JP7802883B2 - Information processing device, information processing method, and program - Google Patents

Description

The present invention relates to an information processing device, information processing method, and program for processing medical images captured using various modalities.

In the medical field, doctors make diagnoses using medical images captured using various modalities. In particular, to monitor the progress of a subject's condition, doctors compare multiple medical images captured using the same modality at different times to observe changes in the subject over time. As a method for supporting the observation of changes in a subject over time, Patent Document 1 discloses technology that generates a medical image (difference image) that depicts the difference between a medical image of the subject to be diagnosed (diagnostic image) and a medical image of the subject to be compared (comparison image), and displays changes in the subject over time.

JP 2013-126575 A

However, when observing changes in a subject over time using multiple medical images, it was necessary to select an appropriate comparison image from multiple medical images taken at different times.

The present invention therefore aims to provide an information processing device that can select a medical image to be compared from multiple medical images based on conditions related to the imaging date and time.

In order to achieve the object of the present invention, an information processing device is provided which has a selection means for selecting a second image to be compared with a first image from a plurality of candidate images, and which has an evaluation means which sets conditions based on the imaging date and time and imaging part of the first image, obtains the imaging date and time and imaging part of each of the plurality of candidate images, and evaluates the plurality of candidate images based on the imaging date and time and imaging part of the plurality of candidate images against the conditions, and the selection means selects at least one candidate image from the plurality of candidate images as the second image based on the evaluation result of the evaluation means, and the evaluation means includes as the condition any one of the following: the imaging date and time being a certain period or more after the imaging date and time of the first image, the imaging date and time being within a predetermined period after the first imaging date and time, or the imaging date and time being within a predetermined period that is a certain period or more after the imaging date and time of the first image and has an upper and lower limit .

According to the present invention, medical images to be compared can be selected from multiple medical images based on conditions related to the imaging date and time.

1 is a diagram showing the configuration of an information processing apparatus according to a first embodiment of the present invention; FIG. 4 is a diagram showing conditions related to image capture date and time set by a condition setting unit of the present invention. 3 is a flowchart showing the operation of the first embodiment of the present invention. FIG. 10 is a diagram showing the configuration of an information processing apparatus according to a second embodiment of the present invention. 10 is a flowchart showing the operation of the second embodiment of the present invention. 10A and 10B are diagrams showing an example of the relationship between the imaging range of a diagnostic object image and the imaging range of a candidate image according to the present invention; 10A and 10B are diagrams showing an example of the relationship between the imaging range of a diagnostic object image and the imaging range of a candidate image according to the present invention; 10A and 10B are diagrams showing an example of the relationship between the imaging range of a diagnostic object image and the imaging range of a candidate image according to the present invention; 10A and 10B are diagrams showing an example of the relationship between the imaging range of a diagnostic object image and the imaging range of a candidate image according to the present invention;

A preferred embodiment of the present invention will now be described with reference to the accompanying drawings.

An information processing device 10 according to a first embodiment will be described with reference to FIGS. 1 to 3. The information processing device 10 according to this embodiment of the present invention provides the following functions to users, such as doctors and radiologists at medical institutions. That is, the information processing device 10 performs a comparison process between a diagnostic target image and a comparison image relating to a patient (subject) on whom the user is about to undergo medical treatment, and provides the results of the comparison process to the user.

In this embodiment, we will explain the case where there are multiple medical images (candidate images) of the same subject captured at different times from the time the diagnostic target image was captured. The information processing device 10 selects a medical image suitable for comparison with the diagnostic target image from the multiple candidate images as a comparison image, and performs a comparison process between the diagnostic target image and the comparison image. Here, comparison processing means, for example, displaying each medical image so that the user can compare them, or performing a difference process between the diagnostic target image and the candidate image and displaying the difference image.

The information processing device 10 depicts the occurrence and progression of lesions in the subject by comparing a medical image captured at the current time or the most recent medical image (diagnostic target image) with a medical image captured earlier than the diagnostic target image (comparison image).

The information processing device 10 acquires the imaging dates and times when multiple candidate images were captured and sets conditions related to the imaging dates and times. Then, based on the set conditions and the imaging dates and times of the multiple candidate images, a comparison image is selected from the multiple candidate images. The information processing device 10 then generates a difference image depicting the difference between the diagnostic target image and the comparison image, and displays the subject's changes over time on the display unit 36. Note that in the following description, the imaging date and time includes not only information indicating both the imaging date and imaging time, but also information indicating only the imaging date. It also includes information indicating only the imaging time. For example, it also includes information recording only the elapsed time from a specific starting time (e.g., January 1, 1970, 00:00:00). The imaging date generally refers to the imaging date, but may also be information indicating only the imaging month and year.

Figure 1 shows the overall configuration of the medical information processing system of the present invention and the configuration of an information processing device 10. The medical information processing system includes an information processing device 10 and a medical image database 23. The information processing device 10 and the medical image database 23 are communicatively connected to each other via communication means. In this embodiment, the communication means is configured as a LAN (Local Area Network) 21, but may also be a WAN (Wide Area Network). Furthermore, the communication means may be connected via a wired connection or a wireless connection.

The medical image database 23 stores multiple medical images related to multiple subjects and their associated information. Medical images are images captured using modalities such as CT scanners or MRI scanners, and can be various images, such as two-dimensional or three-dimensional, monochrome or color. The associated information for medical images includes information such as subject ID (patient ID), imaging date and time, imaging area, reconstruction conditions, and imaging conditions such as tube voltage and tube current values. Other additional information can also include the purpose of the examination, the name of the organ being examined, and the name of the disease being examined.

A unique identifier (examination ID) is assigned to each medical image and its associated information to enable it to be distinguished from other information, and the information is read by the information processing device 10 based on the examination ID. If the medical image is a three-dimensional volume image composed of multiple two-dimensional cross-sectional images (tomographic images), an examination ID is assigned to each two-dimensional cross-sectional image (tomographic image) and each three-dimensional volume image, which is a collection of tomographic images. In addition to reading information, the medical image database 23 works in conjunction with the information processing device 10 to provide functions such as displaying a list of medical images, displaying thumbnails, searching, and writing information. In this embodiment, medical images refer to medical image data.

The information processing device 10 acquires information held in the medical image database 23 via the LAN 21. The information processing device 10 includes a communication IF 31, ROM 32, RAM 33, a storage unit 34, an operation unit 35, a display unit 36, and a control unit 50.

The communication IF (Interface) 31 is realized, for example, by a LAN card, and communicates between the external device (medical image database 23) and the information processing device 10 via the LAN 21. The ROM (Read Only Memory) 32 is realized, for example, by non-volatile memory, and stores various programs. The RAM (Random Access Memory) 33 is realized, for example, by volatile memory, and temporarily stores various information. The storage unit 34 is realized, for example, by an HDD (Hard Disk Drive), and stores various information. The operation unit 35 is realized, for example, by a keyboard or mouse, and the user can input various information via the operation unit 35. The display unit 36 is, for example, a display. The operation unit 35 and display unit 36 function as a graphical user interface (GUI) under the control of the control unit 50.

The control unit 50 is realized, for example, by a CPU (Central Processing Unit) and controls the overall processing in the information processing device 10. The control unit 50 includes a diagnostic target image acquisition unit 51, a candidate image acquisition unit 52, an imaging date and time acquisition unit 53, a condition setting unit 54, a selection unit 55, an image processing unit 56, and a display control unit 57.

The diagnostic target image acquisition unit 51 acquires a medical image of the subject as a diagnostic target image from the medical image database 23 via the communication IF 31 and LAN 21 in accordance with user operations input via the operation unit 35. The diagnostic target image acquisition unit 51 then outputs the acquired diagnostic target image together with accompanying information to the image processing unit 56.

The candidate image acquisition unit 52 acquires multiple medical images of the same subject previously captured from the medical image database 23 via the communication IF 31 and LAN 21 as medical images (candidate images) that are candidates for comparison images for the diagnostic target image acquired by the diagnostic target image acquisition unit 51. The candidate image acquisition unit 52 then outputs supplementary information for the multiple acquired candidate images to the imaging date and time acquisition unit 53, and outputs the multiple candidate images to the selection unit 55.

The image capture date and time acquisition unit 53 acquires the image capture date and time of multiple candidate images. Since the image capture date and time is included in the additional information attached to the multiple candidate images, the image capture date and time acquisition unit 53 can acquire the image capture date and time of multiple candidate images by reading the additional information attached to the multiple candidate images.

The condition setting unit 54 sets conditions related to the imaging date and time in order to select a medical image (comparison image) to be compared from multiple candidate images. Specifically, the condition setting unit 54 sets the imaging date and time based on the imaging date and time of the diagnostic target image.

Figure 2 shows an example of a condition related to imaging date and time set by the condition setting unit 54. The imaging date and time includes the date and time when the subject was imaged. However, for simplicity, the following explanation will use the date. The condition setting unit 54 can set, for example, the most recent imaging date and time (the imaging date and time closest to the imaging date and time of the diagnostic target image) as the imaging date and time condition. This enables comparison with the most recent examination, which is the basis of comparative image interpretation. In this condition, the selection unit 55 selects the candidate image corresponding to the most recent imaging date and time as the comparison image. In the example of Figure 2, the imaging date and time of the diagnostic target image is April 15, 2019. Among multiple candidate images 1 to 5, candidate image 5 was selected by the selection unit 55 because its imaging date and time was October 20, 2018, making it the candidate image with the most recent imaging date and time. The selection unit 55 outputs the selected candidate image 5 to the image processing unit 56 as the comparison image. The oldest imaging date and time (the imaging date and time furthest from the imaging date and time of the diagnostic target image) can also be set as the imaging date and time condition. This allows the image that has changed the most from the diagnostic target image to be selected as the comparison image. Under this condition, the selection unit 55 selects the candidate image with the oldest imaging date and time as the comparison image.

The condition setting unit 54 can also set an imaging date and time within a predetermined period based on the imaging date and time of the diagnostic target image as a condition related to the imaging date and time. For example, a lower limit can be set as the predetermined period. For example, a condition can be set that the examination was conducted at least a predetermined period (e.g., one year) after the imaging date and time of the diagnostic target image. This makes it possible to select candidate images from examinations conducted within an appropriate period as comparison images when it is known that a certain amount of time is required for the development of a lesion of interest. Furthermore, an additional condition related to the imaging date and time can be set that the imaging date and time is the most recent within the predetermined period. This makes it possible to select as the comparison image a candidate image from an examination that satisfies the period condition and is closest to the diagnostic target image and has no additional changes.

Conversely, an upper limit can be set as the predetermined period. For example, as shown in FIG. 2(a), the condition setting unit 54 can set the period to, for example, within two years based on the imaging date and time of the diagnostic target image (April 15, 2019). This reduces the increase in noise caused by past changes appearing in the difference image due to a large time difference between the diagnostic target image and the comparison image. Alternatively, the predetermined period can be set with an upper and lower limit, such as one year or more but less than two years. When a predetermined period is set as a condition, the selection unit 55 selects, as comparison images, candidate images whose imaging dates fall within the predetermined period. In the example of FIG. 2(a), among multiple candidate images, candidate image 4 was imaged on September 5, 2018, and candidate image 5 was imaged on October 20, 2018. Therefore, candidate image 4 and candidate image 5 are selected by the selection unit 55. The selection unit 55 outputs the selected candidate images 4 and 5 to the image processing unit 56 as comparison images. Furthermore, an additional condition related to the imaging date and time can be set that the imaging date and time must be the earliest within a specified period. This allows the examination that will produce the greatest change within the period condition to be selected. In this case, in the example of Figure 2(a), candidate image 4 is selected as the comparison image.

The condition setting unit 54 may be able to set multiple conditions with different priorities as conditions related to the image capture date and time. For example, multiple levels of conditions can be set, such that if there are no candidate images that satisfy the first priority condition, a candidate image that satisfies the second priority condition is selected. For example, the first priority condition could be that the image capture date and time is between one year and two years old, and the second priority condition could be between two years and three years old. Furthermore, any number of conditions, three or more, may be set with priorities, such as between three years and four years old. This has the effect of reducing the number of cases where images cannot be compared, because even if there are no comparison images with ideal conditions, it is possible to select a comparison image that satisfies the next best conditions.

The condition setting unit 54 can also set, as a condition related to the imaging date and time, an imaging date and time that does not include a predetermined treatment such as surgery between the imaging date and time of the diagnostic target image. In this case, the selection unit 55 selects, as the comparison image, a candidate image corresponding to an imaging date and time that does not include a predetermined treatment such as surgery between the imaging date and time of the diagnostic target image.

If treatment such as surgery is performed between the image capture date and time of the candidate image and the image capture date and time of the diagnostic target image, the pixel values around the treatment site in the candidate image and the diagnostic target image will differ significantly, making the candidate image unsuitable for comparison. Figure 2(b) shows that surgery was performed at the image capture date and time of candidate image 4 and the image capture date and time of candidate image 5. Among multiple candidate images, candidate image 5 was selected by the selection unit 55 because its capture date and time does not include any treatment such as surgery between the image capture date and time of the diagnostic target image. The selection unit 55 outputs the selected candidate image 5 to the image processing unit 56 as the comparison image. Note that the image capture date and time may also be set as a condition that it does not include any specific treatment such as surgery and is as old as possible. This allows the image with the greatest change from the diagnostic target image to be selected as the comparison image, within the range where no major changes due to surgery or other factors occur.

The condition setting unit 54 can also set the imaging date and time condition to be as recent as possible before the start of a specified treatment. In this case, the selection unit 55 selects, as the comparison image, a candidate image captured on the most recent date and time before the specified date and time of the specified treatment. This makes it possible to select an appropriate comparison image when observing changes due to the effects of treatment.

The condition setting unit 54 can also set, as a condition related to the imaging date and time, an imaging date and time when the modality model has not been changed between the imaging date and time of the diagnostic target image.

Under this condition, the selection unit 55 selects as the comparison image a candidate image corresponding to an imaging date and time when the modality model has not been changed between the imaging date and time of the diagnostic target image.

This is because if the modality is changed between the capture date and time of the candidate image and the capture date and time of the diagnostic target image, the image quality of the candidate image and the diagnostic target image will differ significantly, making the candidate image unsuitable for comparison. Figure 2(c) shows that a modality change occurred between the capture date and time of candidate image 3 and the capture date and time of candidate image 4. Among the multiple candidate images, candidate image 4 or candidate image 5 is selected by the selection unit 55 because the capture dates and times of candidate image 4 and candidate image 5 do not include a modality change between the capture dates and times of the diagnostic target image. The selection unit 55 outputs the selected candidate image 4 or candidate image 5 to the image processing unit 56 as the comparison image. Note that the conditions for the capture date and time may be that the modality has not been changed and that the capture date and time is as old as possible. This allows the image that has changed the most from the diagnostic target image to be selected as the comparison image, within the range where no significant changes occur due to the modality change. Additionally, while a change in modality model is used as an example here, the date and time of any other event that affects image quality, such as a change in imaging protocol or an upgrade of the reconstruction algorithm, can also be used as a condition.

Note that the conditions regarding the image capture date and time may be a combination of the above conditions.

Generally, multiple images are captured during an examination at the same imaging date and time. Therefore, it may be difficult to identify a single comparison image from the candidate images solely based on the imaging date and time. Even if multiple examinations satisfy the imaging date and time condition, it may be difficult to identify a single comparison image from the candidate images solely based on the imaging date and time. In such cases, the selection unit 55 may be configured to further narrow down the comparison images by setting conditions other than the imaging date and time. In this case, the condition setting unit 54 may select a comparison image from the candidate images that satisfy the imaging date and time condition based on conditions other than the imaging date and time (e.g., imaging range, imaging site, imaging conditions). The imaging site may be, for example, the lung field, abdomen, or head. For example, the condition setting unit 54 may set a condition for selecting a candidate image of the same imaging site as the diagnostic target image. For example, if the diagnostic target image is an abdominal image, the condition setting unit 54 may set the imaging site to be the abdomen as a condition, and the selection unit 55 may select a candidate image of the abdomen as the comparison image.
Similarly, if the diagnostic target image is an image of the head, the condition setting unit 54 sets the condition that the imaging region is the head, and the selection unit 55 selects a candidate image of the head as the comparison image. Similarly, the selection condition may be that the purpose of the examination, the name of the disease to be examined, etc., match with the diagnostic target image.

Furthermore, in the case of imaging conditions, reconstruction conditions, slice thickness, contrast conditions (presence or absence of contrast agent), etc. can be set as selection conditions for comparison images. For example, the condition setting unit 54 can set, as a condition for selecting comparison images, that the images are images under imaging conditions suitable for subtraction processing. For example, the conditions can be set as follows: a slice thickness of 5 mm or less, a reconstruction function that satisfies the mediastinal condition, no contrast agent, a range of tube voltage or tube current, etc.
The condition setting unit 54 can also set a condition for selecting a comparison image that the image is captured under the same imaging conditions as the diagnostic target image. For example, if the slice thickness of the diagnostic target image is 2.5 mm, the condition setting unit 54 sets a slice thickness of 2.5 mm as a condition for selecting a comparison image from among the candidate images. In this case, the selection unit 55 selects, from among the multiple candidate images, a candidate image with a slice thickness of 2.5 mm as the comparison image. Similarly, a condition for selecting a comparison image can also be set that the reconstruction function, tube voltage, and tube current of the diagnostic target image match, or that the difference is within a predetermined range. A condition combining multiple items can also be set. In this case, for example, a comprehensive evaluation value can be calculated based on the degree of match of each item, using the product or sum of these values, and the candidate image with the best evaluation value can be selected as the comparison image. Alternatively, all candidate images with evaluation values exceeding a predetermined threshold can be selected as the comparison image. Note that in the above example, a comparison image is selected from candidate images that satisfy the imaging date and time condition based on conditions other than the imaging date and time. However, the order in which the conditions are applied is not limited to this. For example, candidate images may be narrowed down based on conditions other than the imaging date and time, and then comparison images may be selected based on the imaging date and time. In particular, when the imaging date and time condition is "oldest" or "newest," it is desirable to narrow down the candidate images based on conditions other than the imaging date and time first. For example, a condition may be set such that, from among candidate images that satisfy conditions other than the imaging date and time, the candidate image with the oldest imaging date and time or the candidate image with the most recent imaging date and time may be selected. This allows comparisons to be made based on a purpose (e.g., to see changes that have occurred in a patient over time, or to see changes since the previous examination for the same purpose) from candidate images that satisfy conditions other than the imaging date and time.

Then, the image processing unit 56 performs a comparison process (difference process) between the diagnostic target image and the comparison image (selected candidate image), and outputs the comparison result of the comparison process to the display control unit 57. Specifically, the image processing unit 56 generates a difference image that depicts the difference between the diagnostic target image and the comparison image.
The display processing unit 57 displays the comparison result (difference image) on the display unit 36 .

Note that at least some of the components of the control unit 50 may be implemented as independent devices. Furthermore, each component may be implemented as software that implements its function. In this case, the software that implements the function may run on a server via a network such as the cloud. In this embodiment, each component is implemented by software in a local environment.

Furthermore, the configuration of the information processing device 10 shown in FIG. 1 is merely an example. For example, the storage unit 34 of the information processing device 10 may have the functionality of the medical image database 23, and the storage unit 34 may store multiple medical images related to multiple subjects and their associated information.

Next, the operation of the information processing device 10 in this embodiment will be described in detail using Figure 3. Note that the following description will be given using an example in which CT images are used as medical images, but the implementation of the present invention is not limited to this. Furthermore, the diagnostic target image and the candidate image do not necessarily have to be images of the same subject and the same modality.

(Step S1010) <Acquisition of diagnostic target image>
In this step, the diagnostic target image acquisition unit 51 executes a process of acquiring a diagnostic target image (first image) to be processed from the medical image database 23. This process is executed by accepting a user's operation via a GUI provided by the operation unit 35 or the display unit 36. Through this process, the diagnostic target image acquisition unit 51 reads the diagnostic target image from the medical image database 23, and the information processing device 10 stores the diagnostic target image using the ROM 32, RAM 33, storage unit 34, etc. In this embodiment, a case where a three-dimensional CT image composed of multiple tomographic images is acquired as the diagnostic target image will be described as an example. However, the implementation of the present invention is not limited to this, and two-dimensional CT images, plain X-ray images, MRI images, camera images of the lesion area of the subject, etc. may also be used. The diagnostic target image is accompanied by additional information such as the subject ID (patient ID), imaging date and time, imaging site, and imaging parameters.

(Step S1020) <Acquisition of Multiple Candidate Images>
In this step, the candidate image acquisition unit 52 executes a process of acquiring a plurality of medical images (candidate images) to be compared with the diagnostic target image acquired in step S1010 from the medical image database 23. As an example of this process, the candidate image acquisition unit 52 acquires, as candidate images, medical images of the same subject as the diagnostic target image from among a plurality of medical images recorded in the medical image database 23. Specifically, the candidate image acquisition unit 52 acquires, using a search function provided in the medical image database 23, medical images having, as incidental information, the same subject ID as the subject ID that is incidental information of the diagnostic target image.

The method of acquiring candidate images is not limited to the above method. For example, the candidate image acquisition unit 52 may refer to the capture date and time of the diagnostic target image and acquire only older (past) medical images. In this embodiment, the above method is used to acquire candidate images that are limited to medical images of the same subject that were captured at a time earlier than the capture date and time of the diagnostic target image and that were captured using the same modality as the diagnostic target image (a CT scanner in this specific example).

(Step S1030) <Obtaining image capture dates and times of multiple candidate images>
In this step, the image capture date and time acquisition unit 53 acquires the image capture dates and times of the multiple candidate images acquired by the candidate image acquisition unit 52. The image capture dates and times recorded together with the candidate images are recorded as additional information in the header area of each candidate image. The image capture date and time acquisition unit 53 acquires the image capture dates and times of the multiple candidate images by reading the additional information of the multiple candidate images.
Here, the imaging date and time acquisition unit 53 may acquire the implementation date (examination date) of the examination on which the candidate images were acquired and use that as the imaging date and time of each candidate image acquired in the examination. Alternatively, information correlated with the imaging date and time, such as the date on which the image reconstruction of the candidate image was performed or the date on which the image was first interpreted, may be used as the imaging date and time.

(Step S1035) <Setting conditions related to image capture date and time>
In this step, the condition setting unit 54 sets conditions related to the image capture date and time for selecting a comparison image from a plurality of candidate images. For example, the conditions are set by having the user select a condition from the variations of the image capture date and time conditions listed above via a GUI (not shown). Alternatively, the condition setting unit 54 may be configured to read conditions previously written in a setting file. Note that it is desirable for the condition setting unit 54 to also set conditions other than the image capture date and time in this step.

(Step S1040) <Select a comparison image>
In this step, the selection unit 55 selects a comparison image from among the plurality of candidate images based on the conditions relating to the image capture date and time set by the condition setting unit 54 and the image capture dates and times of the plurality of candidate images.

Note that the processing by the selection unit 55 does not necessarily require narrowing down the comparison image to one, and multiple comparison images may be selected. In this case, the configuration may be such that the user can select the comparison image to be output to the image processing unit 56 from the multiple comparison images selected by the selection unit 55 via a GUI (not shown). Alternatively, the configuration may be such that all of the selected multiple comparison images are output to the image processing unit 56.

(Step S1050) <Comparison Process>
In this step, the image processing unit 56 performs a comparison process between the diagnostic target image and the comparison image. Specifically, the image processing unit 56 performs a difference process between the diagnostic target image and the comparison image and calculates a difference image as the comparison result. Here, if the selection unit 55 selects multiple comparison images, the image processing unit 56 generates a difference image between the diagnostic target image and the comparison image for each comparison image. That is, the image processing unit 56 compares a first image, which is the diagnostic target image, with a second image, which is the comparison image, to generate a difference image between the first image and the second image. The difference process between the diagnostic target image and the comparison image is preferably performed after correcting image differences that are unnecessary for image diagnosis, such as differences in the position and shape of the subject depicted in the diagnostic target image and the comparison image, and differences in image quality between the images. Correction of differences between images may be performed using any known method, and detailed description thereof will be omitted here.

Note that the process for calculating the comparison results is not limited to the above-mentioned subtraction process. For example, the image processing unit 56 may generate a superimposed image of the diagnostic target image and the comparison image. In this case, different color channels may be assigned to the diagnostic target image and the comparison image, and the superimposed image may be generated as a color image that combines them. Alternatively, a deformed image may be generated by deforming the comparison image so that the shapes of the diagnostic target image and the subject match.

(Step S1060) <Display of comparison results>
In this step, the display control unit 57 executes processing to display the comparison result obtained by the processing of step S1050 on the display unit 36. As a specific example, the display control unit 57 performs volume rendering on each of the diagnostic target image, comparison image, and difference image, and displays each of the images created by volume rendering (hereinafter, volume rendering images) side by side on the display unit 36.

The method of displaying the comparison results is not limited to the above method, and the display control unit 57 may display a tomographic image of an arbitrary position in each image on the display unit 36, and the tomographic image to be displayed may be changed by a user operation. The display control unit 57 can perform processing such as identifying corresponding positions in the diagnostic target image and the comparison image, and displaying the tomographic image of that position (displaying the corresponding cross section).
That is, when the user changes the tomographic image of the diagnostic object image to be displayed, the display control unit 57 displays the tomographic image at the position of the comparison image corresponding to the position of the changed tomographic image on the display unit 36. Furthermore, when the diagnostic object image and the comparison image are aligned in step S1050, it is desirable that the display control unit 57 perform processing such as switching the tomographic images to be displayed in the diagnostic object image and the difference image in conjunction with each other based on the alignment result.

As described above, according to the present invention, the information processing device 10 has a selection unit 55 that selects a comparison image (second image) to be compared with a diagnostic target image (first image) from a plurality of candidate images, and is equipped with an imaging date and time acquisition unit 53 that acquires the imaging dates and times at which the plurality of candidate images were captured, and a condition setting unit 54 that sets conditions related to the imaging dates and times, and the selection unit 55 selects at least one candidate image from the plurality of candidate images as a comparison image (second image) based on the imaging date and time conditions set by the condition setting unit 54 and the imaging dates and times of the plurality of candidate images. Thus, the selection unit 55 can select a candidate image (comparison image) to be compared from the plurality of candidate images based on the imaging date and time conditions, and the display unit 36 can quickly display the comparison results using the comparison images.

As described above, this embodiment has the advantage of being able to select comparison images that can be suitably compared with the diagnostic target image without requiring the user to perform cumbersome operations, and to provide the user with the results of the comparison process of those images.

In the above embodiment, the results of the comparison process between the selected comparison image and the diagnostic target image are displayed, but displaying the results is not necessarily required. For example, a configuration may be used in which the comparison process results and information identifying the comparison image are associated with the diagnostic target image and saved in the medical image database 23. In this case, when observing the diagnostic target image using a different image viewer, the associated comparison process results and the comparison image can be loaded and displayed. Furthermore, performing the comparison process is not necessarily required. For example, a configuration may be used in which the information identifying the selected comparison image is associated with the diagnostic target image and saved in the medical image database 23. In this case, when observing the diagnostic target image using a different image viewer, the associated comparison image can be loaded and displayed in a comparable format, or a superimposed image or difference image can be generated between the diagnostic target image and the comparison image.

Next, the information processing device 10 of Example 2 will be described using Figures 4 to 9. It differs from Example 1 in that it evaluates a combination of multiple candidate images and selects that combination. Figure 4 shows the overall configuration of the medical information processing system of the present invention and the configuration of the information processing device 10. Here, the configuration that differs from Figure 1 will be mainly described.

The candidate image acquisition unit 52 acquires multiple medical images (candidate images) that are candidates for comparison images for the diagnostic target image acquired by the diagnostic target image acquisition unit 51 from the medical image database 23 via the communication IF 31 and LAN 21. The candidate image acquisition unit 52 then outputs the multiple acquired candidate images to the combination generation unit 60 and the selection unit 55.

The combination generation unit 60 generates at least one combination of the candidate images acquired by the candidate image acquisition unit 52. Here, a combination of candidate images is a combination of one or more candidate images. Furthermore, the combinations of candidate images generated by the combination generation unit 60 include at least one combination of multiple (i.e., two or more) candidate images. The combination generation unit 60 then outputs the generated combination of candidate images to the evaluation unit 61.

The evaluation unit 61 calculates an evaluation value for each combination of candidate images generated by the combination generation unit 60, relating to the appropriateness of comparison with the diagnostic target image. The evaluation unit 61 then outputs each calculated evaluation value to the selection unit 55.

The selection unit 55 selects a combination of candidate images to be compared with the diagnostic target image from among the combinations of candidate images based on each evaluation value. The selected combination of candidate images is composed of one or more candidate images. Hereinafter, the selected combination of candidate images will be referred to as a set of comparison images, and the candidate images that make up the set of comparison images will be referred to as comparison images. The selection unit 55 then outputs the set of comparison images to the image processing unit 56.

The image processing unit 56 performs a comparison process between the diagnostic target image and the set of comparison images, and outputs the comparison results to the display control unit 57. The display control unit 57 displays the comparison results of the image processing unit 56 on the display unit 36.

Next, the overall processing procedure by the information processing device 10 in this embodiment will be described in detail using Figure 5. Steps S1010, S1020, S1050, and S1060 are generally the same as those in Figure 3.

Here, candidate images are represented as I_r,i (1≦i≦N). N represents the number of candidate images obtained in step S1020. In this example, the following explanation will be given using the case where N=3 as a specific example.

Note that if the number N of candidate images acquired in this step is 1, the processing of steps S2000 to S2020 described below may be omitted, and the acquired candidate images may be considered as a set of comparison images, which will be described in detail later, and the processing from step S1050 onwards may be executed.

(Step S2000) <Generate combinations of candidate images>
In this step, the combination generation unit 60 generates multiple combinations of one or more candidate images from the N candidate images acquired in step S1020, including at least one combination of two or more candidate images. In other words, the combination generation unit 60 corresponds to a setting unit that sets multiple combinations including two or more candidate images from the multiple candidate images. In this embodiment, an example is shown in which the combinations of candidate images are generated by combining two different candidate images. In a specific example where N=3, the following three combinations are generated: <I_r,1,I_r,2>, <I_r,1,I_r,3>, and <I_r,2,I_r,3>. In this embodiment, the combinations of candidate images are represented as P_j (1≦j≦M), where M is the total number of combinations. In this embodiment, the following explanation will be given using the case where M=3 as an example.

(Step S2010) <Calculation of evaluation value>
In this step, the evaluation unit 61 executes a process of calculating an evaluation value for each of the combinations of candidate images generated in step S2000 in relation to a comparison with the diagnostic target image.
In this embodiment, a specific example will be described in which an evaluation value is calculated based on the overlapping range (common imaging range) between the imaging range of the combination of candidate images to be evaluated and the imaging range of the image to be diagnosed.

The processing of this step will now be described in detail using Figures 6 to 9. Figures 6 to 9 are diagrams showing the relationship between the imaging range of the diagnostic target image and the imaging range of the candidate images. In Figure 6, 400 is the subject in this embodiment. 410 indicates the range in the body axis direction of subject 400 in the imaging range of the diagnostic target image acquired in step S1010. Here, the imaging range of the diagnostic target image is shown to include the range from subject 400's chest to abdomen. Also in Figure 6, 420, 430, and 440 indicate the imaging range (in the body axis direction) of each of candidate images I_r,1, I_r,2, and I_r,3 acquired in step S1020, respectively. Here, I_r,1 indicates the candidate image of the subject's head, I_r,2 indicates the abdomen, and I_r,3 indicates the chest.

In Figure 7, 450 indicates the overlapping range between the imaging range (the sum of 420 and 430) of candidate image combination P_1 (combination of candidate images I_r,1, I_r,2) and imaging range 410 of the diagnostic target image. In other words, 450 represents the range in which image comparison processing with the diagnostic target image can be performed using candidate image combination P_1. Figure 7 shows that the imaging ranges of the diagnostic target image and candidate image I_r,1 do not overlap, while the imaging ranges of the diagnostic target image and candidate image I_r,2 partially overlap. Therefore, the overlapping range of the imaging ranges of the diagnostic target image and candidate image I_r,2 is the overlapping range of the diagnostic target image and candidate image combination P_1.

In Figure 8, 460 indicates the overlapping range (the sum of 420 and 440) of the imaging range of candidate image combination P_2 (combination of candidate images I_r,1, I_r,3) with the imaging range 410 of the diagnostic target image, expressed in the same way as in Figure 7. Figure 8 shows that the imaging ranges of the diagnostic target image and candidate image I_r,1 do not overlap, while the imaging ranges of the diagnostic target image and candidate image I_r,3 partially overlap. Therefore, the overlapping range of the imaging ranges of the diagnostic target image and candidate image I_r,3 is the overlapping range of the diagnostic target image and candidate image combination P_2.

In Figure 9, 470 indicates the overlapping range (the sum of 430 and 440) of the imaging range of the candidate image combination P_3 (the combination of candidate images I_r,2 and I_r,3) with the imaging range 410 of the diagnostic target image, expressed in the same manner as in Figure 7. Figure 9 shows that the imaging range 410 of the diagnostic target image partially overlaps with the imaging ranges 430 and 440 of candidate images I_r,2 and I_r,3. In this step, the evaluation unit 61 calculates the size of each range (overlapping range) 450 in Figure 7, 460 in Figure 8, and 470 in Figure 9, and calculates an evaluation value E_j (1 ≦ j ≦ M) for each combination of candidate images based on the size of the overlapping range. The evaluation value E_j may be the length of the overlapping range in the body axis direction of the subject 400, as shown in Figures 7, 8, and 9, or the volume of the overlapping range. It may also be the ratio of the overlapping range to the entire imaging range of the diagnostic target image. That is, the evaluation unit 61 corresponds to an example of an overlap range calculation means that calculates the overlap range with the first image for each of a plurality of combinations, and an evaluation value calculation means that calculates an evaluation value for each of a plurality of combinations based on the overlap range. In particular, the evaluation unit 61 corresponds to an example of an overlap range calculation means that calculates, as the overlap range, the range of overlap between the sum of the imaging ranges of two or more candidate images included in the combination and the imaging range of the first image.

The above explanation assumes that the relative positional relationship between the imaging areas of the diagnostic target image and the candidate image is clear. If the relative positional relationship between the imaging areas of each image is not clear, the information processing device 10 can calculate the relative positional relationship by aligning the images in this step. This alignment process can be performed using any well-known method, and a detailed explanation will be omitted here.

Furthermore, the calculation of the relative positional relationship between the imaging ranges of each image is not limited to a method of aligning the images, but may also be calculated based on the additional information of each image. For example, if the diagnostic target image and candidate images are accompanied by information on the "imaging region," the positional relationship between the images can be calculated based on that information. For example, suppose the additional information regarding the imaging region of the diagnostic target image is "chest, abdomen," the additional information regarding the imaging region of candidate image I_r,1 is "head," the additional information regarding the imaging region of candidate image I_r,2 is "abdomen," and the additional information regarding the imaging region of candidate image I_r,3 is "chest." In this case, the imaging range of candidate image combination P_1 (the combination of I_r,1 and I_r,2) is estimated to be "head and abdomen," and the overlapping area with the diagnostic target image is estimated to be "abdomen." Furthermore, the imaging range of candidate image combination P_2 (the combination of I_r,1 and I_r,3) is estimated to be "head and chest," and the overlapping area with the diagnostic target image is estimated to be "chest." Furthermore, it is estimated that the imaging range of the candidate image combination P_3 (the combination of I_r,2 and I_r,3) is "chest and abdomen," and the overlapping range with the diagnostic target image is "chest and abdomen." In this case, the evaluation unit 61 can use the number of parts included in the overlapping range (1 if the overlapping range is "chest" or "abdomen," and 2 if "chest and abdomen") as the evaluation value. Alternatively, a predetermined coefficient can be set in advance for each part of the human body, such as the "chest" or "abdomen," and the evaluation value can be calculated by multiplying each overlapping range by this coefficient. As a more specific example, for the "chest" or "abdomen," the size of the corresponding part on a standard human body is set as the coefficient. This allows the size of the overlapping range to be calculated using a simpler method and used as the evaluation value.

Using the method described above, the evaluation unit 61 calculates an evaluation value E_j (1≦j≦M) for each combination P_j of candidate images generated in step S2010.

(Step S2020) <Select a set of comparison images>
In this step, the selection unit 55 executes a process of selecting a combination of candidate images suitable for comparison with the diagnostic target image from among the M combinations of candidate images based on each evaluation value calculated in step S2010. Specifically, the selection unit 55 selects the combination with the highest evaluation value among the evaluation values E_j (1≦j≦M) for the M combinations of candidate images. The selected combination of candidate images becomes a set of comparison images, and the candidate images constituting the set of comparison images become comparison images. In other words, the selection unit 55 corresponds to an example of a selection means that selects, from among the plurality of candidate images, a plurality of second images to be compared with the first image. In particular, the selection unit 55 corresponds to an example of a selection means characterized by selecting any of the plurality of combinations as a plurality of second images based on the evaluation value.

(Step S1050) <Comparison Process>
In this step, the image processing unit 56 performs a comparison process between the diagnostic target image and the set of comparison images to calculate a comparison result. Specifically, for example, the image processing unit 56 performs a difference process between the diagnostic target image and the set of comparison images to calculate a difference image as the comparison result. That is, the image processing unit 56 corresponds to an example of a generating unit that generates a difference image between the first image and the second image by comparing the first image and the second image. In this embodiment, the set of comparison images is composed of multiple comparison images. As an example of a processing method in this step, the image processing unit 56 may perform a difference process between each of the multiple comparison images included in the set of comparison images and the diagnostic target image. In this case, a difference image between each of the multiple comparison images and the diagnostic target image is calculated, and this set of difference images can be used as the processing result of this step. It is desirable to perform the difference process between the diagnostic target image and each of the multiple comparison images after correcting image differences that are unnecessary for image diagnosis, such as differences in the position and shape of the subject depicted in the diagnostic target image and each of the multiple comparison images, and differences in image quality between the images. Furthermore, the comparison process can process only the overlapping range between the imaging range of each of the multiple comparison images and the imaging range of the diagnostic target image. The overlapping range can be easily obtained from the processing results of steps S2000 and S2010, and therefore a detailed description thereof will be omitted. It is also possible to combine multiple comparison images constituting a set of comparison images into a single comparison image by stitching, and then compare the combined comparison image with the diagnostic target image. In this case, for overlapping areas between multiple comparison images, any comparison image including the overlapping area can be selected and combined. It is also possible to select a comparison image captured at a similar date and time to the diagnostic target image, or to select a comparison image captured under similar conditions to the diagnostic target image, or to select a comparison image with better image quality than the others. In addition to selecting and combining any comparison images, a method of combining overlapping images by smoothly calculating a weighted average can also be an embodiment of the present invention.

Note that the process for calculating the comparison results is not limited to the above-mentioned difference processing. For example, the image processing unit 56 may generate a superimposed image for each of the diagnostic target image and multiple comparison images. In this case, different color channels can be assigned to the diagnostic target image and the comparison images, and the superimposed image can be generated as a color image that combines them.

(Step S1060) <Display of comparison results>
In this step, the display control unit 57 executes processing to display the comparison result obtained by the processing of step S1050 on the display unit 36. As a specific example, the display control unit 57 may perform volume rendering on each of the diagnostic target image, the set of comparison images, and the set of difference images that are the comparison results, and display each of the images created by volume rendering (hereinafter, volume-rendered images) side by side on the display unit 36. In this case, for the set of comparison images or the set of difference images, the display control unit 57 may perform volume rendering on each of the multiple comparison images or difference images that make up the set of comparison images or the set of difference images, and display the multiple volume-rendered images side by side. Alternatively, the display control unit 57 may perform volume rendering on the multiple images that make up each set based on the result of step S1050 or the like, by stitching them together into a single three-dimensional image.

The method of displaying the comparison results is not limited to the above method. The display control unit 57 may display a tomographic image of an arbitrary position in each image on the display unit 36, and the user may change the tomographic image to be displayed. In this case, it is desirable that the display control unit 57 identify corresponding positions in each of the diagnostic target image and comparison image pairs based on the positional relationship between the diagnostic target image and comparison image pair, and display the tomographic image at that position (display the corresponding cross section). In other words, if the user changes the tomographic image of the diagnostic target image to be displayed, the display control unit 57 displays a tomographic image at the position in the comparison image pair that corresponds to the position of the changed tomographic image on the display unit 36. Furthermore, if the diagnostic target image and comparison image pair are aligned in step S1050, it is desirable that the display control unit 57 perform processing such as switching the tomographic images displayed in the diagnostic target image and the difference image in conjunction with each other based on the alignment results.

The information processing device 10 also includes a combination generation unit 60 that sets multiple combinations of multiple candidate images, an overlap range calculation means (evaluation unit 61) that calculates the overlap range between the candidate images included in the multiple combinations and the diagnostic target image (first image), and a selection unit 55 that selects at least one combination from the multiple combinations based on the overlap range, and the selection unit 55 further selects a candidate image included in the selected combination as a comparison image (second image).

As described above, this embodiment has the advantage of being able to select multiple comparison images that can be suitably compared with the diagnostic target image without requiring the user to perform cumbersome operations, and to provide the user with the results of the comparison process of those images.

(Modification 2-1: Combination of Candidate Images <Evaluation Based on Number of Candidate Images>)
In the process of step S2000 of this embodiment, two candidate images are combined as a combination of candidate images. However, this embodiment is not limited to this. For example, three or more candidate images may be combined. In this case, in the process of step S2010, the evaluation unit 61 calculates an evaluation value by comparing the imaging range of the combination of three or more candidate images with the imaging range of the diagnostic target image. The number of candidate images included in each combination of candidate images may be a predetermined constant or may be different numbers. In this case, the combination of candidate images is not necessarily limited to a combination of multiple candidate images; each candidate image may be treated as one of the combinations of candidate images. For example, if the total number of candidate images N is 3, the combination generation unit 60 generates three combinations including only one candidate image, three combinations including two candidate images, and one combination including three candidate images. The subsequent processes may then be performed with the total number of combinations M set to 7. In this case, the greater the number of candidate images combined, the wider the overlapping range tends to be. Therefore, it is desirable to calculate the evaluation value in step S2010 based not only on the overlapping range information but also on the number of candidate images included in the combination of candidate images to be evaluated. For example, when there are two combinations of candidate images with approximately the same evaluation value due to the overlapping range (the difference between the evaluation values is equal to or less than a predetermined threshold), the evaluation unit 61 may increase the evaluation value of the combination of candidate images consisting of the fewer number of candidate images. For example, among combinations of candidate images in which the entire area of the diagnostic target image overlaps, the selection unit 55 may select the combination of candidate images consisting of the fewer number of candidate images as the set of comparison images. Alternatively, the evaluation value based on the overlapping range may be multiplied by a predetermined correction coefficient (the value decreases as the number of images combined increases) depending on the number of candidate images to be combined, and the corrected evaluation value may be used. This allows the comparison process in step S1050 to be performed on a smaller number of comparison images, thereby improving the efficiency of the process in step S1050 and obtaining higher-quality comparison results.

(Variation 2-2: Evaluation Value Calculation Method <Image Quality/Image Capture Conditions>)
The processing of step S2010 in this embodiment has been described as an example of calculating an evaluation value based on an overlapping range between the imaging range of the diagnostic target image and the imaging range of the combination of candidate images (hereinafter referred to as overlapping range evaluation), but the implementation of the present invention is not limited to this.

For example, in addition to evaluating the overlapping range, the evaluation unit 61 may also perform an evaluation based on the image quality of each candidate image included in the combination of candidate images (hereinafter, "image quality evaluation"), and then calculate an evaluation value by integrating the overlapping range evaluation and the image quality evaluation. In other words, the evaluation unit 61 corresponds to an example of an image quality evaluation means for evaluating the image quality of each candidate image included in multiple combinations. Specifically, the evaluation unit 61 estimates the noise level of each candidate image included in each combination of candidate images and reduces the evaluation value for combinations that include candidate images with high noise levels. For example, a configuration may be used in which the evaluation value obtained by the overlapping range evaluation is multiplied by a predetermined correction coefficient (the higher the noise level, the smaller the value) depending on the noise level of the candidate image, and the corrected evaluation value is used. In this case, it is desirable for the evaluation unit 61 to calculate the size of the overlapping range between each candidate image and the diagnostic target image, and to calculate the evaluation value by placing greater emphasis on the noise level of candidate images with larger overlapping ranges. Furthermore, the evaluation unit 61 may calculate the evaluation value by placing greater emphasis on the noise level of the candidate image with the highest noise level among the candidate images that at least partially overlap with the imaging range of the diagnostic target image. Additionally, for example, the evaluation unit 61 may lower the evaluation value of the combination when the image quality of multiple candidate images included in the same combination of candidate images is different. For example, the evaluation value based on the overlapping range may be multiplied by a predetermined correction coefficient (a value that becomes smaller the lower the evaluation value of image quality) depending on the image quality of the candidate images, and the resulting value may be used as the corrected evaluation value.

Furthermore, without being limited to the above-mentioned method, the evaluation unit 61 may evaluate the candidate images based on, for example, reconstruction conditions that are supplementary information of the candidate images, or imaging conditions such as the tube voltage and tube current of the X-ray radiation device. The evaluation unit 61 may also evaluate the candidate images based on the resolution of the candidate images (such as the pixel size of the two-dimensional slices and the slice interval). For example, the evaluation unit 61 may correct the evaluation value of the combination so that the evaluation value of a candidate image with a narrow slice interval is higher.
Furthermore, when the image capturing conditions of multiple candidate images included in the same combination of candidate images are different, the evaluation unit 61 may lower the evaluation value of the combination.

Furthermore, although the above description has been given as an example of a method for evaluation based on the image quality of the candidate image, evaluation may also be based on the image quality of the diagnostic target image. For example, the evaluation unit 61 may evaluate the image quality of the diagnostic target image using a method similar to that described above, and calculate an evaluation value based on both the evaluation of the image quality of the candidate image and the evaluation of the image quality of the diagnostic target image.
More specifically, the evaluation value of a combination of candidate images including candidate images with image quality lower than that of the diagnostic target image may be lowered, or the evaluation value of a combination of candidate images including candidate images with image quality close to that of the diagnostic target image may be higher.

The evaluation unit 61 may also acquire supplementary information such as the purpose of examination, the name of the organ being examined, and the name of the disease being examined for each candidate image included in the combination of candidate images from a medical image database as examination information, and evaluate the candidate images based on this examination information. For example, the evaluation unit 61 may be configured to correct the evaluation value so that the evaluation value of a combination of candidate images including a candidate image having supplementary information similar to the examination information of the diagnostic target image is higher. In other words, the evaluation unit 61 corresponds to an example of an examination information acquisition means that acquires examination information for each candidate image included in multiple combinations. The evaluation value may also be multiplied by a correction coefficient determined according to the degree of match between the diagnostic target image and the supplementary information of the candidate image (a value that becomes smaller the lower the degree of match), and the value obtained by multiplying the evaluation value by this coefficient may be used as the corrected evaluation value.

(Modification 2-3: Selection Method)
In step S2020 of this embodiment, the selection unit 55 selects one combination of candidate images with the highest evaluation value calculated in step S2010, but the present invention is not limited to this. For example, in step S2020, the selection unit 55 may select multiple combinations of candidate images with evaluation values higher than a predetermined value as sets of comparison images.
In this case, the image processing unit 56 may perform a similar comparison process on each of the plurality of sets of comparison images, and calculate the same number of comparison results as the number of sets of comparison images. The above method has the advantage that the user can arbitrarily select and observe a desired comparison result from the comparison results for the plurality of sets of comparison images that meet predetermined conditions.

(Modification 2-4: Comparison Method)
In the processing of step S1050 in this embodiment, the image processing unit 56 generates a difference image or a superimposed image as a comparison process between a diagnostic target image and a set of comparison images. However, the present invention is not limited to this. For example, the image processing unit 56 may perform a deformation and registration process on each of the comparison images constituting the set of comparison images so that the anatomical structures depicted in each comparison image substantially coincide with the anatomical structures depicted in the diagnostic target image, and then deform each comparison image based on the results of the deformation and registration process to generate an image as the result of the comparison process. This has the effect of allowing the user to easily grasp the position in each comparison image constituting the set of comparison images that corresponds to the region of interest to the user in the diagnostic target image.

The present invention is not limited to this implementation. In step S1060, the display control unit 57 may display a diagnostic target image and a set of comparison images side by side on the display unit 36. This has the effect of allowing a set of comparison images selected from multiple candidate images of the same subject as images suitable for comparison with the diagnostic target image to be displayed in a format that allows for easy comparison with the diagnostic target image.

The information processing device 10 may also be configured to perform only the process of linking information identifying the selected set of comparison images with the diagnostic target image and saving it in the medical image database 23. In this case, when observing the diagnostic target image using a separate image viewer, it is possible to read the linked set of comparison images and display them in a comparable format, or to generate superimposed images or difference images between the diagnostic target image and the set of comparison images.

(Variation 2-5: Save in association)
In this embodiment, the display control unit 57 displays the comparison processing results between the diagnostic target image and the set of comparison images as the processing of step S1060. However, the implementation of the present invention is not limited to this. For example, the comparison results calculated by the image processing unit 56 in step S1050 may be stored in the medical image database 23, and the control unit 37 may acquire them through a user operation or the like. A mechanism may also be provided in which the display control unit 57 displays the comparison results acquired from the medical image database 23 through processing similar to step S1060. In this case, it is desirable for the control unit 37 to save the comparison processing results in association with the diagnostic target image acquired in step S1010 and the comparison image included in the set of comparison images selected in step S2020.

The information processing device 10 according to this embodiment performs a comparison process between a diagnostic target image and a comparison image relating to a subject on whom a user is about to perform a medical procedure, and provides the user with the results of the comparison process. As a specific example, we will explain a case where there are multiple images (hereinafter referred to as candidate images) of the same subject that were captured on dates different from the date and time the diagnostic target image was captured. The information processing device 10 according to this embodiment selects, as the comparison image, an image suitable for comparison with the diagnostic target image from among the multiple candidate images, based on information relating to the date and time the diagnostic target image and the date and time the candidate images were captured.

The overall configuration of the medical information processing system according to this embodiment is the same as that shown in Figure 4 used in the description of Example 2. However, only the details of the processing performed by the evaluation unit 61 differ from those in Example 2. Furthermore, the processing procedure executed by the information processing device 10 according to this embodiment is the same as that shown in Figure 5 used in the description of Example 2. From step S1010 to step S2000 in Figure 5, the information processing device 10 executes the same processing as in Example 2. A detailed description will not be given here.

(Step S2010)
In this step, the evaluation unit 61 executes a process of calculating an evaluation value for each of the combinations of candidate images generated in step S2000 in relation to a comparison with the diagnostic target image.

In this embodiment, the evaluation unit 61 calculates the following evaluation value in addition to the evaluation value calculated in the same manner as in step S2010 of embodiment 2 (evaluation value based on the overlapping range). That is, the evaluation unit 61 also calculates the imaging date and time (imaging date) of each candidate image included in the combination of candidate images to be evaluated, and an evaluation value based on the imaging date and time of the image to be diagnosed (evaluation value based on the imaging date and time).
Then, the evaluation unit 61 calculates an evaluation value by integrating (for example, adding or multiplying) the two evaluation values.

As a specific example of calculating an evaluation value based on the imaging date and time, a method of calculating the evaluation value based on the difference between the imaging date and time of the diagnostic target image and the imaging date and time (imaging date) of each candidate image included in the combination of candidate images can be considered. More specifically, a high evaluation value can be calculated when the period (elapsed time) between the imaging date and time of the diagnostic target image and the representative value of the imaging dates and times of the candidate images included in the combination of candidate images (the representative imaging date and time of the combination of candidate images) is long. Here, the representative value may be the average imaging date and time of the candidate images included in the combination of candidate images, or the oldest or newest imaging date and time. In this way, in step S2020 described below, a combination of candidate images with a long elapsed time based on the imaging date and time of the diagnostic target image can be preferentially selected. Note that, in addition to the conditions based on the imaging date and time of the diagnostic target image and the representative imaging date and time of the combination of candidate images, various conditions based on the imaging date and time of the diagnostic target image and the imaging date and time of the candidate images described in Example 1 can also be used. Here, if a condition is set that a predetermined treatment such as surgery does not occur between the diagnostic target images and the combination of candidate images, it is desirable to set the imaging date and time of the oldest candidate image included in the combination of candidate images as the representative imaging date and time of the combination of candidate images.

Note that the method of calculating the evaluation value by the evaluation unit 61 is not limited to the method exemplified above. For example, the evaluation unit 61 may calculate a low evaluation value when there is a large variation in the imaging dates and times (imaging dates) of the candidate images included in the combination of candidate images (for example, the interval between the earliest and latest imaging dates and times). In this case, the evaluation value does not depend on the imaging date and time of the diagnostic target image. Furthermore, the evaluation unit 61 may calculate a low evaluation value when the ratio between the imaging date and time of the diagnostic target image and the imaging date and time of the oldest candidate image and the imaging date and time of the diagnostic target image and the latest candidate image is large. In this way, combinations made up of candidate images with small differences in imaging dates and times are preferentially selected. In this way, combinations of candidate images in which the differences between the candidate images are as small as possible are preferentially selected.

Note that the conditions regarding the image capture date and time may be a combination of the above conditions.

The conditions related to the image capture date and time that the evaluation unit 61 uses to calculate the evaluation value may be preset as fixed conditions, or, as in Example 1, a condition setting unit (not shown) may be provided so that the user can set the conditions according to their purpose.

From step S2020 to step S1060, the information processing device 10 performs the same processing as in Example 2. Detailed explanation will be omitted here.

As described above, in this embodiment, the information processing device 10 has a selection unit 55 that selects a comparison image (second image) to be compared with a diagnostic target image (first image) from a plurality of candidate images. The information processing device 10 includes an imaging date and time acquisition unit 53 that acquires the imaging dates and times at which the plurality of candidate images are captured, a combination generation unit 60 that sets a plurality of combinations of the plurality of candidate images, a condition setting unit 54 that sets conditions related to the imaging dates and times, and the selection unit 55 that selects at least one combination from the plurality of combinations based on the imaging date and time conditions set by the condition setting unit 54 and the imaging dates and times of the candidate images. The selection unit 55 further selects a candidate image included in the selected combination as a comparison image (second image). The evaluation unit 61 may also acquire evaluation values for the imaging dates and times of the plurality of candidate images with respect to the imaging date and time conditions, and the selection unit 55 may select at least one candidate image from the plurality of candidate images based on the evaluation values.

The process of the information processing device 10 in this embodiment is executed according to the procedure described above.
In this embodiment, it is possible to select a plurality of comparison images that can be more suitably compared based on the imaging dates and times of the diagnostic object image and the candidate images, and to provide the user with the results of the comparison process of these images.

(Variation 3-1)
In this embodiment, the evaluation value calculation process executed in step S2010 uses the imaging date and time of the diagnostic target image and the imaging date and time of the candidate image. However, the present invention is not limited to this. For example, when generating a combination of candidate images in step 2000, a condition may be set regarding the length of the period between the imaging date and time of the diagnostic target image and the imaging date and time of the candidate image, and a combination of candidate images that satisfies the condition may be generated. Alternatively, a condition may be set regarding the variation in imaging date and time between the candidate images, and a combination of candidate images that satisfies the condition may be generated. Alternatively, a condition may be set regarding whether or not the subject underwent surgery or treatment during the period between the imaging date and time of the diagnostic target image and the imaging date and time of the candidate image, and a combination of candidate images that satisfies the condition may be generated. For example, when generating a combination of candidate images, the imaging dates and times of the multiple comparison images that make up the combination may be set to fall within a predetermined range (e.g., within one year). This makes it possible to omit the process of generating a combination of candidate images that does not satisfy the predetermined condition and the process of calculating the evaluation value, thereby improving processing efficiency.

(Variation 3-2)
The evaluation unit 61 can calculate an evaluation value based on the contrast conditions of the diagnostic target image and the candidate images. Specifically, a high evaluation value can be calculated for a combination of candidate images consisting of candidate images with the same contrast conditions as those of the diagnostic target image. Alternatively, for example, the evaluation value can be calculated based on the variation in the contrast conditions of the multiple candidate images included in the combination of candidate images. For example, when the presence or absence of contrast is used as the contrast condition, a high evaluation value can be calculated for a combination of candidate images consisting only of images with contrast (or with the same contrast phase) or a combination of candidate images consisting only of images without contrast. Furthermore, the evaluation value can be calculated based on both an evaluation based on the difference between the contrast conditions of the diagnostic target image and those of the candidate images, and the variation in the contrast conditions of the multiple candidate images included in the combination of candidate images. For example, a higher evaluation value can be calculated for a combination of candidate images consisting only of images with the same contrast conditions as those of the diagnostic target image. Furthermore, when the diagnostic target image and the candidate images are MRI images, the evaluation value can be calculated based on the type of image sequence (e.g., T1 or T2) in a manner similar to that described as an example of the contrast condition. According to these methods, a high evaluation value is calculated for a combination of candidate images consisting of candidate images whose image characteristics match those of the diagnostic target image, thereby providing the user with favorable comparison processing results.

(Variation 3-3)
In step S2010, the evaluation unit 61 may calculate the evaluation value based on the image capture date and time of each image, rather than based on the overlapping range between the candidate image and the diagnostic target image. Here, the method of calculating the evaluation value based on the image capture date and time of each image can be performed using the method described above. This has the effect of enabling more efficient processing when the evaluation value based on the overlapping range is not important, such as when it is known that the overlapping range with the diagnostic target image is approximately the same between each of the multiple combinations of candidate images generated in step S2000.

<Other Examples>
It is also possible to combine at least two of the above-described variations.

The disclosed technology may also be embodied as, for example, a system, device, method, program, or recording medium (storage medium). Specifically, it may be applied to a system consisting of multiple devices (e.g., a host computer, an interface device, an imaging device, a web application, etc.), or it may be applied to an apparatus consisting of a single device.

It goes without saying that the object of the present invention can be achieved by the following: A recording medium (or storage medium) on which software program code (computer program) that realizes the functions of the above-described embodiments is recorded is supplied to a system or device. This storage medium is, of course, a computer-readable storage medium. The computer (or CPU or MPU) of the system or device then reads and executes the program code stored on the recording medium. In this case, the program code read from the recording medium itself realizes the functions of the above-described embodiments, and the recording medium on which the program code is recorded constitutes the present invention.

10 Information processing device 21 LAN
23 Medical image database 31 Communication IF
32 ROM
33 RAM
34 Storage unit 35 Operation unit 36 Display unit 50 Control unit 51 Diagnostic object image acquisition unit 52 Comparison candidate image acquisition unit 53 Shooting date and time acquisition unit 54 Condition setting unit 55 Selection unit 56 Image processing unit 57 Display control unit

Claims (14)

An information processing device having a selection means for selecting a second image to be compared with a first image from a plurality of candidate images,
an evaluation means for setting conditions based on the imaging date and time and imaging region of the first image, acquiring the imaging date and time and imaging region of each of the plurality of candidate images, and evaluating the plurality of candidate images based on the imaging date and time and imaging region of the plurality of candidate images in response to the conditions;
the selecting means selects at least one candidate image as the second image from the plurality of candidate images based on the evaluation result of the evaluating means ;
The information processing device is characterized in that the evaluation means includes as its conditions any one of the following: the imaging date and time being a certain period or more after the imaging date and time of the first image; the imaging date and time being within a predetermined period after the imaging date and time of the first image; or the imaging date and time being within a predetermined period that is a certain period or more after the imaging date and time of the first image and has an upper and lower limit . The information processing device described in claim 1, characterized in that the evaluation means includes in the condition that the imaging area is the same as the imaging area of the first image. The information processing device described in claim 1, characterized in that the evaluation means further evaluates the plurality of candidate images based on the imaging conditions for the plurality of candidate images. The information processing device described in claim 1, characterized in that the evaluation means further evaluates the plurality of candidate images based on the resolution of the plurality of candidate images. The information processing device described in claim 1, characterized in that the evaluation means further evaluates the plurality of candidate images based on the pixel size of the slices or the slice spacing in the plurality of candidate images. The information processing device described in claim 1, wherein the evaluation means further evaluates the plurality of candidate images based on the image quality of the plurality of candidate images. The information processing device described in claim 1, characterized in that the evaluation means further evaluates the plurality of candidate images based on the noise level in the plurality of candidate images. An information processing device according to any one of claims 1 to 7, characterized in that the evaluation means obtains an evaluation value based on the evaluation results, and the selection means selects candidate images with an evaluation value higher than a predetermined value. The condition includes that the image capture date and time is within a predetermined period of time that is a period having an upper limit and a lower limit from the image capture date and time of the first image, and
2. The information processing apparatus according to claim 1, wherein the selection means selects a candidate image that satisfies the condition from among the plurality of candidate images. The information processing device described in claim 8, characterized in that the evaluation means acquires the overlapping range between the first image and each of the multiple candidate images, and further performs evaluation based on the overlapping range , and if there is a combination of multiple candidate images among the multiple candidate images where the difference in evaluation value due to the overlapping range is less than a predetermined threshold, the evaluation means evaluates so that the evaluation value of the combination of candidate images consisting of a smaller number of candidate images among the multiple candidate image combinations is higher. the evaluation means includes in the condition that the image capture date and time is the most recent image capture date and time and is a certain period of time or more after the image capture date and time of the first image,
2. The information processing apparatus according to claim 1, wherein the selection means selects a candidate image that satisfies the condition from among the plurality of candidate images. the evaluation means includes in the condition that the image capture date and time is within a predetermined period from the first image capture date and time and is the earliest image capture date and time,
2. The information processing apparatus according to claim 1, wherein the selection means selects a candidate image that satisfies the condition from among the plurality of candidate images. An information processing method for selecting a second image to be compared with a first image from a plurality of candidate images, comprising:
a step of setting conditions based on the imaging date and time and imaging region of the first image, acquiring the imaging date and time and imaging region of each of the plurality of candidate images, and evaluating the plurality of candidate images based on the imaging date and time and imaging region of the plurality of candidate images relative to the conditions;
selecting at least one candidate image from the plurality of candidate images as the second image based on a result of the evaluation ;
An information processing method characterized in that in the step of performing the evaluation, the conditions include any of the following: the image capture date and time being a certain period or more after the image capture date and time of the first image; the image capture date and time being within a predetermined period after the image capture date and time of the first image; or the image capture date and time being within a predetermined period that is a certain period or more after the image capture date and time of the first image and has an upper and lower limit . A program for causing a computer to execute the information processing method according to claim 13 .