JP2013052245A - Information processing device and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for efficiently understanding both images of a photographed disease site and diagnosis support information for the images.SOLUTION: An information processing device for supporting diagnosis includes: first means for obtaining position information for indicating the disease site recorded in a schema of a subject; and second means for extracting an area corresponding to the disease site from medical image data of the subject based on the position information.

Description

  The present invention relates to an information processing apparatus for supporting creation of a medical record (medical record) and an image diagnosis report. In particular, the present invention relates to a schema display technique that a doctor draws on a medical document.

  In the medical field, CR (Computed Radiography) apparatus, CT (Computed Tomography) apparatus, MRI (Magnetic Resonance Imaging) apparatus, and ultrasonic apparatus (US: US) are used as apparatuses for generating medical images for examining a patient's body. Ultrasound System). The doctor displays the medical image of the patient photographed by these imaging devices on the monitor, observes the state of the diseased part and the change over time from the displayed medical image, and uses the results of the image diagnosis as finding information Fill out the diagnostic imaging report. Then, a doctor who performs a definitive diagnosis for the disease enters the patient's symptoms and disease name in the medical record. At this time, a medical document such as an image diagnosis report or medical chart is created and attached with a schema depicting characteristics of the judgment on the diseased part so that the judgment made by the doctor on the diseased part can be easily seen visually. be able to.

  Traditionally, schemas are handwritten on a medical document made of paper. However, in recent years, as medical information systems such as hospital information systems (HIS) and image storage communication systems (PACS) have become widespread, electronic medical records and other electronic documents have been digitized. The schema digitization is also progressing gradually. When creating a schema as electronic data, an arbitrary shape figure can be input to a computer as line drawing information by using an input device such as a mouse or a tablet. However, when creating an illustration (referred to as a basic schema) representing a body structure in a schema, it is necessary to draw a human body structure having a complicated shape, and drawing using a mouse or tablet is not easy.

  Therefore, Patent Document 1 discloses a technique in which a number of basic schema templates are stored in the apparatus in advance, and a doctor selects a desired basic schema. In this method, after a doctor selects a basic schema, a diagram showing the state of the diseased part in a simple graphic and explanatory text (called the diseased part schema) is drawn on the basic schema to easily create the schema. it can. Furthermore, Patent Document 2 proposes an apparatus that stores not only the basic schema but also the diseased part diagram constituting the diseased part schema, the explanatory text, and the arrows connecting them as separated electronic data. As a result, it is possible to easily edit a created schema with operations such as changing, moving, and deleting characters and figures included in the schema.

  Further, Non-Patent Document 1 and Non-Patent Document 2 propose SHIFT feature values and PCA-SHIFT feature values, respectively. For example, by calculating these feature values in the basic schema, the result can be used as one of the structural information of the body part drawn in the basic schema. Further, Non-Patent Document 3 proposes a region dividing method using a variable shape model. For example, a diseased region surrounded by edges can be extracted by applying a deformable model to an image or a schema.

  By the way, in the clinical field, a diagnosis for the purpose of observing the progress of a disease state or a healing state of a diseased part such as a tumor, which is one of the purposes of image diagnosis, is performed. In follow-up observation, not only images taken in the past of the same patient, but also schemas created in the past are checked, and the doctor's mental image and text are linked to information on past decisions made by the doctor. It can be referred to as information and used as a reference for diagnosis.

  Conventionally, as shown in Patent Document 3, Patent Document 4, and the like, a doctor displays a plurality of different images at the time of imaging related to the same subject (referred to as a patient to be examined) as an image on a CRT display device or the like. By displaying on the display means, the change with time of the diseased part was judged. In this case, the image display means simultaneously displays or switches a plurality of images at the time of photographing, and an observer such as a doctor makes a comparative observation of the plurality of display images to determine temporal changes in the diseased part. I was trying to do it.

JP 2006-181146 A JP 2000-222503 A Japanese Patent Laid-Open No. 1-107739 JP-A-8-263625

D. G. Lowe, "Object recognition from local scale-invariant features", Proc. Of IEEE International Conference on Computer Vision (ICCV), pp.1150-1157, 1999 Y. Ke, R. Sukthankar, "PCA-SHIFT: A more distinctive representation for local image descriptors", Proc. Of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.511-517, 2004 Kass M, Witkin A, Terzopoulos D: Snakes: active contour models.International Journal of Computer Vision 1 (4): 321-331, 1988

  However, in the above-described method, although the temporal change in the state of the diseased part can be grasped from the image of the diseased part taken, the diagnostic information (how the doctor characterizes and judges the diseased part at the time of each photographing) ( (Schema) cannot be grasped. For this reason, doctors need to separately check various information corresponding to captured images.

  The present invention has been made in view of the above, and an object of the present invention is to provide a technique capable of efficiently grasping a captured image of a diseased part and diagnosis support information for the image.

  In order to solve the above-described problems, the information processing apparatus of the present invention has the following configuration. That is, in an information processing apparatus that supports diagnosis, the first means for obtaining position information indicating the position of a diseased part recorded in a schema of the subject, and the medical image data of the subject based on the position information And a second means for extracting a region corresponding to the diseased part.

  Alternatively, in the information processing apparatus, an acquisition unit that acquires a plurality of schemas recorded at different time points regarding the subject and medical image data related to diseased part information recorded in each of the plurality of schemas, and the plurality of the plurality of schemas Display control means for displaying the schema and the corresponding medical image data in association with each other on the display means.

  Furthermore, in the information processing apparatus, an acquisition unit configured to acquire a plurality of schemas described at different time points regarding the subject and medical image data related to diseased part information recorded in each of the plurality of schemas; Deriving means for deriving an image feature amount of the diseased part region for each of the medical image data related to each of the schemas, time information of each of the plurality of schemas, and medical image data related to each of the plurality of schemas A time-series image feature quantity generating means for generating a time-series image feature quantity relating to the diseased part area based on an image feature quantity of the diseased part area, and a time-series image relating to the plurality of schemas and the diseased part area Display control means for synchronizing and displaying the feature quantity on the display means.

  ADVANTAGE OF THE INVENTION According to this invention, the technique which enables efficiently grasp | ascertaining the image of the image | photographed diseased part and the diagnostic assistance information with respect to the said image can be provided.

It is a figure which shows the apparatus structure of the medical treatment assistance apparatus 1 which concerns on 1st Embodiment. It is a process flowchart of the medical treatment assistance apparatus 1 which concerns on 1st Embodiment. It is a figure showing the feature point on a basic schema. It is a figure showing the disease part schema based on 1st Embodiment. It is a figure which shows the interlocking display of a disease part time series schema and a disease part time series image. It is a figure which shows the other example of the interlocking display of a disease part time series schema and a disease part time series image. It is a process flowchart of the medical treatment assistance apparatus 1 which concerns on 2nd Embodiment. It is a figure showing the disease part time-dependent change graph which concerns on 2nd Embodiment. It is a figure which shows the interlocking display of a disease part time series schema and a disease part time-dependent change graph. It is a figure which shows a diseased part time-dependent change table.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. The following embodiments are merely examples, and are not intended to limit the scope of the present invention.

(First embodiment)
As a first embodiment of an information processing apparatus according to the present invention, a medical assistance apparatus in a data processing system will be described below as an example.

<Device configuration>
FIG. 1 is a diagram illustrating a device configuration of the medical assistance device according to the first embodiment.

  The medical assistance device 1 includes a control unit 10, a monitor (display unit) 104, a mouse 105, and a keyboard 106. The control unit 10 includes a central processing unit (CPU) 100, a main memory 101, a magnetic disk 102, and a display memory 103. When the CPU 100 executes the program stored in the main memory 101, various controls such as communication with the medical document database 2 and the medical image database 3 and overall control of the medical assistance device 1 are executed.

  The medical assistance device 1 is connected to a medical image database 3 that can capture an image of a subject. The medical image database 3 stores image data obtained by a medical image photographing apparatus such as an X-ray CT apparatus, an MRI apparatus, a US apparatus, an X-ray apparatus, or a nuclear medicine apparatus. The medical document database 2 (examination record database) stores document data such as an electronic medical record, an image diagnosis report, and a created schema.

  The CPU 100 mainly controls the operation of each component of the medical assistance device 1. The main memory 101 stores a control program executed by the CPU 100 and provides a work area when the CPU 100 executes the program. The magnetic disk 102 stores an operating system (OS), device drives for peripheral devices, various application software including a program for performing diagnosis support processing, which will be described later, and the like. The display memory 103 temporarily stores display data for the monitor 104. The monitor 104 is, for example, a CRT monitor or a liquid crystal monitor, and displays an image based on data from the display memory 103. The mouse 105 and the keyboard 106 are used by the user for pointing input and character input, respectively. The above components are connected to each other via a common bus 107 so that they can communicate with each other.

  In the first embodiment, the medical assistance device 1 can access the medical document database 2 and the medical image database 3 via the LAN 4 and read medical document data such as an electronic medical record and medical image data. Note that a storage device such as an FDD, a CD-RW drive, an MO drive, a ZIP drive, or the like may be connected to the medical assistance device 1 and medical images or the like may be read from these drives. Alternatively, a medical image or the like may be acquired directly from the medical image capturing apparatus via the LAN 4.

<Operation of the device>
FIG. 2 is an operation flowchart of the medical assistance device 1 according to the first embodiment. 2 is realized by the CPU 100 executing a program stored in the main memory 101.

  In step S <b> 201, the CPU 100 performs processing for inputting medical examination data (medical image and electronic medical record) to be subjected to image diagnosis to the medical treatment support apparatus 1 in accordance with an input from the mouse 105 or the keyboard 106. This medical examination data input process is, for example, a process in which the CPU 100 receives medical images and electronic medical records from the medical document database 2 and the medical image database 3 via the LAN 4 as described above. Further, as described above, the CPU 100 may be configured to read similar data from various storage media such as a FDD, a CD-RW drive, an MO drive, and a ZIP drive connected to the medical assistance device 1. Good.

  In step S202, the CPU 100 performs processing for recognizing patient information to be examined from the medical examination data read in step 201. The patient information recognition process is, for example, a process of extracting information (subject specifying information) for identifying a patient such as a patient ID from header information of a medical image (DICOM header if DICOM image) or an electronic medical record. I do.

  In step S203, the CPU 100 searches the medical document database 2 and the medical image database 3 using the patient information recognized in the patient information recognition process in step S202 as a search condition. And the process which extracts the data of the past 1 or more schema and medical image of a patient to be examined is performed. This will be referred to as past examination record extraction processing. Then, the extracted data is stored in the magnetic disk 102. At this time, when data of a plurality of examination dates are extracted, examination data numbers i are newly assigned to the schema and medical image data in the order of the examination dates.

  As for the schema and medical images stored in the medical document database 2 and the medical image database 3, basically, a schema depicting a diseased part is created every time a doctor diagnoses an image at a certain shooting date and time. Suppose you are. That is, regarding diagnosed data, basically, all schemas and medical images are stored in each database in a one-to-one correspondence. In addition, information on the corresponding medical image is recorded along with the schema. As the accompanying information, for example, the ID of a medical image and the shooting date (time information) are recorded. Therefore, in this process, after extracting the corresponding patient's schema and medical image from each database, the i-th schema and the i-th medical image are associated with each other using the information recorded in each schema as a key. Save to the magnetic disk 102.

  However, there may be a case where information of either the schema or the image is missing. In this case, for example, an inspection data number including data for which there is no corresponding partner may be given, and an inspection data number may be given as blank data to the missing data side. Alternatively, the inspection data number may be counted by skipping data for which no other party exists.

  In addition, the schema recorded in the medical document database 2 selects a basic schema template stored in advance in the apparatus and draws the diseased part schema on the schema in order to save the labor of drawing the body part. Suppose that Accordingly, each schema is assumed to be separated into a “basic schema” and a “sick part schema” and stored in the medical document database 2.

  In step S204, the CPU 100 stores the value 1 in the examination data number i to be processed, and stores the total number of past examination records extracted for the examination subject patient in the constant n.

  In step S <b> 205, the CPU 100 reads the i th examination data related to the patient to be examined, that is, the i th schema and the i th medical image from the magnetic disk 102 to the main memory 101.

  In step S206, the CPU 100 performs a process of extracting the structure information of the body part drawn on the schema and the location / range information of the diseased part regarding the i-th schema read out to the main memory 101 in step S205. As described above, each schema is stored in the basic schema and the diseased part schema separately.

  FIG. 3 is a diagram showing an example in which coordinates of characteristic points (shape features) (positions indicated by ◯ in the figure) on the outline of the body part are extracted as the structure information of the body part drawn on the basic schema. is there. Features such as contour branch points and points with maximum curvature are extracted as feature points.

  There are several different means for extracting the structure information of the body part drawn on the basic schema. As one of them, for example, there is a method in which structure information of a body part is input as basic schema attribute information in advance when a basic schema is created. Also, for example, from the group name to which the basic schema belongs in the hierarchical list of basic schemas or the identification name given to the basic schema, the position or organ name of the body part and the direction of the body part (front, left side, right side, etc.) ) Can be identified. Further, as described above, the CPU 100 calculates the SHIFT feature value proposed in Non-Patent Document 1 or the PCA-SHIFT feature value proposed in Non-Patent Document 2 as a method for automatically extracting the structure information of the body part. There is a way to do it. However, SHIFT is an abbreviation for Scale-Invariant Feature Transform, and PCA is an abbreviation for Principal Component Analysis.

  In addition, as a method for extracting the location / range information (disease part information) of the diseased part depicted in the diseased part schema, the following methods can be considered.

  FIG. 4 shows a diseased part schema drawn on a basic schema 401 representing a body part where the diseased part exists. The diseased part schema generally comprises the following three types of information (parts).

A figure 402 representing the state of the diseased part as a picture of a doctor (referred to as a figure drawn by the diseased part)
A sentence 403 for explaining a diseased part drawing figure (referred to as a diseased part explanation)
An arrow 404 (referred to as a connecting arrow) that connects a disease part description and a disease part drawing figure and faces the disease part drawing figure from the disease part description
That is, the diseased part drawing figure 402 may be extracted from the diseased part existing position / range information. At this time, as shown in FIG. 4, the disease part description 403 and the connecting arrow 404 are generally composed of line components, but the disease part drawing figure 402 is used for the doctor to clearly indicate the position and range of existence. In many cases, it is represented by a figure surrounded by a closed boundary line. Therefore, for example, by applying the deformable model proposed in Non-Patent Document 3 described above to an image of a diseased part schema, a region surrounded by edges can be extracted. The figure can be taken out. Next, by calculating the coordinates of the central part of the drawing of the diseased part, the approximate location of the diseased part on the basic schema (upper lobe, middle lobe, lower lobe, etc. in the lung field) can be obtained. .

  In addition, as described in Patent Document 2, when the three components of the diseased part schema are stored as electronic data separated in advance, the presence position / range of the diseased part drawing figure is directly read. Good.

  In step S <b> 207, the CPU 100 performs a process of extracting structural information of the body part and candidate information of the existence position / range of the diseased part in the image regarding the i-th medical image read out to the main memory 101 in step S <b> 205. When extracting the structure information of the body part shown in the medical image, the structure information of the body part should be information that can be compared with the structure information of the body part drawn on the basic schema.

  Similar to the method for extracting the structure information of the body part drawn in the basic schema, there are several different methods for extracting the structure information of the body part shown in the medical image. For example, there is a method of displaying a medical image on the monitor 104 and having the doctor specify the position of the feature point on the outline of the body part shown in the medical image. At this time, the feature point specified by the doctor is specified at a position corresponding to the feature point on the outline of the basic schema illustrated in FIG. Alternatively, there is a method in which the CPU 100 automatically calculates the aforementioned SHIFT feature value or PCA-SHIFT feature value for a medical image.

  As a method for extracting candidate information of the existing position / range related to the diseased part shown in the medical image, there is a method of extracting by using a computer-aided diagnosis (CAD) technique. For example, lung cancer is one example of a disease with a relatively high occurrence frequency and a high risk of death, and this is taken as an example. Primary lung cancer appears on the image with a nodular shadow close to the shape of a sphere. Therefore, by applying a filter that emphasizes a spherical shadow and performing threshold processing on the enhanced image, a candidate shadow region of lung cancer, that is, a candidate shadow region of a diseased part can be automatically extracted.

  In step S208, the CPU 100, the structure information of the body part and the diseased part drawn in the i-th schema extracted in step S206, and the structure of the body part and the diseased part shown in the i-th medical image extracted in step S207. Compare information. Then, body part association is performed by calculating geometric correspondence (referred to as structure information association processing). At this time, when the medical image is composed of a three-dimensional image or a plurality of two-dimensional images, a cross-sectional image in the same direction as the schema is specified from the image data, and the schema is associated with the detected cross-sectional image.

  As examples of the structure information, the above-described feature point coordinates, SHIFT feature amount, or PCA-SHIFT feature amount can be said to be vector information composed of a plurality of components. Therefore, the specific process of step S206 is to search for vector information that is paired between the basic schema and the medical image. In this process, a vector distance between one vector information of the basic schema and one vector information of the medical image may be calculated, and a vector combination that minimizes the distance among all vector combinations may be obtained. However, when the distance between vectors is larger than a predetermined threshold, it is determined that there is no paired vector. Furthermore, as a contrivance to reduce the search error of the corresponding vector, it is possible to combine several vectors in the vicinity of each other instead of one vector and handle them as higher-dimensional vectors. At this time, the inter-vector distance between the high-dimensional vector of the basic schema and the high-dimensional vector of the medical image is calculated.

  Here, the orientation of the body part drawn on the basic schema is extracted as part of the structure information of the body part in step S206. When the medical image is a three-dimensional image, when the orientation of the body part is determined, the direction of the image cross section such as an axial cross section, a coronal cross section, or a sagittal cross section is determined. However, in the three-dimensional image, there are still a plurality of cross-sectional images even if the direction of the image cross-section is determined. Therefore, among these cross-sectional images, a cross-sectional image in which the diseased part is best reflected is selected. For this purpose, it is necessary to identify the actual diseased part from the three-dimensional image. At this time, since the candidate shadow region of the diseased part has already been extracted in step S207, the actual diseased part is identified from the candidate shadows.

  As a method for identifying a diseased part, there is a method of using information obtained in advance regarding a diseased part. In follow-up observation, if there is an image taken before the i-th medical image to be processed, the shape, size, location, and luminance distribution of the diseased part can be predicted to some extent from the previous image. If these pieces of information are recorded as examination records prior to the i-th examination, the actual diseased part can be narrowed down from the candidate shadows based on the information. Each cross-sectional image is in the same direction as the schema, and the position and range of the diseased part in the schema are known to some extent in step S206. Thus, for each cross-sectional image, the position and range of the candidate shadow in the image are compared with the approximate position and range of the diseased part in the schema, and the matching candidate shadow is detected, so that the diseased part can be identified. This makes it possible to associate a diseased part between a schema and an image. Next, as a method for selecting a cross-sectional image in which the diseased part is best reflected, there is a method for selecting a cross-sectional image in which the area of the specified diseased part region is the largest.

  In step S209, the CPU 100 increments the inspection data number i (increases by 1).

  In step S210, the CPU 100 proceeds to step S205 if the value of the inspection data number i is less than or equal to the total number n of past inspection records, and otherwise proceeds to the next step S211 and step S212.

  In step S211, the CPU 100 uses the diseased part schema data in the 1st to nth schemas to generate a schema (referred to as a diseased part time-series schema) capable of grasping the temporal transition of these diseased part schemas. Perform the process. Furthermore, this processing is referred to as diseased part time-series schema generation processing.

  Here, when a set of diseased part schemas is associated and the above three parts constituting the diseased part schema are separated and their coordinates are known, the time series is corrected by correcting them to a form suitable for display. Generate a schema.

  As for the diseased part drawing figure in the diseased part schema, if the examination is for follow-up observation, the location of the diseased part will not change basically. Should be drawn in the same position. Accordingly, the coordinates at which the diseased part drawing figure is arranged are used without change. However, the description of the diseased part and the connected graphic are not always unified between the diseased part schemas because the doctor can arrange the figure at a free position. Therefore, for example, when the diseased part schema is switched and displayed at the same position as the diseased part time-series schema, the positions of the diseased part explanatory texts are displayed in a discrete manner, which makes it very difficult to see. Therefore, the average value of the coordinates of the disease part description in the first to nth disease part schema is calculated, and the coordinates of the disease part description in each disease part schema are changed to the average value. And it is good to comprise so that the connection arrow in each disease part schema may be redrawn so that the disease part drawing figure and the disease part description after a coordinate change may be connected.

  In step S212, the CPU 100 extracts an ROI (region of interest) image surrounding the diseased part as image information of the diseased part from the cross-sectional images associated with the schema in step S208 for the 1st to nth medical images. And the process which produces | generates the image (it is called disease part time-sequential image data) which can grasp | ascertain the temporal transition of these disease part ROI images is performed. Furthermore, this process is called a diseased part time-series image data generation process.

  As a method for determining the ROI image extraction range, for example, there is the following method. In step S208, since the region of the diseased part in the medical image has already been extracted, it is possible to obtain the largest diseased part among the 1st to nth medical images. By calculating the smallest rectangular region or circular region surrounding the diseased part with a certain width margin for the diseased part of the maximum size, a common ROI image extraction range can be obtained for the 1st to nth medical images. Can be determined.

  In step S213, the CPU 100 performs a process of displaying and outputting on the monitor 104 the two time-series data of the diseased part time-series schema generated in step S211 and the diseased part time-series image data generated in step S212. This process is referred to as diseased part time-series data linked presentation process.

  FIG. 5 shows an example of displaying a pair of diseased part time-series schema and diseased part time-series image data in association with each other. Reference numeral 501 in FIG. 5 represents a basic schema depicting a body part to which the diseased part belongs, and the position of the diseased part is also displayed on the basic schema. 502 represents a diseased part time-series schema, and 503 represents an image in the diseased part time-series image data corresponding thereto.

  The diseased part time series schema 502 is simultaneously displayed in order from the smallest value of the examination data number i (1 ≦ i ≦ n), that is, in a state arranged in time series. Similarly, the diseased part time-series image data 503 is also displayed in order from the smallest examination data number i. By these, the operator of the medical treatment support apparatus 1 can observe the temporal change of the diseased part regarding both the diseased part schema and the diseased part image.

  The i-th diseased part schema and the i-th diseased part image are associated with each other. Using this, if the i-th data in one type is specified among the two types of display information associated with each other, the corresponding data can be highlighted in relation to the other display information. For example, when the i-th point of the diseased part time-series schema is designated, the associated i-th diseased part image is highlighted. The reverse designation and display are also possible. As a highlighting method, there are methods such as displaying the i-th diseased part schema and the image display frame by coloring them, or displaying other diseased part schema and images translucently. Thus, the operator of the medical assistance device 1 can simultaneously grasp the disease part time-series schema and the disease part time-series image data having the same examination data number.

  Furthermore, by extracting (selecting) a plurality of data in one type from two types of display information (disease part time-series schema and disease part time-series image data), a plurality of corresponding display information can also be obtained. Data can be highlighted. Thereby, two types of display information can be simultaneously compared between a plurality of data having different inspection data numbers. For example, when two points having different examination data numbers are designated in the diseased part time-series schema, the corresponding diseased part images are highlighted and can be easily compared with each other.

  As described above, according to the medical care support apparatus 1 according to the first embodiment, the operator of the medical care support apparatus 1 observes temporal changes of the diseased part regarding both the diseased part schema and the diseased part image. I can do it. By referring to the diseased part time series schema together with the diseased part image, it is possible to easily grasp the transition of how the doctor characterizes and judges the diseased part. In addition, since the diseased part time-series schema and the diseased part time-series image data can be referred to in conjunction with each other, observation of changes over time while simultaneously grasping the information of the judgment by the doctor and the original image of the diseased part be able to. Thereby, since it is possible to refer to both subjective evaluations in the past and raw information, it is possible to provide information useful for diagnosis in follow-up observation.

(Modification 1)
The linked display of the pair of diseased part time-series schema and diseased part time-series image data in step S213 may take the following form.

  FIG. 6 shows an example of displaying the diseased part time-series schema superimposed on the basic schema in the interlocking display. Reference numeral 601 denotes a basic schema depicting a body part to which a diseased part belongs, and a diseased part time-series schema 602 is also displayed on the basic schema. Reference numeral 603 denotes an image in the disease part time-series image data corresponding to the disease part time-series schema. The diseased part time series schema 602 displays the diseased part schema of the designated examination data number i on a specific display area (on the basic schema), and the corresponding diseased part schema is changed every time the value of i is changed. It is displayed on the display area. Similarly, in the diseased part time-series image data 603, the diseased part image of the designated examination data number i is displayed on a specific display area, and the corresponding diseased part image is displayed every time the value of i is changed. Displayed on the area.

  Similar to FIG. 5, the two display information items are associated with the inspection data number i. When the inspection data number i is designated, the two display information items are sequentially displayed to display the i-th data. As a method of specifying the data of the inspection data number i, there are methods such as automatically adding the value of i continuously on the program or changing the value of i in conjunction with mouse click or drag. is there. Thereby, the images in the diseased part time-series schema and the diseased part time-series image data can be displayed as a moving image (animated display) on the display area.

  By comprising in this way, a continuous change can be grasped | ascertained more intuitively. In addition, by switching and displaying the disease part time-series schema on the basic schema, it becomes possible to seamlessly grasp the information on the body part and the disease part.

(Second Embodiment)
In the second embodiment, a mode in which a schema and a feature amount of a diseased part image corresponding to the schema are displayed in conjunction with each other will be described. Note that the configuration of the medical assistance device according to the second embodiment is the same as that of the first embodiment, and a description thereof will be omitted. However, as will be described below, the control program executed by the CPU 100 is different from that of the first embodiment. That is, the programs stored in the magnetic disk 102 are different.

  FIG. 7 is an operation flowchart of the medical assistance device 1 according to the second embodiment. Note that detailed description of steps in which processing similar to that in the flowchart of FIG. 2 of the first embodiment is performed will be omitted.

  The processing from step S701 to step S711 is the same as the processing from step S201 to step S211 of the first embodiment, respectively.

  In step S <b> 712, the CPU 100 calculates quantitative data of the diseased part region in the image associated with the diseased part in the schema for the 1st to nth medical images. And processing which produces | generates the data (it calls a disease part time-sequential image feature-value (quantitative data)) which can grasp | ascertain quantitative transition of these diseased part area | regions using quantitative data is performed. This process will be referred to as diseased part time-series image feature value generation means.

  Here, quantitative information useful for diagnosis is extracted from the diseased part region data extracted in step S707. For example, when a nodular shadow suspected to be lung cancer is found as a finding in a chest CT image, the size of the information is effective as one of the information useful for diagnosis. Therefore, in this case, the effective radius R of the diseased part region divided in step S708 is calculated. The effective radius R is given by, for example, the following formula (1) or formula (1 ′).

R = (radius of a circle having the same area as the diseased area derived based on the image) (1)
Or R = (radius of a sphere having the same volume as the diseased part volume derived based on the image) (1 ′)
By determining the effective radius R for each medical image having a different shooting date and time based on Expression (1), it is possible to generate disease part time-series quantitative data. As the diseased part time-series quantitative data, for example, a graph (referred to as a diseased part temporal change graph) representing a quantitative transition of the size of a nodule as shown in FIG. 8 can be created. FIG. 8 shows a graph in which points are plotted with the horizontal axis as the imaging date and time for each medical image and the vertical axis as the value of the effective radius R.

  In step S713, the CPU 100 performs processing to display the diseased part time-series schema generated in step S711 and the diseased part time-series quantitative data generated in step S712 in conjunction with each other. Similar to step S213 according to the first embodiment, this process is referred to as a diseased part time-series data linked presentation process.

  FIG. 9 is a diagram illustrating an example in which a diseased part time-series change graph generated as a specific example of the diseased part time-series schema and the diseased part time-series quantitative data is displayed in conjunction with each other.

  Reference numeral 901 denotes a basic schema depicting a body part to which the diseased part belongs, and the location of the diseased part is also displayed on the basic schema. Reference numeral 902 denotes a diseased part time series schema, and reference numeral 903 denotes a corresponding diseased part time series graph. In FIG. 9, the nodule shadow size is applied as the quantitative value of the diseased part. Similar to the linked presentation in FIG. 9, the diseased part time series schema 902 and the diseased part time series graph 903 can be presented together. For example, when the i-th point on the diseased part change graph 903 is designated, the i-th diseased part schema associated therewith can be highlighted. For example, as a highlighting method, there is a method of displaying a colored display frame of an i-th diseased part schema or displaying other diseased part schemas in a translucent manner.

  Further, for example, by changing the point designated along the time series on the diseased part change graph 903, the transition of the two pieces of information can be highlighted in synchronization. As a method for changing a specified point on the graph, for example, a point on the graph is sequentially clicked with a mouse, or a method of dragging from one point to another point is available. Alternatively, there is a method of preparing a control bar corresponding to the horizontal axis of the graph and changing the designated point by moving the control bar. Note that when two points having different test data numbers are designated on the graph, the corresponding two diseased part schemas may be highlighted.

  As described above, according to the medical assistance device 1 according to the second embodiment, the operator of the medical assistance device 1 can use the disease part regarding both the disease part schema and the quantitative data derived from the corresponding disease part image. The change with time can be observed.

(Modification 1)
The generation of disease part time-series quantitative data in step S712 and the linked presentation of time-series data in step S713 may take the following forms.

  FIG. 10 is a diagram showing a disease part chronological change table in which quantitative data of diseased parts are summarized as a table. Similar to the diseased part change graph in step S712, the effective radius of the nodule shadow is shown as quantitative data of the diseased part.

  In step S713, the diseased part time-series schema and the diseased part chronological change table are displayed in a linked manner. For example, when a specific examination data number in the disease part chronological change table is designated, the data in the corresponding disease part time-series schema may be highlighted.

(Modification 2)
The generation of disease part time-series quantitative data in step S712 may take the following form.

  As the quantitative data of the diseased part, the effective radius of the nodule shadow extracted from the image data is applied in step S712, but applicable quantitative data of the diseased part is not limited to this. For example, when the soft region such as a tumor is included in the internal region of the lymph node and when it includes a necrotic liquid tissue, the average value and dispersion of the concentration values in the region are different. In general, the soft tissue tends to have a slightly higher concentration value than the liquid, and the liquid tends to be less dispersed than the soft tissue because the concentration value is uniform. Therefore, it is also preferable to generate the diseased part time-series quantitative data by applying the average or variance value of the image density values inside the diseased part region in the image data as the quantitative data characterizing the state of the diseased part.

(Third embodiment)
In the third embodiment, an example in which two or more pieces of other information (data) to be displayed in conjunction with the schema will be described. That is, in the first and second embodiments, the disease part time-series image and the disease part time-series quantitative data are used as information to be displayed in conjunction with the disease part time-series schema, respectively. However, the information displayed in conjunction with the disease part time-series schema is not limited to one.

  For example, three pieces of information of a diseased part time-series schema, a diseased part time-series image, and a diseased part time-series quantitative data can be presented in conjunction with each other. When a diseased part change graph is used as the diseased part time-series quantitative data, there are the following linked display methods. That is, a point on the graph may be designated in order with a mouse or the like, and the diseased part time-series schema and the diseased part time-series image may be displayed as a moving image (animation display).

  As described above, according to the medical assistance device 1 according to the third embodiment, the operator of the medical assistance device 1 confirms a plurality of other data at the time corresponding to each schema in conjunction with the schema. It becomes possible.

(Fourth embodiment)
In the fourth embodiment, an example in which similar case data is used as information (data) displayed in conjunction with a schema will be described. That is, in the above-described embodiment, information obtained only from past examination images of the patient to be examined is applied as information presented simultaneously with the diseased part time-series schema. However, the information presented at the same time as the diseased part time series schema is not limited to the examination data relating to the patient to be examined.

  For example, there is a method of simultaneously presenting past examination images (referred to as similar images) of other patients similar to the case of the patient to be examined. As a method for acquiring a similar image, for example, using a confirmed diagnosis information described in a medical record of a patient to be examined as a keyword, a patient medical record in which matching diagnostic information is described is searched from a database, and a test image of the patient is acquired. There is a technique to do. Alternatively, there is a method of extracting an image feature amount from an image of a diseased part of a patient to be examined and acquiring an image having the most similar image feature amount as a similar image from a database in which a large amount of various past examination images are stored. .

  Then, after acquiring a similar image, when there are other examination images regarding the patient related to the image, a time-series similar image (referred to as similar case time-series image data) can be generated. The processing so far will be referred to as similar case time-series image data generation processing. At that time, as one method of presenting the time series similar image, it may be configured to present it in conjunction with the diseased part time series image of the patient to be examined on the basis of the progress (degree) of the disease. As an association method based on the progress of the disease, for example, when a nodule shadow suspected to be lung cancer is confirmed as a finding, there is a method of associating those having the same effective radius of the nodule shadow.

  As described above, according to the medical care support apparatus 1 according to the fourth embodiment, the operator of the medical care support apparatus 1 compares the diseased part schema or the corresponding diseased part image with the diseased part image of a similar case. Can be observed.

(Other embodiments)
Although the embodiments of the present invention have been described in detail above, the present invention may be applied to a system constituted by a plurality of devices or may be applied to an apparatus constituted by one device.

  The present invention can also be achieved by supplying a program that realizes the functions of the above-described embodiments directly or remotely to a system or apparatus, and the system or apparatus reads and executes the supplied program code. The Accordingly, the program code itself installed in the computer in order to realize the functional processing of the present invention by the computer is also included in the technical scope of the present invention.

  In this case, the program may be in any form as long as it has a program function, such as an object code, a program executed by an interpreter, or script data supplied to the OS.

  Examples of the recording medium for supplying the program include a floppy (registered trademark) disk, a hard disk, an optical disk (CD, DVD), a magneto-optical disk, a magnetic tape, a nonvolatile memory card, and a ROM.

  Further, the functions of the above-described embodiments are realized by the computer executing the read program. In addition, based on the instructions of the program, an OS or the like running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments can also be realized by the processing.

  Further, the program read from the recording medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. Thereafter, the CPU of the function expansion board or function expansion unit performs part or all of the actual processing based on the instructions of the program, and the functions of the above-described embodiments are realized by the processing.

1 Medical Support Device 2 Medical Document Database 3 Medical Image Database 4 LAN
10 Control unit 100 CPU
101 Main memory 102 Magnetic disk 103 Display memory 104 Monitor 105 Mouse 106 Keyboard 107 Common bus

Claims (18)

An information processing apparatus that supports diagnosis,
A first means for obtaining position information indicating a position of a diseased part recorded in a schema of a subject;
Second means for extracting a region corresponding to the diseased part from the medical image data of the subject based on the position information;
An information processing apparatus comprising: The information processing apparatus according to claim 1, further comprising a third unit that performs control to display the extracted area on a display unit. The schema includes a plurality of schemas recorded at different times;
3. The information processing according to claim 2, wherein the third means performs control to display the plurality of schemas and the regions extracted by the second means in a display form associated in time series. apparatus. When the medical image data is a three-dimensional image or a plurality of two-dimensional images, the second means specifies a cross-sectional image in the same direction as the schema and extracts a region corresponding to the diseased part. The information processing apparatus according to claim 1, wherein the information processing apparatus is characterized. The second means restricts the existence position and range of candidate shadows obtained from the medical image data of the subject by using position information indicating the position of a diseased part recorded in the schema of the subject. The information processing apparatus according to claim 1, wherein: Acquisition means for acquiring a plurality of schemas recorded at different time points on the subject and medical image data related to diseased part information recorded in each of the plurality of schemas;
Display control means for displaying the plurality of schemas and corresponding medical image data in association with each other on display means;
An information processing apparatus comprising: Based on the diseased part information recorded in each of the plurality of schemas, further comprising a specifying means for specifying a diseased part region in medical image data related to each of the plurality of schemas,
The information processing apparatus according to claim 6, wherein the acquisition unit acquires medical image data based on the identified diseased part region. And further comprising an association means for geometrically associating the shape feature of the body part drawn on the schema with the shape feature of the body part in the medical image data related to the schema,
The specifying unit specifies a diseased part region in medical image data related to the schema based on a result of geometric association of body parts by the matching unit and diseased part information drawn on the schema. The information processing apparatus according to claim 7. The display control means extracts a plurality of pairs of the schema and the corresponding medical image data from the plurality of schemas and the corresponding medical image data acquired by the acquisition means, The information processing apparatus according to any one of claims 6 to 8, wherein the pairs are displayed in the display unit in chronological order. The information processing apparatus according to any one of claims 6 to 9, wherein the display control unit synchronizes and displays the plurality of schemas and corresponding medical image data on the display unit. An examination record database for storing a schema related to the subject and medical image data related to the schema;
11. The acquisition unit according to claim 6, wherein the acquisition unit acquires a plurality of schemas and medical image data relating to the subject from the examination record database based on subject specifying information for specifying the subject. The information processing apparatus according to any one of claims. Acquisition means for acquiring a plurality of schemas described at different time points regarding the subject and medical image data related to diseased part information recorded in each of the plurality of schemas;
Deriving means for deriving an image feature amount of the diseased part region for each of the medical image data related to each of the plurality of schemas;
When generating time-series image feature values for the diseased part region based on time information of each of the plurality of schemas and an image feature value of the diseased part region of medical image data related to each of the plurality of schemas A series image feature value generation means;
Display control means for synchronizing and displaying the plurality of schemas and time-series image feature quantities related to the diseased part area on a display means;
An information processing apparatus comprising:   The information processing apparatus according to claim 12, wherein the image feature amount is a size of a diseased part region.   The information processing apparatus according to claim 12, wherein the image feature amount is a density value of a diseased part region. An information processing method for supporting diagnosis,
A first step of obtaining position information indicating the position of the diseased part recorded in the schema of the subject;
A second step of extracting a region corresponding to the diseased part from the medical image data of the subject based on the position information;
An information processing method comprising: An acquisition step of acquiring a plurality of schemas described at different time points regarding the subject and medical image data related to diseased part information recorded in each of the plurality of schemas;
A display control step of associating and displaying the plurality of schemas with corresponding medical image data on a display unit;
An information processing method comprising: An acquisition step of acquiring a plurality of schemas described at different time points regarding the subject and medical image data related to diseased part information recorded in each of the plurality of schemas;
A derivation step of deriving an image feature amount of the diseased part region for each of the medical image data related to each of the plurality of schemas;
When generating time-series image feature values for the diseased part region based on time information of each of the plurality of schemas and an image feature value of the diseased part region of medical image data related to each of the plurality of schemas A sequence image feature generation step;
A display control step of synchronizing and displaying the plurality of schemas and time-series image feature amounts related to the diseased part region on a display means;
An information processing method comprising:   The program for functioning a computer as each means of the information processing apparatus as described in any one of Claims 1 thru | or 14.