JP6251945B1 - Diagnosis support system, medical diagnosis support apparatus, and diagnosis support method - Google Patents

Abstract

The diagnosis support apparatus 1 includes an image transmission unit that transmits a medical image used for determination of a physical state to an image analysis apparatus 2 that analyzes the medical image, a medical image transmitted by the image transmission unit, and an imaging method of the medical image. A storage unit that associates and stores the method information to be shown, and a result receiving unit that receives a plurality of certainty factor calculation results indicating what kind of physical state the medical image indicates from the image analysis device, and Among the plurality of certainty factor calculation results, based on the certainty factor calculation result corresponding to the method information stored in the storage unit in association with the medical image transmitted to the image analysis device, the diagnosis support information related to the physical state indicated by the medical image And an output unit for outputting.

Description

  The present invention relates to a diagnosis support system, a medical diagnosis support apparatus, and a diagnosis support method for determining a body state based on a medical image.

  2. Description of the Related Art Conventionally, a system that assists diagnosis of a body condition by analyzing a medical image using a computer is known. Japanese Patent Application Laid-Open No. 2004-228561 discloses a method for assisting diagnosis of a medical image in a computer that has learned a method for identifying a part included in the medical image using a medical image for learning.

Japanese translation of PCT publication No. 2008-520317

  A medical image analyzed by a computer in a conventional method is an image obtained by photographing an organ from the same direction. For example, by using the method disclosed in Patent Document 1, the position of the tumor can be specified based on the chest image taken from the same direction.

  However, depending on the organ, multiple types of medical images taken from different directions are used. For example, when diagnosis is performed using an MRI (Magnetic Resonance Imaging) image of the heart, various medical images such as a horizontal cross-sectional image and a vertical cross-sectional image are used. In this way, when various medical images are used in a mixed manner, the computer recognizes that the image is an abnormal image although it is a normal image due to different conditions such as the direction in which the image was taken. There was a problem that there was something.

  Accordingly, the present invention has been made in view of these points, and an object thereof is to improve the accuracy of diagnosis support based on medical images by a computer.

  A diagnosis support system according to a first aspect of the present invention is a diagnosis support system including a diagnosis support device that supports diagnosis of a physical condition indicated by a medical image, and an image analysis device that analyzes the medical image. The image analysis device is configured to calculate a certainty factor of each of a plurality of physical states indicated by the medical image by analyzing the medical image and an image receiving unit that receives the medical image from the diagnosis support device And a result transmitting unit that transmits a plurality of certainty factor calculation results indicating a plurality of certainty factors corresponding to the plurality of physical states calculated by the calculating unit to the diagnosis support apparatus. Further, the diagnosis support device includes an image transmission unit that transmits the medical image to be analyzed by the image analysis device to the image analysis device, the medical image transmitted by the image transmission unit, and the medical image A storage unit that associates and stores method information indicating an imaging method, a result receiving unit that receives the plurality of certainty factor calculation results transmitted by the image analysis device, and the image among the plurality of certainty factor calculation results An output unit that outputs diagnosis support information related to a physical condition indicated by the medical image based on a certainty factor calculation result corresponding to the method information stored in the storage unit in association with the medical image transmitted to the analysis device; Have

  The result receiving unit receives the plurality of certainty factor calculation results in which the certainty factor information indicating the certainty factor indicating the normal state or the certainty factor indicating the abnormal state is associated with the method information. And the output unit corresponds to the same method information corresponding to the same method information as the method information stored in the storage unit in association with the medical image transmitted by the image transmission unit among the plurality of certainty factor calculation results. The diagnosis support information may be output based on the degree calculation result.

  When there are a plurality of certainty factor calculation results corresponding to the method information associated with the medical image transmitted by the image transmitting unit, the output unit indicates the certainty factor information among the plurality of certainty factor calculation results. The diagnosis support information including one certainty factor calculation result having the highest certainty factor may be output.

  The image transmitting unit transmits a medical image group including a plurality of the medical images, the result receiving unit receives the plurality of certainty factor calculation results in association with each of the plurality of medical images, and the output unit After selecting one image-specific reliability calculation result from the plurality of reliability calculation results for each of the plurality of medical images, the plurality of image-specific beliefs associated with each of the plurality of medical images The diagnosis support information corresponding to the medical image group may be output based on the degree calculation result.

  The output unit selects based on the time when each of the plurality of medical images is generated when different certainty factor calculation results have been obtained for the plurality of medical images included in the same medical image group The certainty factor calculation result corresponding to the one or more medical images may be selected.

  The image transmitting unit transmits a medical image group including a plurality of the medical images, and the result receiving unit includes a plurality of image certainty factor calculation results associated with each of the plurality of medical images. Receiving a plurality of certainty factor calculation results, the output unit from the plurality of certainty factor calculation results corresponding to the same method information, a plurality of images corresponding to the plurality of medical images in each of the certainty factor calculation results The maximum value of the certainty factor indicated by another certainty factor calculation result and the sum total of the certainty factor calculation result corresponding to the plurality of medical images in the respective certainty factor calculation results is the largest in the certainty factor calculation result The diagnosis support information corresponding to the medical image group may be output based on the ratio of the sum in other certainty factor calculation results.

  The image transmission unit is configured to acquire a learning image, which is the medical image to be learned by the image analysis apparatus, the method information indicating an imaging method of the learning image, and diagnosis support information indicating a physical state corresponding to the learning image. And the calculation unit is configured to transmit the diagnostic support information indicated by the learning image to the degree of certainty of the body state indicated by the medical image to be analyzed received by the image receiving unit. You may determine based on the result learned based on.

  The output unit, based on the method information corresponding to the medical image transmitted by the image transmission unit to the image analysis device, from a plurality of results included in the certainty factor calculation result received by the result reception unit, You may remove the result corresponding to the system information linked | related with the medical image which the image transmission part transmitted.

  Further, the calculation unit acquires a plurality of learning medical images in which method information indicating an imaging method of the medical image and state information indicating whether the medical image is normal or abnormal are associated, and the imaging method The difference between the normal pattern and the abnormal pattern of the medical image may be learned for each type.

  The medical diagnosis support apparatus according to the second aspect of the present invention includes an image transmission unit that transmits a medical image used for diagnosis of a physical condition to an image analysis device that analyzes the medical image, and the image transmission unit that transmits the medical image. A storage unit that associates and stores a medical image and method information indicating an imaging method of the medical image, and indicates from the image analysis device what kind of body state the medical image indicates is highly reliable A result receiving unit that receives a plurality of certainty factor calculation results, and among the plurality of certainty factor calculation results, corresponds to the method information stored in the storage unit in association with the medical image transmitted to the image analysis device An output unit that outputs diagnosis support information related to the physical state indicated by the medical image based on the certainty factor calculation result.

  A diagnosis support method according to a third aspect of the present invention is a diagnosis support method for supporting diagnosis of a physical condition indicated by a medical image, and is an object to be analyzed by an image analysis apparatus that analyzes the medical image. Transmitting a medical image to the image analysis device; receiving from the image analysis device a plurality of certainty factor calculation results indicating a high certainty factor indicating the physical state of the medical image; and Based on the certainty factor calculation result corresponding to the method information indicating the imaging method of the medical image transmitted to the image analysis device among the plurality of certainty factor calculation results, the diagnostic support information regarding the physical condition indicated by the medical image And a step of outputting.

  According to the present invention, there is an effect that accuracy of diagnosis support based on a medical image by a computer is improved.

1 is a diagram illustrating a configuration of a diagnosis support system S. FIG. It is a figure for demonstrating learning in the image analysis apparatus. It is a figure for demonstrating the method of outputting the reliability calculation result with respect to a medical image after learning of the image analysis apparatus. 1 is a diagram illustrating a configuration of a diagnosis support apparatus 1. FIG. 2 is a diagram illustrating a configuration of an image analysis device 2. FIG. It is a figure for demonstrating the method the diagnosis assistance apparatus 1 specifies the physical state corresponding to a medical image group. It is a figure for demonstrating the other method in which the output part 143 outputs diagnostic assistance information.

[Overview of this embodiment]
FIG. 1 is a diagram illustrating a configuration of a diagnosis support system S according to the present embodiment. The image determination system S includes a diagnosis support apparatus 1 and an image analysis apparatus 2.

  The diagnosis support apparatus 1 acquires a medical image output from a medical device such as an MRI apparatus, a CT (Computer Tomography) apparatus, an X-ray apparatus, or an ultrasonic inspection apparatus, and the acquired medical image has an abnormality in the body. Assists in diagnosing whether the condition is indicated. The diagnosis support apparatus 1 is a server, for example, and transmits a medical image acquired from a medical device to the image analysis apparatus 2. The diagnosis support apparatus 1 receives, from the image analysis apparatus 2, a certainty factor calculation result that is estimated based on the transmitted medical image and indicates a high probability that the body is in a certain state. The diagnosis support apparatus 1 determines whether or not there is a high possibility that an abnormality has occurred in the body based on the received calculation result of the certainty factor.

  The image analysis apparatus 2 is a computer capable of deep learning using, for example, artificial intelligence software. The image analysis device 2 learns the relationship between the content of the medical image and the presence / absence of a physical abnormality using a large number of medical images for learning, and the medical image received from the diagnosis support device 1 based on the learned result Calculate the certainty that the body is in an abnormal state and the certainty that the body is in a normal state. The image analysis device 2 notifies the diagnosis support device 1 of a certainty factor calculation result including the calculated certainty factor.

  The image determination system S is characterized in that various types of medical images can be used together. At the time of learning, the image analysis apparatus 2 acquires a medical image in which method information indicating a medical image imaging method and state information indicating whether the medical image is normal or abnormal are associated with each other, and normalization of various types of medical images is performed. Learn the difference between patterns and abnormal patterns.

  The method information includes information for specifying the type of the imaging method, such as the imaging direction (horizontal sectional direction or vertical sectional direction) of the organ included in the medical image and the imaging timing (during rotation or stopping). include. After performing learning, the image analysis device 2 is convinced that the medical image received from the diagnosis support device 1 is a normality pattern of each type of medical image and an abnormal pattern of each type of medical image. Calculate the degree.

  FIG. 2 is a diagram for explaining learning in the image analysis apparatus 2. When the image analysis device 2 learns, a medical image in which the type of medical image and state information indicating normality / abnormality are associated is input to the image analysis device 2. In the example of FIG. 2, three types of information “a”, “b”, and “c” are shown as the method information indicating the imaging method of the medical image. Further, N is associated with the medical image indicating the normal state, and E1 or E2 is associated with the medical image indicating the abnormal state. E1 and E2 correspond to different abnormal states, respectively.

  The image analysis apparatus 2 recognizes what characteristics each combination of the method information and the state information of the medical image has based on a large number of medical images, and maps the relationship between the combination and the characteristics. The image analysis apparatus 2 stores, in a storage medium, learning results obtained by mapping the relationship between combinations and features for various features.

  FIG. 3 is a diagram for explaining a method of outputting a certainty factor calculation result for a medical image after learning by the image analysis apparatus 2. The image analysis device 2 analyzes the characteristics of the medical image acquired after learning, and calculates the certainty factor that the characteristics of the medical image belong to each combination. The image analysis device 2 calculates a certainty factor belonging to each combination for each medical image, and calculates a certainty factor that associates information for specifying the medical image (for example, a medical image ID) and a certainty factor belonging to each combination. Output the result.

  In the example shown in FIG. 3, the certainty factor calculation result output from the image analysis apparatus 2 is 0.9 when the combination of the state information and the method information is Na, and the certainty factor when E1a is 0.4. It is shown that the certainty factor of E2b is 0.3. In the following description, a numerical value indicating the certainty level may be referred to as certainty factor information. When the diagnosis support apparatus 1 receives such a certainty factor calculation result, the diagnosis support apparatus 1 transmits it from a plurality of results included in the certainty factor calculation result based on the method information corresponding to the medical image transmitted to the image analysis apparatus 2. Results corresponding to method information different from medical information may be removed.

  For example, when the method information of the medical image transmitted by the diagnosis support apparatus 1 is a, the diagnosis support apparatus 1 removes the result E2b included in the certainty calculation result. In the diagnosis support apparatus 1, the state information indicated by the combination (Na in FIG. 3) corresponding to the result with the highest certainty among the remaining results indicates the physical state corresponding to the medical image transmitted to the image analysis apparatus 2. The diagnosis support information indicating the determined physical condition is output. In this way, the diagnosis support apparatus 1 can output diagnosis support information indicating the state of the organ indicated by the medical image without notifying the image analysis apparatus 2 of the method information of the medical image. For example, the diagnosis support apparatus 1 associates the medical image with the diagnosis support information and transmits it to the doctor's computer.

The numerical value indicated by the certainty factor calculation result does not need to be a numerical value indicating the certainty factor itself, and may be a numerical value indicating a relative certainty among a plurality of combinations. As shown in FIG. 3, if the certainty factor that is Na is 0.9, the certainty factor that is E1a is 0.4, and the certainty factor that is Eb is 0.3, it is diagnosed. The support device 1 is normalized so that the sum of the certainty factors of all the combinations becomes 1.0, the certainty factor of Na is 56%, the certainty factor of E1a is 25%, and the certainty factor of E2b is 19%. %.
Hereinafter, configurations and operations of the diagnosis support apparatus 1 and the image analysis apparatus 2 will be described in detail.

[Configuration of Diagnosis Support Device 1]
FIG. 4 is a diagram illustrating a configuration of the diagnosis support apparatus 1. The diagnosis support apparatus 1 includes an image acquisition unit 11, a communication unit 12, a storage unit 13, and a control unit 14. The image acquisition unit 11 is an interface for acquiring a medical image from an external device such as a medical device. The image acquisition unit 11 is an interface corresponding to, for example, DICOM (Digital Imaging and Communication in Medicine) standard.

  The communication unit 12 is an interface for transmitting and receiving data to and from the image analysis apparatus 2 and includes, for example, a LAN (Local Area Network) controller. The communication unit 12 transmits a plurality of medical images to the image analysis device 2 via a network such as a LAN or the Internet, and receives a plurality of certainty factor calculation results from the image analysis device 2.

  The storage unit 13 includes storage media such as a ROM (Read Only Memory), a RAM (Random Access Memory), and a hard disk. The storage unit 13 stores a program executed by the control unit 14. In addition, the storage unit 13 temporarily stores a plurality of medical images received via the image acquisition unit 11 and a plurality of certainty factor calculation results received from the image analysis device 2 via the communication unit 12. The storage unit 13 stores, for example, a plurality of medical images transmitted to the image analysis device 2 and the method information of each medical image in association with each other.

  The control unit 14 is, for example, a CPU (Central Processing Unit). The control unit 14 functions as an image transmission unit 141, a result reception unit 142, and an output unit 143 by executing a program stored in the storage unit 13.

  The image transmission unit 141 transmits a medical image to be analyzed by the image analysis device 2 to the image analysis device 2. The image transmission unit 141 removes identification information that can identify a person corresponding to the medical image from the medical image acquired by the image acquisition unit 11, and communicates a plurality of medical images after the identification information is removed. The image is transmitted to the image analysis apparatus 2 via the unit 12. The image transmission unit 141 corresponds to the learning image, which is a medical image for the image analysis device 2 to learn when the image analysis device 2 learns, and the learning information. The image may be transmitted to the image analysis apparatus 2 in association with the diagnosis support information indicating the physical condition to be performed.

  The result receiving unit 142 receives a plurality of certainty factor calculation results transmitted by the image analysis device 2 via the communication unit 12. For example, as described with reference to FIG. 3, the plurality of certainty factor calculation results are certainty factor calculation results corresponding to each combination of a plurality of method information and a plurality of state information. The result receiving unit 142 receives a plurality of certainty factor calculation results in which the certainty factor information indicating the certainty factor indicating the normal state or the certainty factor indicating the abnormal state and the method information are associated with each other. The result receiving unit 142 inputs a plurality of received certainty factor calculation results to the output unit 143.

  The output unit 143 outputs a certainty factor calculation result corresponding to the method information stored in the storage unit 13 in association with the medical image transmitted to the image analysis device 2 among the plurality of certainty factor calculation results received by the result receiving unit 142. Based on this, the diagnosis support information related to the physical condition indicated by the medical image is output. Specifically, the output unit 143 has the same method information as the method information stored in the storage unit 13 in association with the medical image transmitted from the image transmission unit 141 to the image analysis device 2 among the plurality of certainty factor calculation results. The diagnosis support information is output based on the certainty factor calculation result corresponding to. For example, in the case of the example illustrated in FIG. 3, the output unit 143 corresponds to “a” that is the method information of the medical image transmitted to the image analysis device 2 among the plurality of certainty factor calculation results corresponding to the method information “a” and “b”. Diagnosis support information corresponding to the certainty calculation result is output.

When there are a plurality of certainty factor calculation results corresponding to the method information associated with the medical image transmitted from the image transmitting unit 141 to the image analysis device 2, the output unit 143 includes the certainty factor information among the plurality of certainty factor calculation results. One confidence level calculation result with the highest certainty level is output. For example, in the example illustrated in FIG. 3, Na and E1a are transmitted from the image analysis device 2 as the certainty calculation result corresponding to the method information a associated with the medical image transmitted to the image analysis device 2. . Therefore, the output unit 143 selects Na having higher certainty than E1a among Na and E1a, and outputs diagnosis support information indicating that the medical image indicates a normal state. When the output unit 143 receives the certainty factor calculation result, the image transmitting unit 141 transmits the plurality of results included in the certainty factor calculation result based on the method information corresponding to the medical image transmitted to the image analysis device 2. The result corresponding to the method information different from the method information associated with the medical image may be removed.

[Configuration of Image Analysis Device 2]
FIG. 5 is a diagram illustrating a configuration of the image analysis apparatus 2. The image analysis device 2 includes an image reception unit 21, a calculation unit 22, a storage unit 23, and a result transmission unit 24.

  The image receiving unit 21 receives a medical image from the diagnosis support apparatus 1. The calculation unit 22 analyzes the medical image to calculate what kind of physical state the medical image indicates is high. The calculation unit 22 is based on the result of learning the certainty factor corresponding to each of a plurality of physical states indicated by the medical image to be analyzed received by the image receiving unit 21 using the diagnosis support information indicated by the learning image. To calculate. The calculation unit 22 is, for example, a CPU, and executes a calculation process by executing a program stored in the storage unit 23.

  In the learning process, the calculation unit 22 acquires a plurality of learning medical images in which method information indicating a medical image capturing method and state information indicating whether the medical image is normal or abnormal are associated with each other. The calculation unit 22 creates a learning model by learning the difference between the normal pattern and the abnormal pattern of the medical image for each type of imaging method, and stores the created learning model in the storage unit 23. When the image reception unit 21 receives a medical image, the calculation unit 22 refers to the learning model stored in the storage unit 23 and uses the medical image and method information received by the image reception unit 21 as inputs to the learning model. Thus, the certainty factor is calculated.

  The storage unit 23 includes storage media such as a ROM, a RAM, and a hard disk. The storage unit 23 stores a program executed by the calculation unit 22. The result transmission unit 24 transmits a plurality of certainty factor calculation results generated by the calculation unit 22 to the diagnosis support apparatus 1.

(Modification)
In the above description, the case where the diagnosis support apparatus 1 transmits one medical image to the image analysis apparatus 2 has been described. However, the diagnosis support apparatus 1 converts a medical image group including a plurality of medical images into the image analysis apparatus 2. A plurality of certainty factor calculation results corresponding to each of the transmitted plurality of medical images may be received from the image analysis device 2. In the medical image group, for example, a plurality of medical image files are converted into a single file in the archive format.

Specifically, the diagnosis support apparatus 1 and the image analysis apparatus 2 operate as follows.
The image transmission unit 141 transmits a medical image group including a plurality of medical images to the image analysis device 2. For example, the image transmission unit 141 transmits the MRI image group of the heart acquired over a predetermined time (for example, 30 seconds) to the image analysis apparatus 2. In the medical image group, identification information for specifying each medical image included in the medical image group is associated with the medical image.

  When the image analysis device 2 receives the medical image group, the calculation unit 22 generates a certainty factor calculation result for each medical image included in the medical image group. The result transmission unit 24 transmits a plurality of certainty factor calculation results to the diagnosis support apparatus 1 in association with the medical image. For example, the result transmission unit 24 transmits the medical image identification information and the certainty factor calculation result in association with each other.

The result receiving unit 142 receives a plurality of certainty factor calculation results in association with each of the plurality of medical images, and stores the received plurality of certainty factor calculation results in the storage unit 13 in association with the identification information of the medical image.
The output unit 143 selects, for example, one image certainty factor calculation result from a plurality of certainty factor calculation results for each of a plurality of medical images, and then selects a plurality of image items associated with each of the plurality of medical images. Based on the certainty calculation result, diagnosis support information corresponding to the medical image group is output.

  FIG. 6 is a diagram for explaining a method by which the diagnosis support apparatus 1 identifies a physical state corresponding to a medical image group. The diagnosis support apparatus 1 transmits a medical image group including five medical images as shown in FIG. 6 to the image analysis apparatus 2, and the result receiving unit 142 receives the certainty calculation result corresponding to each medical image. And In such a case, first, the output unit 143 selects a certainty factor calculation result corresponding to the method information of the medical image stored in the storage unit 13 among the certainty factor calculation results corresponding to the respective medical images. For example, in the example illustrated in FIG. 6, the output unit 143 selects Na and E1a from the certainty factor calculation results corresponding to the medical image 1.

  Subsequently, when there are a plurality of selected certainty factor calculation results, the output unit 143 selects a certainty factor calculation result indicating the highest certainty factor among the plurality of certainty factor calculation results. In the case of the medical image 1 in FIG. 6, since the certainty factor of Na is 0.7 and the certainty factor of E1a is 0.6, the output unit 143 selects Na having the highest certainty factor. Then, since the certainty factor calculation result is Na, the output unit 143 outputs diagnosis support information indicating that the body condition is normal based on the medical image 1.

  The output unit 143 selects the certainty factor calculation result Na for the medical images 2 to 4 by the same procedure. On the other hand, the output unit 143 selects the certainty factor calculation result E1a for the medical image 5. As described above, when different certainty factor calculation results are obtained in a plurality of medical images included in the same medical image group, the output unit 143 displays the certainty factor calculation results corresponding to the most medical images. The certainty factor calculation result corresponding to the medical image group is determined. By doing in this way, even if it is difficult to determine the physical state based on some medical images, the output unit 143 can specify the physical state with high accuracy.

  The output unit 143 selects one based on the time when each of the plurality of medical images is generated when different certainty factor calculation results are obtained in the plurality of medical images included in the same medical image group. You may select the certainty calculation result corresponding to the above medical image. The medical image generated in the first predetermined period in the medical image group and the medical image generated in the last predetermined period tend to be less reliable than the other medical images. Therefore, the output unit 143 generates, for example, a medical image included in a predetermined period after the time when the medical image is first generated, and finally a medical image among a plurality of medical images included in the medical image group. The body state is specified using the certainty factor calculation result of the medical image other than the medical image included in the predetermined period before the determined time.

  Further, the output unit 143 is based on the certainty factor calculation result having the largest sum of certainty factors indicated by the certainty factor calculation results corresponding to the plurality of medical images among the plural certainty factor calculation results corresponding to the same method information. The diagnosis support information corresponding to the medical image group may be output. For example, in the case of the example shown in FIG. 6, the total sum of the certainty factor calculation results Na corresponding to each medical image is 0.7 + 0.9 + 0.9 + 0.8 + 0.5 = 3.8, and the total sum of the certainty factor calculation results E1a. Is 0.6 + 0.4 + 0.3 + 0.5 + 0.6 = 2.4, and the sum of the certainty calculation results E2a is 0.3. In this case, the output unit 143 selects Na as the probability calculation result of the medical image group, and outputs diagnosis support information indicating that the physical state indicated by the medical image group is normal.

  Further, the output unit 143 determines the maximum certainty factor indicated by a plurality of image certainty factor calculation results corresponding to a plurality of medical images in each certainty factor calculation result from a plurality of certainty factor calculation results corresponding to the same method information. Value and the ratio (contribution) of the sum in the other belief calculation results to the sum in the belief calculation result with the largest sum of the belief calculation results for each image corresponding to a plurality of medical images in each belief calculation result Based on the above, diagnosis support information corresponding to the medical image group may be output.

  FIG. 7 is a diagram for explaining the above method in which the output unit 143 outputs diagnosis support information. First, the output unit 143 calculates the sum of a plurality of image-specific certainty calculation results corresponding to a plurality of medical images with respect to each reliability calculation result of Na, E1a, E2b, and the like. For example, in the example shown in FIG. 7A, in the case of Na, the sum is 0.7 + 0.9 + 0.9 = 2.5, and in the case of E1a, the sum is 0.6 + 0.4 + 0.3 = 1.3, E2b. In this case, the sum is 0.3 + 0.3 + 0.4 = 1.0. In this case, the reliability calculation result with the largest sum is Na, and the sum is 2.5. In this case, the ratio (contribution) of the sum of E1a to the sum of Na is 0.5, and the ratio (contribution) of E2b to the sum of Na is 0.4.

  Subsequently, the output unit 143 selects the maximum value of the certainty factor indicated by the plurality of image certainty factor calculation results corresponding to the plurality of medical images in the respective certainty factor calculation results. The maximum value of the certainty factor in the certainty factor calculation result Na is 0.9, the maximum value of the certainty factor in the certainty factor calculation result E1a is 0.6, and the maximum value of the certainty factor in the certainty factor calculation result E2b is 0.4.

  Subsequently, as illustrated in FIG. 7B, the output unit 143 calculates a determination value by multiplying the maximum value of the certainty factor selected for each certainty factor calculation result by the contribution factor. In the example shown in FIG. 7B, the determination value of Na is 0.9 × 1.0 = 0.90, the determination value of E1a is 0.6 × 0.5 = 0.30, and the determination value of E2b. The value is 0.4 × 0.4 = 0.16. The output unit 143 specifies that the physical condition corresponding to the certainty factor calculation result having the largest determination value is the physical condition indicated by the medical image, and outputs diagnosis support information indicating that.

  Since the method described with reference to FIG. 7 described above is used, the high reliability of each confidence factor calculation result is taken into account by using the sum of a plurality of image confidence factor calculation results corresponding to a plurality of medical images. The determination accuracy can be further improved.

[Effects of diagnosis support system S]
As described above, in the diagnosis support system S according to the present embodiment, the diagnosis support apparatus 1 transmits a medical image used for the determination of the body state to the image analysis apparatus 2, and the diagnosis support apparatus 1 includes the image analysis apparatus. Among the plurality of certainty factor calculation results received from 2, the body indicated by the medical image based on the certainty factor calculation result corresponding to the method information stored in the storage unit 13 in association with the medical image transmitted to the image analysis device 2 Outputs diagnostic support information about the condition. In this way, the diagnosis support apparatus 1 uses the artificial intelligence of the image analysis apparatus 2 to determine the physical state based on the medical image without transmitting method information to the image analysis apparatus 2 in association with the medical image. It can be determined with high accuracy.

As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.
For example, in the above description, an example in which a captured image output from an MRI apparatus is used as a medical image has been described. However, the medical image type is not limited to an MRI image, and is output from a medical device other than the MRI apparatus. Any image may be used. The medical image may be an image of a part other than the heart.

DESCRIPTION OF SYMBOLS 1 Diagnosis support apparatus 2 Image analysis apparatus 11 Image acquisition part 12 Communication part 13 Storage part 14 Control part 21 Image reception part 22 Calculation part 23 Storage part 24 Result transmission part 141 Image transmission part 142 Result reception part 143 Output part

Claims (11)

A diagnosis support system comprising: a diagnosis support device that supports diagnosis of a physical condition indicated by a medical image; and an image analysis device that analyzes the medical image,
The image analysis device includes:
An image receiving unit for receiving the medical image from the diagnosis support apparatus;
By calculating the certainty factor of each of a plurality of physical states indicated by the medical image by analyzing the medical image;
A result transmitting unit that transmits a plurality of certainty factor calculation results indicating a plurality of certainty factors corresponding to the plurality of physical states calculated by the calculating unit;
Have
The diagnosis support apparatus includes:
An image transmission unit that transmits the medical image to be analyzed by the image analysis device to the image analysis device;
A storage unit that associates and stores the medical image transmitted by the image transmission unit and method information indicating an imaging method of the medical image;
A result receiving unit for receiving the plurality of certainty factor calculation results transmitted by the image analysis device;
Of the plurality of certainty factor calculation results, the body indicated by the medical image is based on the certainty factor calculation result corresponding to the method information stored in the storage unit in association with the medical image transmitted to the image analysis device. An output unit for outputting diagnosis support information about the state;
A diagnostic support system. The result receiving unit receives the plurality of certainty factor calculation results in which the certainty factor information indicating the certainty factor indicating the normal state or the certainty factor indicating the abnormal state is associated with the method information. And
The output unit calculates the certainty factor corresponding to the same method information as the method information stored in the storage unit in association with the medical image transmitted by the image transmission unit among the plurality of certainty factor calculation results. Outputting the diagnosis support information based on a result;
The diagnosis support system according to claim 1. When there are a plurality of certainty factor calculation results corresponding to the method information associated with the medical image transmitted by the image transmitting unit, the output unit indicates the certainty factor information among the plurality of certainty factor calculation results. Outputting the diagnosis support information including one certainty factor calculation result having the highest certainty factor,
The diagnosis support system according to claim 2. The image transmission unit transmits a medical image group including a plurality of the medical images,
The result receiving unit receives the plurality of certainty factor calculation results in association with each of the plurality of medical images,
The output unit selects one image certainty factor calculation result from the plurality of certainty factor calculation results for each of the plurality of medical images, and then selects a plurality of the plurality of the medical images associated with each of the plurality of medical images. Outputting the diagnosis support information corresponding to the group of medical images based on the certainty calculation result for each image,
The diagnosis support system according to any one of claims 1 to 3. The output unit selects based on the time when each of the plurality of medical images is generated when different certainty factor calculation results have been obtained for the plurality of medical images included in the same medical image group Selecting the certainty calculation result corresponding to the one or more medical images,
The diagnosis support system according to claim 4. The image transmission unit transmits a medical image group including a plurality of the medical images,
The result receiving unit receives the plurality of certainty factor calculation results each including a plurality of image certainty factor calculation results associated with each of the plurality of medical images;
The output unit has the certainty factor indicated by the plurality of certainty factor calculation results corresponding to the plurality of medical images in the certainty factor calculation result from the plurality of certainty factor calculation results corresponding to the same method information. And the ratio of the sum in the other belief calculation results to the sum in the belief calculation result having the largest sum of the image belief calculation results corresponding to the plurality of medical images in the respective belief calculation results And outputting the diagnosis support information corresponding to the medical image group, based on
The diagnosis support system according to any one of claims 1 to 3. The image transmission unit is configured to acquire a learning image, which is the medical image to be learned by the image analysis apparatus, the method information indicating an imaging method of the learning image, and diagnosis support information indicating a physical state corresponding to the learning image. And send it to the image analysis device in association with
The calculation unit determines a certainty factor of a physical state indicated by the medical image to be analyzed received by the image receiving unit based on a learning result based on the diagnosis support information indicated by the learning image.
The diagnosis support system according to any one of claims 1 to 6. The output unit, based on the method information corresponding to the medical image transmitted by the image transmission unit to the image analysis device, from a plurality of results included in the certainty factor calculation result received by the result reception unit, Removing results corresponding to method information different from the method information associated with the medical image transmitted by the image transmission unit;
The diagnosis support system according to any one of claims 1 to 7. The calculation unit acquires a plurality of learning medical images in which method information indicating an imaging method of the medical image and state information indicating whether the medical image is normal or abnormal are associated, and the type of the imaging method Learn the difference between normal and abnormal patterns of medical images for each
The diagnosis support system according to any one of claims 1 to 8. An image transmission unit that transmits a medical image used for diagnosis of a physical condition to an image analysis device that analyzes the medical image;
A storage unit that associates and stores the medical image transmitted by the image transmission unit and method information indicating an imaging method of the medical image;
A result receiving unit that receives a plurality of certainty factor calculation results indicating what kind of physical state the medical image indicates from the image analysis device;
Of the plurality of certainty factor calculation results, the body indicated by the medical image is based on the certainty factor calculation result corresponding to the method information stored in the storage unit in association with the medical image transmitted to the image analysis device. An output unit for outputting diagnosis support information about the state;
A diagnosis support apparatus having A diagnosis support method for supporting diagnosis of a physical condition indicated by a medical image,
Transmitting the medical image to be analyzed by an image analysis device that analyzes the medical image to the image analysis device;
Receiving, from the image analysis device, a plurality of certainty factor calculation results indicating a high certainty factor indicating the physical state of the medical image;
Based on the certainty factor calculation result corresponding to the method information indicating the imaging method of the medical image transmitted to the image analysis device among the plurality of certainty factor calculation results, diagnosis support information regarding the physical condition indicated by the medical image is obtained. Output step;
A diagnostic support method comprising:

Images