.
  • FIG. 13 is a diagram showing a schematic configuration of an image interpretation doctor determination apparatus according to a third embodiment of the present invention.
  • the same reference numerals in FIG. 13 denote the same parts as in FIG. 2 and a detailed description will be omitted.
  • the image interpretation doctor determination apparatus 10 according to the third embodiment is different from the first embodiment in that the determination unit 22 includes the first evaluation unit 22A and the second evaluation unit 22B. .
  • the first evaluation unit 22A outputs a first evaluation value for determining at least one of the degree of difficulty and the degree of urgency by the input of the analysis result by the analysis unit 21. In the third embodiment, both the degree of difficulty and the degree of urgency are determined.
  • the first evaluation unit 22A is, for example, a discriminator that receives an analysis result and is trained to output a first evaluation value for determining the degree of difficulty and the degree of urgency.
  • the first evaluation value represents, for example, the difficulty level and the urgency level as values between 0 and 1, for example. The higher the value, the higher the degree of difficulty and urgency.
  • the determination unit 22 determines the degree of difficulty and the degree of urgency according to the first evaluation value output by the first evaluation unit 22A.
  • the first evaluation value of the difficulty level is 0 or more and less than 0.33 for difficulty C, 0.33 or more and less than 0.66 for difficulty B, and 0.66 or more and 1.0 or less for difficulty A Determine the level of difficulty.
  • the first evaluation value for the degree of urgency is 0 or more and less than 0.33, the degree of urgency C, 0.33 or more and less than 0.66 the degree of urgency B, 0.66 or more and 1.0 or less the degree of urgency A Determine the degree of urgency.
  • the second evaluation unit 22B outputs a second evaluation value for determining a radiologist or radiologist group by inputting at least one of the degree of difficulty and the degree of urgency.
  • the second evaluation value is output by inputting both the degree of difficulty and the degree of urgency.
  • the second evaluation unit 22B includes, for example, a discriminator trained with the degree of difficulty and the degree of urgency as inputs and outputting a second evaluation value for determining a radiologist or radiologist group.
  • the difficulty level and the urgency level to be input may be A, B, and C described above, but the first evaluation value itself may be input.
  • the second evaluation value is represented, for example, as a value between 0 and 1 for each image reading doctor or each image reading doctor group. It is assumed that the image reading doctor or image reading doctor group suitable for image reading as the value is larger.
  • the determination unit 22 determines a radiologist or radiologist group according to the second evaluation value output by the second evaluation unit 22B. That is, the radiologist or radiologist group whose second evaluation value is the largest is determined as the radiologist or radiologist group who interprets the medical image.
  • the first evaluation unit 22A and the second evaluation unit 22B it is possible to determine a radiologist or radiologist group who interprets a medical image.
  • the determination unit 22 may be configured to include only the first evaluation unit 22A. Good. In this case, the degree of difficulty and the degree of urgency are determined based on the first evaluation value output by the first evaluation unit 22A, and then the second database DB2 is referred to as in the first embodiment. A radiologist or radiologist group may be determined. Further, the determination unit 22 may be configured to include only the second evaluation unit 22B. In this case, as in the first embodiment, the difficulty level and the urgency level are determined with reference to the first database DB1 based on the analysis result, and thereafter, the second embodiment is the same as the third embodiment. The degree of difficulty and the degree of urgency may be input to the evaluation unit 22B, and the image reading doctor or the image reading doctor group may be determined based on the second evaluation value output by the second evaluation unit 22B.
  • the third evaluation unit 22A and the second evaluation unit 22B are provided, a third system for determining an image interpretation doctor or an image interpretation doctor group by inputting an analysis result is described.
  • the image evaluation doctor or the image reading doctor group may be determined based on the third evaluation value.
  • the third evaluation unit is a discriminator that is trained to receive an analysis result and output a third evaluation value for determining an image reading doctor or an image reading doctor group.
  • the third evaluation value is represented, for example, as a value between 0 and 1 for each image reading doctor or each image reading doctor group. It is assumed that the image reading doctor or image reading doctor group suitable for image reading as the value is larger.
  • the determination unit 22 determines an image interpretation doctor or an image interpretation doctor group according to the third evaluation value output by the third evaluation unit. That is, the radiologist or radiologist group having the largest third evaluation value is determined as the radiologist or radiologist group who interprets the medical image.
  • the image interpretation doctor or the image interpretation doctor group is given a score, and the image interpretation doctor or the image interpretation doctor group having the highest score is determined as the image interpretation doctor or the image interpretation doctor group who interprets the medical image. It is good also as things. For example, the distance to a hospital when an image interpreting doctor or an image interpreting doctor belonging to an image interpreting doctor group (hereinafter referred to simply as an image interpreting doctor) is not in a hospital, there is a problem with the presence of awareness entered by emergency personnel in case of emergency, Information such as region and age may be scored, a weighted score for these scores may be calculated, and a radiologist or radiologist group may be determined based on the weighted score.
  • the information on the image interpretation doctor or the image interpretation doctor group is transmitted to the image interpretation report server 7, but the image interpretation doctor determination apparatus 10 generates the image interpretation list L1 shown in FIG. It may be transmitted to the image interpretation doctor who belongs to the read image interpretation doctor or the image interpretation doctor group.
  • the image interpretation doctor may be able to correct the second database DB2. That is, when the image reading doctor's image reading skill is improved and it becomes possible to cope with the reading of medical images with high degree of difficulty and urgency, it is possible to cope with the level of difficulty and the degree of possible response in the second database DB2.
  • the image reading doctor may be able to make corrections.
  • the image interpretation doctor may access the image interpretation doctor determination apparatus 10 from his / her image interpretation WS3 to correct the second database DB2.
  • the first database DB1 and the second database DB2 are stored in the storage 13 of the image interpretation doctor determination apparatus 10, but are stored in the image server 5 or the image interpretation report server 7 etc.
  • the doctor determining apparatus 10 may access the image server 5 or the image interpretation report server 7 to refer to the first database DB1 and the second database DB2.
  • the medical image is an image including the brain, and each pixel of the medical image is classified into any of a bleeding area, an infarct area, and a background area.
  • the analysis unit 21 outputs a probability for each of the hemorrhage area and the infarct area, but it is possible to identify which of the hemorrhage area and the infarct area is suspected with the probability for the target medical image. . Therefore, by referring to the probability obtained by analysis of the medical image, it may be identified which of the bleeding area and the infarct area is suspected with respect to the medical image.
  • hemorrhage is often in charge of neurosurgery
  • infarct is often in charge of neurology.
  • the analysis unit 21 identifies which of the bleeding area and the infarct area is suspected, and the image reading doctor or the image reading doctor group is determined in accordance with the disease identified as having a high intensity in the determining section 22. May be Specifically, if the probability of bleeding is high, it is decided to be a neurosurgery or a radiologist group, and if the probability of infarction is high, it is decided to be a neurologist or a radiologist group. do it.
  • the second database DB2 may be prepared by correlating a neurosurgery and a neurologist or a group of radiologists with each of the hemorrhage and the infarct.
  • the medical image is an image including a brain, but may be an image including a heart and other organs.
  • the analysis unit 21 may specify the ischemic area of the heart.
  • a radiologist or radiologist group can be determined according to at least one of the degree of difficulty and the degree of urgency of diagnosis. For example, when the degree of difficulty of image interpretation is high or when the degree of urgency of image interpretation is high, if the degree of difficulty is high or the degree of urgency is high by determining a skilled image reading doctor or an image reading doctor group In addition, medical images can be read appropriately.

Abstract

Selon la présente invention, une unité d'analyse (21) d'un dispositif de détermination de radiologue (10) analyse une image médicale et acquiert des résultats d'analyse. Sur la base des résultats d'analyse, une unité de détermination (22) détermine un radiologue ou un groupe de radiologues chargés d'interpréter l'image médicale. Par exemple, sur la base des résultats d'analyse, l'unité de détermination (22) acquiert la difficulté d'interprétation et/ou l'urgence d'interprétation de l'image médicale, et sur la base de la difficulté et/ou de l'urgence, détermine un radiologue ou un groupe de radiologues.
PCT/JP2018/042142 2017-11-21 2018-11-14 Dispositif, procédé et programme de détermination de radiologue WO2019102917A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017-223628 2017-11-21
JP2017223628 2017-11-21

Publications (1)

Publication Number Publication Date
WO2019102917A1 true WO2019102917A1 (fr) 2019-05-31

Family

ID=66631887

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/042142 WO2019102917A1 (fr) 2017-11-21 2018-11-14 Dispositif, procédé et programme de détermination de radiologue

Country Status (1)

Country Link
WO (1) WO2019102917A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7370735B2 (ja) 2019-06-06 2023-10-30 キヤノン株式会社 情報処理装置、方法、及びプログラム
JP7465342B2 (ja) 2019-09-27 2024-04-10 ホロジック， インコーポレイテッド ２ｄ／３ｄ乳房画像を精査するための読み取り時間および読み取り複雑性を予測するためのａｉシステム

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005131011A (ja) * 2003-10-29 2005-05-26 Toshiba Corp 急性脳梗塞診断治療支援装置
JP2009157527A (ja) * 2007-12-25 2009-07-16 Canon Inc 医用画像処理装置、医用画像処理方法ならびにプログラム
JP2010165109A (ja) * 2009-01-14 2010-07-29 Toshiba Corp 医用画像診断装置
JP2012088828A (ja) * 2010-10-18 2012-05-10 Sony Corp 情報処理装置及び方法、並びにプログラム
JP2016035739A (ja) * 2014-08-01 2016-03-17 キヤノンマーケティングジャパン株式会社 読影依頼管理システムおよびその制御方法、読影依頼管理装置およびその制御方法、並びにプログラム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005131011A (ja) * 2003-10-29 2005-05-26 Toshiba Corp 急性脳梗塞診断治療支援装置
JP2009157527A (ja) * 2007-12-25 2009-07-16 Canon Inc 医用画像処理装置、医用画像処理方法ならびにプログラム
JP2010165109A (ja) * 2009-01-14 2010-07-29 Toshiba Corp 医用画像診断装置
JP2012088828A (ja) * 2010-10-18 2012-05-10 Sony Corp 情報処理装置及び方法、並びにプログラム
JP2016035739A (ja) * 2014-08-01 2016-03-17 キヤノンマーケティングジャパン株式会社 読影依頼管理システムおよびその制御方法、読影依頼管理装置およびその制御方法、並びにプログラム

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7370735B2 (ja) 2019-06-06 2023-10-30 キヤノン株式会社 情報処理装置、方法、及びプログラム
JP7465342B2 (ja) 2019-09-27 2024-04-10 ホロジック， インコーポレイテッド ２ｄ／３ｄ乳房画像を精査するための読み取り時間および読み取り複雑性を予測するためのａｉシステム

Similar Documents

Publication Publication Date Title
US11139067B2 (en) Medical image display device, method, and program
JP2019153250A (ja) 医療文書作成支援装置、方法およびプログラム
US10950019B2 (en) Automatic layout apparatus, automatic layout method, and automatic layout program
JP7080304B2 (ja) 学習支援装置、学習支援方法、学習支援プログラム、関心領域判別装置、関心領域判別方法、関心領域判別プログラム及び学習済みモデル
US10803986B2 (en) Automatic layout apparatus, automatic layout method, and automatic layout program
WO2020209382A1 (fr) Dispositif de generation de documents médicaux, procédé et programme
JP7237089B2 (ja) 医療文書作成支援装置、方法およびプログラム
JP2020032045A (ja) 梗塞領域を判別する判別器の学習装置、方法およびプログラム、梗塞領域を判別する判別器、並びに梗塞領域判別装置、方法およびプログラム
WO2019102917A1 (fr) Dispositif, procédé et programme de détermination de radiologue
WO2019102903A1 (fr) Dispositif, procédé et programme d'affichage d'informations d'examen
US11551351B2 (en) Priority judgement device, method, and program
JP2019195426A (ja) 医用画像処理装置、方法およびプログラム
JP7007469B2 (ja) 医療文書作成支援装置、方法およびプログラム、学習済みモデル、並びに学習装置、方法およびプログラム
US20220392619A1 (en) Information processing apparatus, method, and program
US11328414B2 (en) Priority judgement device, method, and program
GB2573193A (en) System and method for using imaging quality metric ranking
JPWO2019107134A1 (ja) 検査情報表示装置、方法およびプログラム
JP2019213785A (ja) 医用画像処理装置、方法およびプログラム
JP2008217426A (ja) 症例登録システム
JP7376715B2 (ja) 経過予測装置、経過予測装置の作動方法および経過予測プログラム
US20230070906A1 (en) Information processing apparatus, method, and program
WO2020241857A1 (fr) Dispositif de création de documents médicaux, procédé, et programme, dispositif d'apprentissage, procédé, et programme, et modèle appris
JP6765396B2 (ja) 医用画像処理装置、方法およびプログラム
JP7361930B2 (ja) 医用画像処理装置、方法およびプログラム
US20230225681A1 (en) Image display apparatus, method, and program

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: 18881778

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18881778

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP