WO2019221586A1

WO2019221586A1 - Medical image management system, method, and computer-readable recording medium

Info

Publication number: WO2019221586A1
Application number: PCT/KR2019/095016
Authority: WO
Inventors: 심우현; 박범우; 박채리
Original assignee: 재단법인 아산사회복지재단; 울산대학교 산학협력단
Priority date: 2018-05-16
Filing date: 2019-05-16
Publication date: 2019-11-21
Also published as: KR102034648B1

Abstract

Disclosed are a medical image management system, method, and computer readable recording medium. The medical image management system according to one embodiment of the present invention comprises: a database containing information on the characteristics of a medical image and the characteristics of artifacts contained in the medical image; an image receiving unit for receiving a medical image; an image classification unit for determining the type of the received medical image by using the information on the characteristics of the medical image stored in the database; and a quality determination unit for determining the quality of the received medical image by using information on the characteristics of the artifacts stored in the database.

Description

Medical Image Management Systems, Methods, and Computer-readable Records

The present invention relates to a medical image management system, a method and a computer readable recording medium, and more particularly, to a medical image management system, a method and a computer readable recording medium capable of quickly and accurately determining the type and quality of a medical image. It is about.

Different types of medical images may be displayed, and the same substance or organ may be displayed with different brightness according to the type of medical image, and noise may be included in the medical image as an unpredictable situation occurs while taking the medical image.

For accurate diagnosis, it is necessary to basically classify the types of medical images. Depending on the degree of noise, there is a medical image that cannot be used for diagnosis.

However, since these processes are performed by the human eye, it may cause a problem that excessive time is taken or accurate judgment is not made.

An object of the present invention is to provide a medical image management system, a method and a computer-readable recording medium capable of quickly and accurately determining the type and quality of medical images.

Medical image management system according to an embodiment of the present invention, a database containing information on the characteristics of the medical image and the characteristics of the artifacts (artifacts) in the medical image, an image receiving unit for receiving a medical image, stored in the database An image classifier that determines the type of the medical image received by using the information about the characteristics of the medical image and a quality that determines the quality of the received medical image by using information about the characteristics of the noise stored in the database. It includes a determination unit.

The database may include information about a result of learning brightness contrast according to a plurality of medical image capturing techniques and a result of learning whether noise is included in the medical image and characteristics of the noise.

The database may further include information about a result of learning header information included in the plurality of medical images, and the learning of the brightness contrast may be performed by the normal regions of the plurality of medical images. It may be to learn brightness contrast.

The image classifying unit may determine the type of the medical image based on the brightness contrast degree of the received medical image.

In addition, except for a facial region separating unit for separating the facial region in the medical image, a request receiving unit for receiving a data providing request for at least one of the eyes, nose, mouth included in the facial region from the client and the requested portion The apparatus may further include a non-identification unit performing de-identification on the facial area.

The database may further include information on eye, nose and mouth learning results included in a medical image, and the face area separator may separate the face area using the eye, nose and mouth learning results.

In addition, the de-identification unit may perform irreversible data conversion for the eyes, nose, and mouth.

On the other hand, the medical image management method according to an embodiment of the present invention, the step of constructing a database including information on the characteristics of the medical image and the characteristics of the artifacts (artifacts) included in the medical image, receiving the medical image Determining a type of the received medical image by using information about the characteristic of the medical image stored in the database, and determining the quality of the received medical image by using information about the characteristic of the noise stored in the database. Determining.

In addition, the building of the database may include learning brightness contrast according to a plurality of medical image capturing techniques, and learning whether noise is included in the medical image and characteristics of the noise.

The establishing of the database may further include learning header information included in the plurality of medical images, and in the learning of the brightness contrast, the plurality of medical images. You can learn the brightness contrast of the normal areas in.

In the determining of the type of the medical image, the type of the medical image may be determined based on the brightness contrast degree of the received medical image.

The method may further include separating a facial area from the received medical image, receiving a request for providing a data for at least one of eyes, nose, and mouth included in the facial area from a client, and remaining facial areas except for the requested area. The method may further include performing de-identification on.

The database may further include information about eye, nose, and mouth learning results included in a medical image, and in the step of separating the face area, the face area using machine learning results for the eyes, nose, and mouth. Can be separated.

In the performing of the de-identification, irreversible data transformation may be performed on the eyes, nose, and mouth.

Meanwhile, a computer-readable recording medium having a program recorded thereon for performing the medical image management method according to the present invention may be provided.

The present invention can provide a medical image management system, a method and a computer-readable recording medium that can quickly and accurately determine the type and quality of medical images.

1 is a view schematically showing the configuration of a medical image management system according to an embodiment of the present invention.

2 and 3 are views illustrating differences in contrast of medical images according to differences in imaging techniques.

4 is a diagram exemplarily illustrating an artifact existing in a medical image.

5 is a view schematically showing the configuration of a medical image management system according to another embodiment of the present invention.

6 is a diagram exemplarily illustrating a result of medical image de-identification.

7 is a diagram schematically illustrating an embodiment of a de-identification method.

8 is a flowchart schematically illustrating a medical image management method according to an exemplary embodiment of the present invention.

9 is a flowchart schematically illustrating a database building step according to an embodiment of the present invention.

10 is a flowchart schematically illustrating a medical image management method according to another exemplary embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving them will be apparent with reference to the embodiments described in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments set forth below, but may be embodied in many different forms and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. do. The embodiments set forth below are provided to make the disclosure of the present invention complete, and to fully inform the scope of the invention to those skilled in the art. In the following description of the present invention, when it is determined that the detailed description of the related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprises" or "having" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof. Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

Referring to FIG. 1, the medical image management system 100 according to an exemplary embodiment of the present invention includes a database 110, an image receiving unit 120, an image classifying unit 130, and a quality determining unit 140. . The database 110 includes information about the characteristics of the medical image and the characteristics of artifacts included in the medical image. The medical image may be an image such as CT (Computed Tomography), MRI (Magnetic Resonance Imaging), or a plurality of 2D images or 3D images.

The information about the characteristics of the medical image may include information about a method of obtaining an image such as whether the medical image is an MRI image, which is a CT image, and information about a situation or condition in which the medical image is obtained.

On the other hand, the information on the characteristics of the medical image stored in the database 110 may include a result of learning the brightness contrast according to the imaging technique for a plurality of medical images, the medical image based on the brightness contrast The type of can be determined.

In one embodiment, the learning of the brightness contrast may be learning the brightness contrast of the normal areas in the plurality of medical images. The normal area means an area where no disease or abnormality exists, and sufficient information suitable for determining the type of medical image may be obtained by learning brightness contrast of the normal areas.

The brightness contrast degree means a difference in brightness for each image type for blood, bone, fat, and edema, and the type of medical image is determined by determining the brightness of the blood, bone, fat, and edema included in the medical image. Can be determined.

The artifacts refer to noise that may be included in the medical image. If the noise exists in the medical image, the diagnosis may not be accurately made through the medical image, and thus the medical image including the noise may not be used. You may not. Therefore, the information about the characteristics of the noise may be an important indicator for determining the quality of the medical image, and the database 110 may include a result of learning whether the noise is included in the medical image and the characteristics of the noise. have.

The image receiver 120 receives the medical image, and the type and quality may be determined based on the information stored in the database 110 for the medical image. The image receiver 120 may receive the medical image from the database 110 or may directly receive the medical image from a user.

In general, when the medical image is acquired, it may be desirable to determine the type and quality of the medical image before the medical image is acquired. Therefore, when the medical image is generated by the medical imaging apparatus, the medical image may be immediately stored in the database 110. The medical image stored in the database 110 may be automatically provided to the image receiving unit 120 or may be provided to the image receiving unit 120 in response to a request of the image receiving unit 120 by a user.

The medical image provided to the image receiving unit 120 is provided to the image classifying unit 130 and the quality determining unit 140, the operation of the image classifying unit 130 and the quality determining unit 140 for the medical image is performed in real time or It may be performed sequentially. In an embodiment, an operation by the image classifying unit 130 may be performed first or an operation of the quality determining unit 140 may be performed first, and the operation according to the present invention is not limited to a specific order.

The image classification unit 130 determines the type of the medical image by using information about the characteristics of the medical image stored in the database 110. When determining the type of medical image by the image classifying unit 130, it may be determined whether the corresponding medical image is a CT image or an MRI image, and if the MRI image is a T1 image or a T2 image.

As described above, a difference in image contrast occurs according to an image capturing technique, and since the database 110 stores the result of learning the difference, the image classifying unit 130 uses the corresponding learning result to perform the corresponding medical treatment. The type of image can be determined.

MRI T1 image means T1-weighted image. Water is black and fat is white on MRI T1 image, water is white and fat is black on MRI T2 image. On the other hand, in the CT image, bones appear white and water appears black.

The image classifying unit 130 considers which part of the medical image provided from the image receiving unit 120 is an image taken by the image, and on the basis of how much brightness of the part included in the medical image is viewed as a reference. The type of can be judged.

Meanwhile, the database 110 may further store a result of learning the header information included in the medical image. The header information may include personal information of the patient, medical information, and the like. It may also include information on whether the CT image or MRI image. In addition, the header information may include information about a part or an organ included in the medical image.

The quality determining unit 140 determines the quality of the medical image by using information about the characteristic of the noise stored in the database 110.

The quality judgment may determine whether the medical image is suitable for use in diagnosis and / or prescription. If the quality of the medical image does not meet a predetermined standard, the medical image may be determined not to be used.

The quality determining unit 140 uses information about the characteristics of the noise stored in the database 110 to determine whether artifacts are present in the medical image provided from the image receiving unit 120 and the noise of the noise present in the medical image. The quality of the medical image may be determined in consideration of the degree.

Meanwhile, as described above, the medical image may be simultaneously provided to the image classifying unit 130 and the quality determining unit 140 from the image receiving unit 120 and provided to the image classifying unit 130 and the quality determining unit 140. May be performed simultaneously, but the process of determining the type may not be necessary for the medical image having a quality that is not suitable for use in diagnosis and / or prescription. After performing the quality determination operation by the first step 140, only the medical image that satisfies the quality of a predetermined level or more may be provided to the image classification unit 130 to determine the type.

However, the operation is performed by the user's selection, and the process of determining the type and determining the quality is not limited to a specific order.

2 exemplarily shows an MRI image of a horizontal cross section of the brain. 2 (a) and 2 (b) respectively show a brain in which a disease occurs and a normal brain, and show MRI images photographed by different methods.

2 (a) and 2 (b) are left MRI T1 images. In the MRI T1 image, the cerebrospinal fluid (Colony-Stimulating Factor, CSF) is shown in black. Referring to the left image of FIG. 2 (a), it can be seen that the cerebrospinal fluid in the ventricles is shown in black. In contrast, cerebrospinal fluid is shown in white in the right image of FIG. 2 (a), which is an MRI T2 image.

In FIG. 2 (b), the MRI T1 image and the MRI T2 image may be distinguished in consideration of the brightness contrast of the cerebrospinal fluid.

Meanwhile, as described with reference to FIG. 1, the database 110 may further include header information, and the header information may include information about a part or an organ included in a corresponding medical image. The image classifying unit 130 may determine which part of the medical image is photographed based on the header information, and determine the type of the medical image through the brightness contrast of the medical image.

Also, in the case of the brain image shown in FIG. 2, even if the corresponding medical image is an image of the brain, the header information may determine which pixel of the medical image should be used to determine the brightness contrast. have. For example, in order to determine the brightness contrast of the cerebrospinal fluid, it is necessary to determine the brightness contrast of the pixels corresponding to the ventricles.

Therefore, the image classifying unit 130 may include information about the shape of the brain, and may include information about the shape of other parts or organs. Information about the shape of the site or organ may be obtained through machine learning.

3 illustrates the characteristics of brightness contrast for each medical image type. Referring to the CT scan image, the cerebrospinal fluid (CSF) is the darkest and the bones are the brightest. Blood also appears darker than bones.

In the MRI T1 image, water is darkest and fat is lightest, as described with reference to FIG. 1. As shown in FIG. 3, even the same material is shown at different brightnesses according to the imaging technique, and thus, the type of medical image may be determined through learning the brightness contrast according to the imaging technique.

Noise in medical images can be classified into cases generated by a patient and artifacts caused by chemical shifts or RF signals. Noise generated by the patient may be generated in the MRI by regular movement by the patient's breathing, movement of the patient during imaging or specific posture of the patient.

4 (a) and 4 (b), it can be seen that the noise in the wave form and the noise in the blur form occur in the image, respectively. As shown in FIG. 4, when noise occurs in a medical image, the image is fatally acted on to diagnose the image and thus interfere with accurate diagnosis.

Therefore, when a medical image is acquired, the medical staff undergoes a process of examining the quality of the medical image with the naked eye. However, accurate quality judgment may not be made due to the skill of the medical staff and may take a considerable time.

Information on the characteristics of the noise stored in the database 110 may be made of information about the shape of the noise or the characteristics of the pixel data as shown in Figure 4, the quality determination unit 140 relates to the characteristics of the noise The information may be used to determine whether noise exists in the medical image, and the quality of the medical image may be determined according to the degree.

Through this, it is possible to select a medical image that can be accurately diagnosed, and to take a retake, etc. for a low-quality medical image, and also to expect the effect of processing this process in a short time.

Referring to FIG. 5, the medical image management system 200 according to another exemplary embodiment of the present invention may include a database 210, an image receiver 220, an image classifier 230, a quality determiner 240, and a face region separation. The unit 250 includes a request receiving unit 260 and a de-identifying unit 270. The database 210, the image receiver 220, the image classifier 230, and the quality determiner 240 may include the database 110, the image receiver 120, and the image classifier 130 described with reference to FIG. 1. , And performs substantially the same function as the quality determination unit 140, so that detailed description thereof will be omitted only for the overlapping contents.

The face area separator 250 receives a medical image from the image receiver 220 and separates a face area from the medical image. The medical image management system 200 illustrated in FIG. 5 may be obtained from the medical image to solve a problem in which the personal information of the patient is unnecessarily disclosed when the medical image of the patient is provided to the outside for research and the like. It is intended to remove personal information that may be.

Therefore, the medical image provided by the image receiving unit 220 is an image for identifying the facial region of the patient, and is an image in which the shape of the face of the patient's eyes, nose, mouth, etc. can be specified. I can understand.

The facial region separator 250 separates the facial region of the patient from the medical image. The facial region may be understood as a facial region including the eyes, nose, mouth, and ears of the patient. In order to separate the face region from the medical image, the face region separator 250 may use a machine learning result of a human eye, nose, mouth, ear, and the like. For example, the facial region separator 250 may be a new eye, nose, or mouth in a newly input medical image through learning to determine positions of eyes, nose, and mouth from CT or MRI images of a head of a plurality of patients. It is possible to determine the position of.

In this case, the object of the machine learning may be a 2D image or a 3D image. When the medical image is a 2D image, the positions of the eyes, the nose, the mouth, and the ear may be learned from a plurality of 2D images. Alternatively, when the medical image is a 3D image, learning may be performed to determine a facial area including an eye, a nose, a mouth, and an ear of a patient.

Therefore, when learning a 2D image and utilizing the corresponding learning result, a plurality of images including an eye, a nose, a mouth, and an ear may be selected by the face region separator 250.

The medical image may further include text information indicating personal information of the patient as well as an image of a body part of the patient. For example, the medical image may include information such as the name, contact information, diagnosis name, etc. of the patient in a text form.

The request receiving unit 260 receives a data providing request for at least one of an eye, a nose, and a mouth included in the face area from a client. The medical image may be used for research and the like, and the client may be a person, an institution, or another system that wants to use the medical image.

The purpose of the study may vary and the body parts of the study may vary. The client may request a medical image of at least one of eyes, nose and mouth. In this case, if the medical image of the patient is provided without additional processing, the patient may be identified through the 3D medical image, and even if the 2D medical image is provided, the 3D image may be generated by combining the patient information. Can be.

The de-identification unit 270 performs de-identification on the remaining facial areas except for the area requested by the client in order to prevent the unintended patient information from being exposed.

For example, when the client requests a data for the nose of the patient, the de-identification unit 140 performs de-identification on eyes, mouths, ears, etc. except the nose. The de-identification is an irreversible transformation and when the client is provided with a medical image on which the nose has been de-identified, it is impossible to restore the patient's nose from the medical image.

The de-identifier 270 may perform de-identification on a plurality of images including the nose by converting the medical image into a 2D image, and de-identifying the nose area directly in the 3D medical image. You can carry out anger. Meanwhile, the de-identifier 270 may convert the medical image into a plurality of 2D images after de-identifying the 3D medical image. That is, the medical image provided to the client may be understood as having a 2D form.

The de-identification may be masking or blurring of a target site, for example, an image processing that masks all data corresponding to the non-identifying target site to '0' or blurs the target site. Can be. As another example, a mosaicing operation may be performed on the target site.

Meanwhile, in another embodiment of the present invention, the non-identifying unit 270 may remove or replace the personal identification text included in the medical image with another text. As described above, the data corresponding to the personal information of the patient can be viewed as a non-identification target according to the present invention, and the personal identification text can be viewed as the personal information of the patient.

Accordingly, when the medical image includes text for identifying the patient individual in the medical image, the de-identification unit 270 may perform de-identification by deleting or replacing the text with another text.

Referring to FIG. 6, a medical image in which de-identification is performed on the entire facial region from a medical image in which de-identification is not performed sequentially from FIGS. 6 (a) to 6 (c) is illustrated. Although the medical image shown in FIG. 6 is a 3D image, the de-identification may be performed directly on the 3D medical image or the 3D image of FIG. 6 may be obtained from the 2D image on which the non-identification has been performed.

Referring to Figure 6 (a), it can be seen that the eyes, nose, mouth of the patient is represented to the extent that the individual can be specified. And, it can be seen that Figure 6 (b) can not identify the individual by performing de-identification for the eyes and mouth of the patient. As described with reference to FIG. 5, the request receiving unit 260 receives a request for providing a data for a specific part from a client. Referring to FIG. 6 (b), the request receiving unit 260 receives a request for providing a data for a nose from a client. I can understand.

On the other hand, Figure 6 (c) can be understood that the de-identification was performed for all eyes, nose, mouth of the patient, it can be understood that the client did not request data for the eyes, nose, mouth. However, as shown in FIG. 6 (c), even if de-identification is performed on all eyes, noses, and mouths of the patient, image damage does not occur in subcutaneous areas including the brain and blood vessels. It cannot be specified and there is no problem with using the necessary information.

In addition, the medical image that has been de-identified according to the client's request may be a 3D image as shown in FIG. 6, but the information provided to the client may be a 2D image as shown in FIG. 3. Therefore, the client may utilize the provided 2D image according to the necessary purpose, but even if the 3D image is generated by using the 2D image, the client cannot obtain information enough to identify the individual patient.

As described with reference to FIG. 5, the de-identification unit 270 may perform de-identification on the medical image, and the de-identification may be performed by masking or blurring the de-identification target site. blurring).

7 exemplarily illustrates a non-identification method using a convolution filter, where a source pixel corresponds to a non-identification target image and a convolution kernel corresponds to a matrix applied to the source pixel.

The source pixel and the central element of the convolution kernel are multiplied with each other, and as a result of the calculation, the sum of the product of the peripheral element of the source pixel and the elements of the corresponding positions of the convolution kernel may be determined as a new pixel value.

In the example of FIG. 7, the product of the source pixel and the center element of the convolution kernel is 0, but since the sum of the products of the remaining elements is -8, the new pixel value is determined to be -8.

On the other hand, the size of the matrix corresponding to the source pixel and the convolution kernel can be variously determined, and is not necessarily limited to 3x3 as shown in FIG. Similarly, the weights of the elements of the convolution kernel are not limited to the values shown in FIG. 7, and thus the degree of de-identification can be controlled by adjusting the kernel size and the kernel weight.

In the de-identification process according to the present invention, when the convolution as described above is performed on the non-identification target pixel, all values are designated as 0 for the boundary region to which the square matrix cannot be applied or designated by duplicating the values assigned to the surroundings. The method may be applied.

If Kernel Convolution is applied, it can be judged that irreversible de-identification has been made. De-identified images can only approximate the approximate shape of the object and cannot be converted to the original data.

On the other hand, hash mapping may be applied to secure additional irreversibility for data on which irreversible de-identification has been performed. The hash mapping may be applied based on the signal intensity of Voxel of the unidentified region.

Referring to FIG. 8, in the medical image management method according to an embodiment of the present invention, a database construction step (S10), a medical image reception step (S20), a medical image type determination step (S30), and a medical image quality determination step (S40).

In the database construction step (S10), a database including information on characteristics of the medical image and characteristics of artifacts included in the medical image is constructed. The medical image may be an image such as CT (Computed Tomography), MRI (Magnetic Resonance Imaging), or a plurality of 2D images or 3D images.

Meanwhile, the information about the characteristics of the medical image stored in the database may include a result of learning brightness contrast according to a photographing technique for a plurality of medical images, the type of medical image based on the brightness contrast degree. Can be determined.

The artifacts refer to noise that may be included in the medical image. If the noise exists in the medical image, the diagnosis may not be accurately made through the medical image, and thus the medical image including the noise may not be used. You may not. Therefore, the information about the characteristics of the noise may be an important indicator for determining the quality of the medical image, and the database may include a result of learning whether the noise is included in the medical image and the characteristics of the noise.

In the medical image receiving step (S20), the medical image is received, and the type and quality of the medical image may be determined based on the information stored in the database. In the medical image receiving step (S20), the medical image may be received from the database or directly from the user.

In general, when a medical image is acquired, it may be desirable to determine the type and quality of the medical image before the medical image is acquired, so that the medical image may be immediately stored in the database when the medical image is generated by the medical imaging apparatus. The medical image stored in the database may be automatically provided to the medical image receiving step S20 or may be provided in response to a request by the user.

In the medical image type determination step (S30), the type of the received medical image is determined by using information about the characteristics of the medical image stored in the database. When the medical image type determination in the medical image type determination step (S30) is performed, it may be determined whether the medical image is a CT image or an MRI image, and in the case of an MRI image, whether the medical image type is a T1 image or a T2 image. .

As described above, a difference in image contrast occurs according to an image capturing technique, and since the database stores the result of learning the difference, the medical image type determination step (S30) uses the learning result. The type of medical image can be determined.

In the step of determining the type of medical image (S30), the type of the medical image is determined based on the brightness of the part included in the medical image in consideration of which part of the medical image is taken. You can judge.

The database may further store a result of learning header information included in the medical image. The header information may include personal information of the patient, medical information, and the like, and the medical image may be a CT image. It may also include information about whether it is an MRI image. In addition, the header information may include information about a part or an organ included in the medical image.

In the medical image quality determining step (S40), the quality of the received medical image is determined by using information about the characteristic of the noise stored in the database. The quality judgment may determine whether the medical image is suitable for use in diagnosis and / or prescription. If the quality of the medical image does not meet a predetermined standard, the medical image may be determined not to be used.

In the medical image quality determining step (S40), the medical image is determined in consideration of whether artifacts exist in the medical image and the degree of noise present in the medical image by using the information on the characteristics of the noise stored in the database. The quality of the image can be determined.

Meanwhile, as described above, the medical image may be simultaneously provided to the medical image type determining step S30 and the medical image quality determining step S40 through the medical image receiving step S20. The operation in step S30 and the medical image quality determining step S40 may be performed at the same time, but the process of determining the type is unnecessary for the medical image having a quality that is not suitable for use in diagnosis and / or prescription. Therefore, according to the user's selection, only after performing the quality determination operation by the medical image quality determination step (S40), the medical image type determination step (S30) is limited to the medical image that satisfies a certain level or more. Can be provided to determine the type.

Referring to FIG. 9, the database construction step according to an embodiment of the present invention may include a medical image type learning step S11, a noise learning step S12, and a header information learning step S13.

In the medical image type learning step (S11), the brightness contrast degree according to the photographing technique of the plurality of medical images is learned. At this time, by learning the brightness of a specific material for the medical image as shown in Figure 2 can learn the brightness contrast according to the imaging technique, through which the medical image type determination step (S30) The type can be judged accurately.

In the noise learning step (S12), whether the noise is included in the medical image and learning the characteristics of the noise. In this step, by learning whether the noise is included in the medical image as shown in Figure 4, if the noise is included in the medical image quality determination step (S40) by learning the characteristics of the noise You can judge.

In the header information learning step (S13), header information included in the plurality of medical images is learned. As described above, the header information may include personal information, medical information, and the like of the patient, and may include information about whether the medical image is a CT image or an MRI image. In addition, the header information may include information about a part or an organ included in the medical image.

Therefore, in the header information learning step (S13), by learning the data constituting the header information and what each data means, it can be used as an auxiliary tool to determine the type of medical image or the quality of the medical image.

9 illustrates the medical image type learning step S11, the noise learning step S12, and the header information learning step S13, but the sequence shown in FIG. 8 is not necessarily required. Regardless of the type, noise and header information may be learned.

10, the medical image management method according to another embodiment of the present invention, database construction step (S10), medical image reception step (S20), medical image type determination step (S30), medical image quality determination step ( S40), a face region separating step S50, a data providing request receiving step S60, and a de-identifying performing step S70. In the database building step (S10), the medical image receiving step (S20), the medical image type determining step (S30), and the medical image quality determining step (S40), the database building step (S10) and the medical image described with reference to FIG. Since the same operations as in the reception step (S20), the medical image type determination step (S30), and the medical image quality determination step (S40) are performed, detailed descriptions thereof will be omitted only for overlapping contents.

In the facial region separating step (S50), the facial region is separated from the medical image. The facial region may be understood as a facial region including the eyes, nose, mouth, and ears of the patient. In order to separate the face region from the medical image, in the face region separation step (S50), machine learning results of a human eye, nose, mouth, and ear may be used. For example, in the face region separation step (S50), eyes, nose, mouth in a newly input medical image through learning to determine the position of eyes, nose, mouth, etc. from CT or MRI images of the head portion of a plurality of patients. It is possible to determine the position of.

Therefore, when learning the 2D image and utilizing the corresponding learning result, a plurality of images including eyes, nose, mouth, and ears may be selected in the face region separation step S50.

In the data providing request receiving step (S60), a data providing request for at least one of an eye, a nose, and a mouth included in the face area is received from a client. The medical image may be used for research and the like, and the client may be a person, an institution, or another system that wants to use the medical image.

The purpose of the study may vary and the body parts of the study may vary. The client may request a medical image of at least one of eyes, nose and mouth. In this case, if the patient is provided as it is without further processing for the medical image of the patient, the patient can be identified through the 3D medical image, and even if the 2D medical image is provided, the 3D image may be generated by combining the patient information, thereby providing patient information. Can be.

In the de-identification step (S70), in order to prevent the unintended patient information from being exposed, the de-identification is performed on the remaining facial areas except for the area requested by the client.

For example, when the client makes a request for data on the nose of the patient, the de-identification step (S70) performs de-identification on eyes, mouths, ears, etc. except the nose. The de-identification is an irreversible transformation and when the client is provided with a medical image on which the nose is de-identified, it is impossible to restore the patient's nose from the medical image.

In the de-identification step (S70), the medical image may be converted into a 2D image, and de-identification may be performed on a plurality of images including the nose. Identification can be performed. In the de-identification step (S70), after de-identification of the 3D medical image, the medical image may be converted into a plurality of 2D images. That is, the medical image provided to the client may be understood as having a 2D form.

On the other hand, in another embodiment of the present invention, in the non-identification step (S70) may be removed or replaced with other text of the personal identification text included in the medical image. As described above, the data corresponding to the personal information of the patient can be viewed as a non-identification target according to the present invention, and the personal identification text can be viewed as the personal information of the patient.

Therefore, in the non-identification step (S70), if the medical image includes the text for identifying the individual patient, de-identification may be performed by deleting another or replacing the text with another text.

Meanwhile, the present invention can be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

If the steps constituting the method according to the invention are not explicitly stated or contrary to the steps, the steps may be performed in a suitable order. The present invention is not necessarily limited to the description order of the above steps.

The use of all examples or exemplary terms (eg, etc.) in the present invention is merely for describing the present invention in detail, and the scope of the present invention is limited by the examples or exemplary terms unless the scope of the claims is defined. It is not limited. Also, one of ordinary skill in the art appreciates that various modifications, combinations and changes can be made in accordance with design conditions and factors within the scope of the appended claims or their equivalents.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and all the scope equivalent to or equivalent to the scope of the claims, as well as the claims to be described later, are defined by the spirit of the invention. It belongs to the category.

[Description of the code]

100, 200: medical image management system

110, 210: database

120, 220: video receiver

130, 230: image classification unit

140, 240: quality judgment

250: face area separator

260: request receiving unit

270: de-identification unit

Claims

A database containing information about characteristics of the medical image and characteristics of artifacts included in the medical image;

Image receiving unit for receiving a medical image;

An image classification unit which determines a type of the received medical image by using information about characteristics of the medical image stored in the database; And

A quality determination unit that determines the quality of the received medical image by using information about the characteristic of the noise stored in the database;

Medical image management system comprising a.
The method of claim 1,

The database includes a medical image management system including information about a result of learning brightness contrast according to a plurality of medical image capturing techniques, and a result of learning whether noise is included in the medical image and characteristics of the noise.
The method of claim 2,

The database further includes information about a result of learning the header information included in the plurality of medical images.
The method of claim 2,

The learning of the brightness contrast degree is a medical image management system for learning the brightness contrast degree of the normal region in the plurality of medical images.
The method of claim 2,

The image classification unit determines a type of the medical image based on the brightness contrast of the received medical image management system.
The method of claim 1,

A face region separator for separating a face region from the medical image;

A request receiving unit which receives a request for providing data for at least one of an eye, a nose and a mouth included in the face area from a client; And

De-identification unit for performing non-identification for the remaining facial region except the requested area;

Medical image management system further comprising.
The method of claim 6,

The database further includes information about eye, nose and mouth learning results included in the medical image,

The face region separator may separate the face region using the eye, nose, and mouth learning results.
The method of claim 6,

The non-identifying unit is a medical image management system for performing irreversible data conversion for the eyes, nose, mouth.
Receiving a medical image;

Determining a type of the received medical image by using information about the characteristic of the medical image stored in a database including information about the characteristic of the medical image and the characteristics of the artifacts included in the medical image; And

Determining the quality of the received medical image by using information about the characteristic of the noise stored in the database;

Medical image management method comprising a.
The method of claim 9,

The medical image management method further comprises the step of building the database;

The step of building the database,

Learning brightness contrast according to a technique of photographing a plurality of medical images; And

Learning whether noise is included in the medical image and characteristics of the noise;

Medical image management method comprising a.
The method of claim 10,

The step of building the database,

Medical image management method further comprising the step of learning the header information (Header Information) included in the plurality of medical images.
The method of claim 10,

In the step of learning the brightness contrast degree, medical image management method for learning the brightness contrast degree of the normal region in the plurality of medical images.
The method of claim 9,

And determining the type of the medical image based on the brightness contrast degree of the received medical image.
The method of claim 9,

Separating a facial region from the received medical image;

Receiving a request for providing data for at least one of an eye, a nose, and a mouth included in the face region from a client; And

Performing de-identification on the remaining facial regions except for the requested site;

Medical image management method further comprising.
The method of claim 14,

The database further includes information about eye, nose and mouth learning results included in the medical image,

In the step of separating the face area medical image management method for separating the face area using the machine learning results for the eyes, nose, mouth.
The method of claim 14,

In the step of performing the de-identification, medical image management method for performing irreversible data conversion for the eyes, nose, mouth.
A computer-readable recording medium having recorded thereon a program for performing the method according to any one of claims 9 to 16.