JP2007287027A - Medical planning support system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To refer to the information of cases substantially similar in conditions and progress of diseases regardless of that observed by doctors in making an inspection plan or a treatment plan while referring to the information of past cases. <P>SOLUTION: An interest area setting means sets respective interest areas in a plurality of medical images acquired in time series. A similar case retrieving means retrieves case information including case images having areas provided with image features similar to the image features of respective interest areas in all the set interest areas. When the corresponding case information is extracted by the retrieval, inspection/treatment history information is extracted from the case information and displayed as a chart (chart 40). Further, the detailed information of each inspection item or each treatment item can be also displayed in accordance with an operator's request. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention is a system that can be used in the planning of examination plans or treatment plans in the medical field, and the examinations performed in similar cases from the case information collected in the past diagnosis and registered in the database. The present invention relates to a system for searching and providing information related to treatment.

  In hospitals and clinics, personal information and medical records of patients who have been examined are managed by recording them in medical records. Recently, medical records have been created and managed as electronic data. By doing so, it is proposed to use it for other purposes, not just as medical records.

For example, in Patent Document 1, the plan of a medical practice to be performed on a patient is based on the medical practice described in the medical records of other patients whose medical records up to a certain point are similar and the timing of the implementation. A medical support system that is automatically created is shown. Here, the medical record includes any one or more of age, sex, chief complaint, lifestyle, family history, etc. at the time of admission, medical practice record, test result, etc. .
JP-A-2001-1118014

  Each of the medical records is information input by a doctor who has performed a medical examination or interpretation. For this reason, the content may be influenced by the doctor's interpretation ability, medical examination ability, communication ability with the patient, and sentence writing ability. That is, the system of Patent Document 1 merely creates a plan of medical practice using information on cases that doctors make similar decisions, and stores information on cases where the actual medical condition of the patient is similar. I do not use it to create a plan.

  The present invention provides a function that allows selection of case information with a similar medical condition based on the patient's actual medical condition, that is, a firm fact that does not depend on human intentions. It is an object to provide a system that can be used as information.

  The planning support system of the present invention is a medical planning support system that supports examination planning and / or treatment planning planned for a predetermined patient by displaying case information registered in a database. The present invention is characterized by comprising a region-of-interest setting means, a similar case search means, and a case information display means described below.

  The region-of-interest setting means displays a plurality of medical images acquired in time series on a monitor screen, and accepts an area specifying operation for at least one medical image among the plurality of medical images. It sets the region of interest. Multiple medical images acquired in chronological order are medical images acquired at different days or at different times for the same part of the same patient. The types of modalities are not necessarily the same.

  In addition, the region of interest is a region that is noticed when interpretation is performed and has a large influence on the diagnosis by the feature of the region, and is usually a region called ROI (Region of Interest). The region of interest does not necessarily have to be a part of the medical image, and may be the whole. For example, when an enlargement tendency is seen in the lung in interpretation of a chest image, the entire lung can be set as a region of interest.

  The region of interest is usually the same as the region specified by the region specifying operation, but it is not necessarily the same. For example, a lesion contour extraction process based on image analysis may be executed in a specified region, and a region in the extracted contour may be set as a region of interest. The area specifying operation does not necessarily have to be performed on all images. For example, in the planning support system, an object represented by the area specified on the first medical image on the second medical image from the coordinate information specifying the area specified on the first medical image. Corresponding region estimation means for estimating a region representing the same object is further provided, and the region of interest setting means accepts designation of a region of interest by region designation operation for only one medical image, and coordinates information specifying the region of interest is The region of interest of each medical image may be set by supplying the region estimation means to estimate the region of interest of another medical image.

  The similar case search means includes case images each having a region having an image feature similar to the image feature in the region of interest for all the regions of interest set by the region of interest setting means by searching the database. Case information is extracted. For example, when the region of interest is set for each of the first, second, and third medical images by the region of interest setting means, an image feature similar to the image feature in the region of interest of the first medical image is provided. A first case image having a region, a second case image having a region similar to the image feature in the region of interest of the second medical image, and an image in the region of interest of the third medical image A search is performed to extract three case images of a third case image having a region with an image feature similar to the feature.

  Various search procedures by the similar case search means can be considered. For example, each region of interest set by the region-of-interest setting means has a region having an image feature similar to the image feature in the region of interest. A case image is a set of case images related to the same patient among case images detected by searching for a case image, and each of the regions of interest is similar to an image feature in the region of interest. A set including case images having a region having an image characteristic to be identified, and case information including case images belonging to the identified set is supplied to the case information display means as case information extracted by searching the database The procedure can be considered.

  Alternatively, for a region of interest of one medical image set by the region-of-interest setting means, a case image having a region having an image characteristic similar to that of the region of interest is searched, and extracted by searching for the case image A case image including an image feature similar to the image feature in the region of interest of the medical image other than the one medical image is further searched using the case information including the case image as a search range, and the search The case information obtained by extracting the case image having a region having an image feature similar to the image feature in the region of interest of the medical image from all the medical images other than the one medical image is extracted by the search. A procedure of supplying the information to the case information display means as the case information that has been made is also conceivable.

  The case information display means supports the planning of examination plans and treatment plans by extracting and displaying the examination and / or treatment history information from the case information extracted by the similar case retrieval means. In the system according to the embodiment of the present invention, the case information display means includes processing for extracting history information of examination and / or treatment from one case information among the case information extracted by the similar case search means, and extraction A process of associating the timing of the examination and / or treatment by comparing the examination item and / or treatment item included in the history information with the examination item and / or treatment item that has been performed on the patient; Based on the association of the execution time, the process for determining the recommended execution time of the test item and / or treatment item that has not been performed for the patient, and the recommended test item and / or treatment item that has not been performed are recommended By executing the processing to be displayed together with the information of the execution time, the planning of the inspection plan and the like is supported.

  The planning support system further includes time interval setting means for setting the acquisition time intervals of a plurality of medical images, and the similar case search means is configured to search all the interests set by the region of interest setting means by searching the database. Each of the regions includes a case image having a region having an image feature similar to the image feature in the region of interest, and the acquisition time interval of the case image is a time interval set by the time interval setting means If the same case information is searched, it is possible to extract cases that have similar disease progression speeds as well as similar changes in lesions (for example, enlargement, discoloration, etc.). This supports the planning of a plan that uses the closer case as reference information.

  The planning support system of the present invention includes a region-of-interest setting unit that can set a region of interest for each of a plurality of images acquired in time series, and all regions of interest set by the region-of-interest setting unit are similar in a batch. Similar case search means is provided for searching, so that temporal changes in image features can be used as similar case search queries, and information on cases with similar disease courses is acquired and provided by the search. be able to. In particular, in the form in which the time interval at which images are acquired is also set as a query for similar case search, it is possible to search for cases that have similar disease progression rates as well as changes in disease states, and extract cases that have very similar examination calendars. be able to. Since the image characteristics of the image obtained by photographing the patient are firm information that does not depend on the ability or action of the doctor, if the system of the present invention is used, With reference to the information, an examination plan and a treatment plan can be made.

  FIG. 1 shows a schematic configuration of a medical information processing system in which a planning support system according to an embodiment of the present invention is introduced. As shown in the figure, this system includes an imaging apparatus (hereinafter referred to as modality) (M) 1, a laboratory technician workstation (WS: WorkStation) 2, a radiologist workstation (WS) 3, and a diagnostic physician workstation (WS). ) 4, an image information database 5, an interpretation report database 6, a medical record information database 7, and a database management server 8 (DBMS: DataBase Management Server) for managing these three types of databases. Connected so that they can communicate with each other. Although FIG. 1 shows a configuration in which only one device is provided for convenience of explanation, a plurality of types of modalities are connected as a modality 1 to the network 9. Also, it is assumed that there are a plurality of inspection technician workstations 2, radiologist workstations 3, and examination physician workstations 4 according to the number of examination technicians, radiologists, and examination physicians.

  In the present embodiment, the network 9 is a local area network that connects various apparatuses in a hospital. However, when the medical doctor workstation 4 is also installed in another hospital or clinic, the network 9 may be configured by connecting local area networks of each hospital via the Internet or a dedicated line. In any case, it is desirable that the network 9 can realize high-speed transfer of image information such as an optical network.

  The modality 1 is a device that generates image data by photographing a subject, adds incidental information defined by the DICOM standard to the image data, and outputs the image information as image information. In this specification, a set of image data representing a subject and accompanying information of the image data is referred to as “image information”. That is, the “image information” includes text information related to the image. The modality 1 is a type of modality that collectively performs processes from subject photographing to image data generation, an imaging device that records a captured image on a recording sheet, and a reading device that reads the image from the recording sheet and generates image data. May be a separate type of modality. Specific examples include an X-ray simple imaging apparatus, CT (Computed Tomography), MR (Magnetic Resonance), PET (Positron Emission Tomography), and an ultrasonic imaging apparatus.

  The inspection technician workstation 2 includes a general-purpose processing device (computer), one or two high-definition displays, and input devices such as a keyboard and a mouse. The processing apparatus incorporates software for supporting the work of the laboratory technician. The inspection engineer's workstation 2 receives the DICOM-compliant image information from the modality 1 by the function of the software program, and displays the image data included in the received image information and the contents of the incidental information on the screen. Encourage engineers to confirm. Then, the image information confirmed by the inspection engineer is transferred to the DBMS 8 via the network 9 and the registration of the image information in the image information database 5 is requested.

  The DBMS 8 is a general-purpose computer having a relatively high processing capability and incorporating a software program that provides the functions of the DBMS. The DBMS 8 stores data requested for registration in a large-capacity storage connected to the DBMS 8, and extracts and provides data that meets the search conditions when there is a browsing request for the registered data. However, the data storage location may be NAS (Network Attached Storage) or SAN (Storage Area Network) connected to the network 9.

  When the DBMS 8 receives a registration request for image information, the DBMS 8 arranges the image information in a database format and registers it in the image information database 5. FIG. 2 shows an example of image information registered in the image information database 5.

  The image information 10 is managed as XML or SGML data, for example, and as illustrated in the figure, image data representing the subject image, at least a patient number for identifying the subject, an examination number for identifying the examination, And supplementary information including an image number for identifying the image. As the incidental information, in addition to this, a unique ID (UID) assigned for each image information, the examination date when the image information was generated, the examination time, and the type of modality used in the examination for obtaining the image information , Patient information such as patient name, age, gender, examination site (imaging site), imaging conditions (contrast agent used / used dye, radionuclide, radiation dose, etc.) Multiple images in one examination Information such as the series number or collection number at the time of acquisition is included.

  When the DBMS 8 receives a browsing request from the radiologist workstation 3 via the network 9, the DBMS 8 searches the image information registered in the image information database 5, and extracts the extracted image information from the requesting radiologist. Sent to workstation 3

  The radiologist workstation 3 is an apparatus used by the radiologist for interpretation of images, and includes a processing device, one or two high-definition displays, and input devices such as a keyboard and a mouse. This processing apparatus incorporates a function for supporting diagnosis by a radiologist, for example, a function for automatically detecting and highlighting a portion that appears to be a lesion in an image, and a function for supporting the creation of an interpretation report. When an operation for requesting browsing of an image to be interpreted or other information used for diagnosis is performed on the interpretation screen, the radiologist workstation 3 transmits a viewing request to the DBMS 8 to obtain image information necessary for interpretation or diagnosis. get. Then, the image information is displayed on the monitor screen, and an automatic lesion detection process is executed in response to a request from the radiologist.

  Furthermore, the radiologist workstation 3 displays a report creation screen for supporting the creation of an interpretation report on the monitor, and text indicating the contents of the diagnosis (findings, etc.) based on the interpretation is input from the radiologist. In some cases, an interpretation report is generated in which the input information and an image to be interpreted (hereinafter referred to as an interpretation target image) are recorded. When there are a plurality of interpretation target images, a representative image (hereinafter referred to as a representative image) having the greatest influence on the diagnosis is recorded in the interpretation report. The radiologist workstation 3 transfers the generated interpretation report to the DBMS 8 via the network 9 and requests registration of the interpretation report in the interpretation report database 6. When the DBMS 8 receives the interpretation report registration request, the DBMS 8 arranges the interpretation report in a database format and registers it in the interpretation report database 6. FIG. 3 shows an example of an interpretation report registered in the interpretation report database 6.

  The interpretation report 11 is managed as, for example, XML or SGML data, and is acquired by referring to image data representing an interpretation target image or a representative image and supplementary information of the image information at the time of interpretation of the image, as illustrated in the figure. It consists of information such as examination number, patient number, and image number, and information indicating findings and certainty of findings. Of these, the image data may be a copy of the image data included in the image information 10 as it is, but the reduced image data having a smaller number of pixels (thinned out) than the image data included in the image information 10. But you can. The image data may be link information indicating a storage location (folder or the like) of the image data in the image information database 5 and a file name. The image interpretation report 11 preferably records coordinate information indicating the ROI set for the image data at the time of image interpretation together with the image data. Note that the coordinate information indicating the ROI may be recorded not only in the image interpretation report 11 but also in the accompanying information of the image information 10.

  When the DBMS 8 receives a browsing request from the diagnostic medical workstation 4 via the network, the DBMS 8 searches the interpretation report registered in the interpretation report database 6, and extracts the extracted image information to the requesting diagnostic medical workstation 4. Send.

  The examiner's workstation 4 is an apparatus used by the examiner to refer to the interpretation report and the image that is the subject of the interpretation, and inputs such as a processing device, one or two high-definition displays, and a keyboard / mouse. Consists of equipment. This processing device includes a medical record information creation function for recording the history of examinations, examinations, and treatments (medications, etc.) of each patient as electronic data, and an order system for radiology by creating examination order information. The inspection order function to transmit to (not shown) is incorporated.

  The medical examination workstation 4 creates new medical record information recording patient information and the purpose of medical examination at the first visit of the patient, transfers it to the DBMS 8 via the network 9, and registers the medical record information in the medical record information database 7. Request. When the DBMS 8 receives the chart information registration request, the DBMS 8 arranges the chart information in the database format and registers it in the chart information database 7.

  Further, when an examination is ordered by the examination order function in the course of the examination, the examination medical workstation 4 adds the examination number, the examination date, and the like notified from the radiology system to the medical chart information. Further, when the examination is completed and an interpretation report 11 is created by the radiologist and the interpretation report 11 is referred to by the doctor, some information recorded in the interpretation report (for example, findings information of the radiologist) ) Is added to the chart information.

  On the other hand, information entered by the examiner himself / herself is added to the chart information as needed. For example, in the case of a test that does not involve imaging, such as a blood test, finding information input by a doctor referring to the test result is added to the chart information. In addition, for the record of medication and other treatments, information input by the doctor is added to the chart information. The medical examination workstation 4 transfers the updated chart information or the difference information indicating only the changed part to the DBMS 8 and requests the update of the corresponding chart information.

  When the DBMS 8 receives an update request for the chart information, the DBMS 8 searches the chart information database 7 to call the corresponding chart information on the memory, updates the contents on the memory, and writes the updated chart information to the storage. Thereby, the contents of the medical chart information database 7 are updated.

  FIG. 4 shows an example of chart information registered in the chart information database 7. The medical record information 12 is managed as XML or SGML data, for example. As described above, the patient information registered at the first visit, the patient's reception history, examination history, treatment history including medication, etc. Information necessary for diagnosis and treatment such as allergies is recorded. In the present embodiment, it is assumed that the chart information 12 includes at least two pieces of information, that is, a patient number and an examination number that identifies examinations performed in the past. The chart information 12 illustrated in the figure does not include image data, but records reduced image data similar to the interpretation report 11 and link information to the image information as one item of the inspection history. Also good.

  Here, as described above, the image information registered in the image information database 5 and the interpretation report registered in the interpretation report database 6 are associated with each other in the course of image diagnosis by the radiologist workstation 3. The interpretation report registered in the interpretation report database 6 and the chart information registered in the chart information database 7 are associated with each other in the process of creating and updating the chart information by the clinician workstation 4. As a result, three types of information, image information, image interpretation report, and medical record information, are associated with each other.

  FIG. 5 is a diagram illustrating data used for associating three types of information and an associating method. In the present embodiment, as shown in the figure, the image information 10a relating to the simple captured image, the image information 10b to 10e constituting the series image obtained by the CT examination, and the image interpretation reports 11a and 11b are all for each examination. Contains information on the assigned examination number. Therefore, when it is desired to obtain an interpretation report of given image information, the interpretation report can be obtained by searching the interpretation report database 6 using the examination number as a search key.

  On the other hand, in the interpretation reports 11a and 11b, the image numbers of the images to be interpreted are recorded. For this reason, when it is desired to obtain the image information referred to when creating the interpretation report from a given interpretation report, the image information can be obtained by searching the image information database 5 using the image number as a search key. Alternatively, when link information to the storage location of image information is recorded in the image interpretation report, it is possible to immediately acquire the image information referred to when creating the image interpretation report by following the link information. The same applies when a plurality of image numbers or link information is recorded in one interpretation report.

  If it is necessary to obtain all images constituting the series from an interpretation report in which only the image number of the representative image in the series image is recorded, such as the interpretation report 11b illustrated in the figure, If the image information database 5 is searched using the number as a search key, all the image information constituting the series can be acquired.

  Furthermore, as shown in the figure, each of the image interpretation reports 11a and 11b and the medical record information 12 includes information on a patient number and an examination number. Therefore, when it is desired to obtain medical chart information from a given interpretation report, the corresponding medical chart information can be obtained by searching the medical chart information database 7 using the patient number as a search key. On the other hand, when it is desired to obtain examination information recorded in the chart information 12 from the chart information 12, the examination report database 6 and further the image information database 5 are retrieved by using the examination number as a retrieval key. Obtained image information and a created interpretation report can be acquired.

  That is, in the present embodiment, the image information 10, the interpretation report 11 and the chart information 12 are registered in separate databases for data management, but are associated with each other. By specifying the image interpretation report 11 or one piece of medical record information 12, all case information related thereto can be acquired and referenced. Hereinafter, this set of information associated with each other is referred to as case information 13.

  Case information can be specified by image information, patient number, or examination number as described above, but it is not always preferable from the viewpoint of personal information protection to be able to specify a patient when referring to a case. It can not be said. For this reason, in the present embodiment, a case number is assigned to the case information 13 separately from the patient number, and if necessary, a person who refers to the case information can distinguish a case by only the case number without knowing the patient number. I am doing so. The case number is recorded in the electronic medical record information 12.

  Hereinafter, the realization method of the similar case search function provided as one of the diagnosis support functions of the radiologist workstation 3 will be further described.

  FIG. 6 is a diagram showing an example of the configuration of the radiologist workstation 3 and the DBMS 8 for realizing the similar case search function. As described above, the radiologist workstation 3 includes a number of diagnosis support functions in addition to the similar case search function, but the description and illustration thereof are omitted here.

  As shown in the drawing, in this embodiment, an input device 14 such as a mouse and a monitor 15 are connected to the radiologist workstation 3. The radiologist workstation 3 also includes a region-of-interest setting means 21 for setting a region image (hereinafter referred to as a query image) to be a query at the time of case search, and case information for controlling display output of case information extracted by the search. Display means 22 is provided. The DBMS 8 searches for case information based on the query image supplied from the region-of-interest setting means 21 of the radiologist workstation 3, and extracts the extracted case information as the case information display means 22 of the radiologist workstation 3. The similar case search means 23 to transfer to is provided.

  In the present embodiment, the functions of the region-of-interest setting unit 21, the case information display unit 22, and the similar case search unit 23 are all realized by the CPU of each device executing the processing specified in the program. However, the functions of the region-of-interest setting means 21, the case information display means 22, and the similar case search means 23 can also be realized by hardware such as a dedicated processor that executes only the processing performed by these means.

  In addition to this, a configuration as shown in FIG. 7 is also conceivable as a configuration for realizing the similar case search function. That is, the similar case search means 23 may be on the side of the radiologist workstation 3. However, in the configuration of FIG. 7, data transfer via the network 9 occurs each time the similar case search means 23 refers to each database. Therefore, the configuration shown in FIG. 6 is preferable in terms of network transmission efficiency and overall system processing efficiency.

  FIG. 8 is a flowchart showing an outline of processing of the region-of-interest setting means 21. When an operation for requesting display of similar cases to the radiologist workstation 3 is performed, the region-of-interest setting unit 21 displays a search condition setting screen on the monitor 15 (S101). An input frame for inputting an image number assigned to each image is displayed on the search condition setting screen. Alternatively, a list of reduced images of selectable images is displayed. The region-of-interest setting unit 11 receives information input to the input frame and a selection operation for the reduced image list (S102).

  Subsequently, the region-of-interest setting unit 21 accesses the image information database 5 and acquires the image information 10 including the designated image. Then, by referring to the modality information (illustrated in FIG. 2) of the auxiliary information of the image information 10, the modality that generated the image data of the image information 10 is determined (S103).

  If the image selected in step S102 is the first selected image (S104), the region-of-interest setting unit 11 outputs the image data of the acquired image information 10 to the screen of the monitor 15, and returns the ROI. The designation operation is accepted (S105).

  If the corresponding image information 6 has already been acquired by the operation before the display of the similar case is requested, and the ROI is set for the image data included in the image information, the region of interest setting is performed. The means 21 displays the medical image on the screen of the monitor 15 together with a frame indicating the set ROI. At this time, the ROI may be corrected as necessary by accepting an ROI resetting operation.

  Subsequently, the region-of-interest setting unit 21 calculates a plurality of types of feature amounts representing the features of the shadow included in the ROI set in step S105 (S106). The feature amount to be calculated includes the size (diameter) of the shadow, a luminance histogram, and the like, but the type of the feature amount that can best represent the feature of the shadow differs for each type of shadow (that is, disease). For this reason, the region-of-interest setting means 21 performs image analysis on the set ROI to determine the type of shadow (for example, a mass, a line, a honeycomb, etc.), and features suitable for identifying each type of shadow After selecting the amount, the feature amount is calculated.

  Next, the region-of-interest setting means 21 displays a message for inquiring whether or not to select and set the next query image on the monitor screen (S107). At this time, an operation button (for example, a YES button and a NO button) for inputting an answer is also displayed on the monitor screen together with a message. In step S107, the region-of-interest setting unit 21 returns to step S102 and accepts a next image selection operation when an operation indicating that the next setting is performed is performed.

  In the present embodiment, the region-of-interest setting means 21 executes the ROI estimation process in step S110 instead of the operation in step S105 in the second and subsequent image selection (S104). The ROI estimation process is a process of estimating the ROI of the second image based on the ROI set in step S105 for the first image.

  FIG. 9 is a diagram for explaining the ROI estimation processing in step S110. For example, as illustrated in FIG. 9A, the selected first image is an X-ray simple photographed image, and the image has a predetermined vertical width V and horizontal width based on the coordinate point (xa, ya). Assume that the ROI of H is specified. In this case, if the selected second image is an X-ray simple photographed image like the first image, the region having the same vertical width V and horizontal width H is set as the second point with the same coordinates (xa, ya) as a base point. Automatically set on the image. Alternatively, after performing alignment by the method disclosed in Japanese Patent Application Laid-Open No. 2002-092589, regions having the same vertical width V and horizontal width H may be automatically set on the second image.

  Further, when the first image is an X-ray simple photographed image and the second image is a series image obtained by CT examination as illustrated in FIG. Based on the included series number and collection number, it is determined at which position of the slice positions (ya, yb, yc) shown in FIG. Then, the coordinate system defined on the X-ray simple photographed image is roughly associated with the coordinate system defined on the CT series image, and the ROI is automatically set on the series image. Another technique for associating regions representing the same object on different images is also shown in, for example, Japanese Patent Application Laid-Open No. 2004-180932 (however, this publication sets an ROI as a query image for similar case search) The detailed description is omitted.

  Here, since the images acquired in time series are selected in step S102, the subject and the ROI should normally be the same, but the change in the size of the ROI is accompanied by the temporal change in the size of the lesion. It may be necessary. Therefore, in the present embodiment, in the ROI estimation process in step S110, a frame indicating the estimated ROI is displayed on the monitor screen so as to overlap the selected image, and is necessary by accepting an enlargement / reduction operation for the displayed frame. Accordingly, the size of the ROI can be adjusted.

  Note that the ROI designation operation acceptance process in step S105 may always be performed regardless of whether the image is the first image or the second and subsequent images (the determination in step S104 is not performed). That is, the ROI estimation process in step S110 is a process for reducing the burden of an operation for designating the ROI on the screen, and is not an essential process in the similar case function.

  Hereinafter, the description will return to FIG. 8 again. If an operation indicating that the next image selection is not performed is performed in step S107, the region-of-interest setting unit 21 calculates a time interval at which the selected image is acquired (S108).

  In order to calculate the time interval, first, among the supplementary information of the image information to which each image belongs, the information on the examination date and examination time illustrated in FIG. Sort in chronological order based on date and inspection time. In the present embodiment, the images are arranged in the order from the latest examination date. Subsequently, by calculating the difference between the values of the inspection dates and times of adjacent images, the time interval at which the plurality of images are acquired is obtained.

  Then, search condition information is generated using the time interval calculated in step S108 and the feature amount calculated in step S106 as a set of search conditions (S109). However, when the time interval is set as a search condition, the difference time obtained by calculation is set after being changed to a predetermined time unit. In this embodiment, the time interval is set in units of half a month. For example, the calculated time interval is 0.5 months when 18 days and 15 hours and 20 minutes, January when 35 days and 3 hours and 13 minutes, and 2.5 months when 72 days and 2 hours and 1 minute. In this way, the half-month time interval closest to the calculated time interval is set.

  The search condition information generated in step S109 is transferred to the DBMS 8 as indicated by the arrow in FIG. 6 and supplied to the similar case search means 23.

  FIG. 10 is a diagram in which images and ROIs set in the region of interest setting process are arranged in order of acquisition date and time. The figure shows a chest CT image with an examination date of March 24, 2006, a chest X-ray image with an examination date of December 20, 2005, and an examination date of April 10, 2005 in step S102 of FIG. A case where a chest simple X-ray image of the day is selected and a rectangular ROI as illustrated in the figure in step S105 or S110 is set as an example. In this example, the image acquisition time interval is 3 months and 8 months. In addition, the set ROI is assumed to be set larger for an image with a newer acquisition date because the shadow with a suspected lesion gradually increases.

  FIG. 11 illustrates search condition information 16 generated by the region-of-interest setting means 21 when the three images and ROI illustrated in FIG. 10 are set. In the present embodiment, when a plurality of query images are set, the region-of-interest setting unit 21 generates search condition information 16 including image acquisition time interval information 161. The image acquisition time interval information 161 is set, for example, as a numerical value separated by commas as shown in the figure. In the example shown in FIG. 10, since the image acquisition time intervals are 3 months and 8 months, as shown in the figure, the acquisition time interval information 161 is set to “3, 8”.

  The search condition information 16 further includes the modality information 162 that generated the image data and the feature value 163 calculated in step S106 of FIG. These pieces of information are arranged in such an order that the inspection dates are in ascending order or descending order. The figure shows an example in which these pieces of information are set in the order in which the examination dates are in descending order.

  In this embodiment, the region-of-interest setting unit calculates the time interval and supplies the time interval information to the similar case search unit 23. However, the calculation of the time interval and the setting as the search query are similar case search. You may carry out by the means 23 side. If the search date / time information is set for each query image instead of the time interval information 161 in the search condition information 16, the similar case search means 23 that has received the search condition information 16 calculates the difference in the test date / time. By doing so, it is possible to obtain the image acquisition time interval and execute the search using the obtained time interval as a search query.

  Hereinafter, the process of the similar case search means 23 will be described. When the search condition information 16 includes only one set of query image information, the similar case search means 23 calculates the feature value of the query image and each image information registered in the image information database 5. By applying the calculated feature value to a predetermined arithmetic expression, the similarity of the image features in the region of interest of the two images is obtained, and image information of an image having a high similarity is extracted.

  Subsequently, the similar case retrieval unit 23 uses the association described with reference to FIG. 5 to interpret the interpretation report, medical record information, and further, the medical record information associated with the extracted case image by searching the database. Get other associated interpretation reports and image information. That is, the case information 13 is acquired and transmitted to the case information display means 22.

  When there are two or more sets of query image information included in the search condition information 16, the similar case search unit 23 acquires case information including similar images for all query images.

  For example, if the three query images illustrated in FIG. 10 are set, the image within the ROI (hereinafter referred to as the ROI image) of the X-ray simple photographed image acquired on April 10, 2005 and the image features are the same. An ROI is applied to an object having similar image characteristics to an X-ray simple image obtained by setting an ROI for a similar object and an ROI image acquired on December 20, 2005. An X-ray CT image in which an ROI is set for an object having similar image characteristics to the set X-ray simple image and the ROI image acquired on March 24, 2006. Case information 13 including three types of images is acquired. This means that case information 13 of cases with similar pathological conditions observed from images is acquired. The acquired case information is transmitted to the case information display means 22.

  Hereinafter, two procedures will be exemplified as a search processing procedure when a plurality of query images are set. However, other search methods are conceivable, and the similar case search means 23 may use any of the methods to perform the search.

  12 and 13 are diagrams for explaining the first search processing procedure, FIG. 12 is a conceptual diagram of the processing, and FIG. 13 is a flowchart. In the first search processing procedure, first, the search range is set as image information in the image information database 5, and the first query image set in the search condition information 16 is set as a search query. Search for similar images. More precisely, image information including image data representing such an image is searched and extracted from the image information database 5 (S201). In the following description, “image information including image data representing an image” is abbreviated as “image” as necessary.

  The image search process in step S201 is performed until there is no query image to be set as a search query regardless of whether or not the corresponding image information is extracted by the search (S202). That is, the image search process in step S201 is performed for all of the plurality of query images set in the search condition information.

  FIG. 12 shows an ROI image (hereinafter referred to as a first query image) of an image acquired on March 24, 2006 for a patient with a patient number 4000 (hereinafter referred to as a patient 4000), as a search query. In the first image search process, four sets of X-ray CT images of four patients photographed on different days are extracted, and ROI images (hereinafter referred to as second images) of images acquired on December 20, 2005. In the second image search process performed as a search query, three X-ray simple images of three patients taken on different days were extracted. On April 10, 2005, Four X-ray single images of four patients photographed on different days are obtained by the third image retrieval process using the ROI image (hereinafter referred to as a third query image) of the photographed image as a retrieval query. Shows an example in which captured image is extracted. In the present embodiment, since the similarity / dissimilarity of the image features is determined based on the feature amount calculated from the query image, that is, the ROI image, even when the position of the ROI is different as illustrated in the figure. The image information of images having similar image characteristics of the ROI image is extracted by the search.

  Next, the similar case search means 23 classifies the search result of step S201, that is, the extracted image based on the patient number (S203). In the example of FIG. 12, since three pieces of image information are extracted for the patient 2501, these images are classified into the same set. Also, since two pieces of image information are extracted for the patient 3254, these images are also sorted into the same set. Similarly, other images are classified into sets for each patient.

  Subsequently, the similar case search means 23 searches for images with similar ROI image characteristics, with the search range as each set defined in step S203 and the search query as the first query image (S204). When an image having similar image characteristics is extracted (S205), it is determined whether or not another query image is set as a search condition in the search condition information (S206), and is further set in the search condition information. The value of the acquisition time interval between the previous query image and the query image is referred to, and the image search processing in step S204 is repeated using the query image and the acquisition time interval as a search query. However, since the acquisition time interval is added as a search query in the second and subsequent image search processing unlike the first image search processing in step S204, images having different acquisition time intervals are extracted even if the images are similar. Not.

  The acquisition time interval is compared in units of time such as one week, one month, or one year. In the present embodiment, as described above, the acquisition time interval is set in half-month units in the search condition information 16, so whether or not the acquisition time intervals are the same is also determined in half-month units. In the example of FIG. 12, the examination date of the patient 2501 is May 9, 2002, February 2, 2002, and June 7, 2001, and the image acquisition time intervals are 3 months and 8 months. Since this is the same as the query image acquisition time interval, it is determined in step S205 that image information having the same acquisition time interval exists.

  When image information that matches two conditions of the query image and the acquisition time interval is extracted (S205), it is determined again whether another query image is set as a search condition (S206). If there is no query image to be used as the search query, it means that image information corresponding to all the query images set in the search condition information has been extracted. Therefore, the similar case search means 23 is extracted at this time. The patient number included in the accompanying information of the image information is referred to and recorded in the memory of the DBMS 8 (S207). For example, in the example of FIG. 12, three sets of images surrounded by a broken line are extracted in the processing of steps S204 to S206 of FIG.

  The similar case search means 23 determines whether or not there is a next group (S208), and if there is a next group, repeats the processing of S204 to S207. For example, in the example of FIG. 12, image search processing for image information included in the group of patients 3254 is performed next. In the group of patients 3254, an image corresponding to the first query image ( The acquisition time interval of the image acquired on December 13, 2001) and the image corresponding to the second query image (acquired on February 14, 2006) is as long as 4 years and 2 months. In the image search process in step S204 performed as described above, an image acquired on February 14, 2006 is not extracted. Therefore, it is determined that there is no corresponding image information for the group of patients 3254 in step S205, and no patient number is recorded in step S207.

  When there are no more pairs to be subjected to the image search processing in step S204 (S208), the similar case search means 23 sets the search range to all of the image information database 5, the interpretation report database 6, and the medical record information database 7, and the search query. Is set to the patient number recorded in the memory in step S207, and the case information 13 is searched (S209). Specifically, first, the medical record information database 7 is searched to obtain the medical record information of the corresponding patient, and as described with reference to FIG. 5, the interpretation number report in the medical record information is used to further interpret the interpretation report. By searching the database 6 and the image information database 5, all case information managed by the chart information is acquired.

  The similar case search means 23 outputs the case information acquired by the search in step S209 as a search result of the similar case search. The outputted case information is transferred to the radiologist workstation 3 and supplied to the case information display means 22.

  14 and 15 are diagrams for explaining a second search processing procedure of the similar case search means 23, FIG. 14 is a conceptual diagram of the processing, and FIG. 15 is a flowchart. The second search processing procedure is the same as the first search processing procedure until the image information database 5 is first set as a search range and an image search is performed using the first query image as a search query. However, the above-described first search processing procedure is to search the image information database 5 until there is no query image that can be set as a search query regardless of whether or not the corresponding image information has been extracted. On the other hand, in the second search processing procedure, if no corresponding image information (similar image) is extracted in the first search (S302), a search result indicating no corresponding image information is output (S309). Exit. Even if it is determined in step S302 that the corresponding image information has been extracted, the image information database 5 is not searched using the second and third query images as search queries.

  In the second search processing procedure, when the corresponding image information is extracted in the processing of steps S301 and S302, the similar case search means 23 acquires the examination number included in the incidental information of the image information, and the examination number The case information is searched by using as a search query (S303). Specifically, by using the association illustrated in FIG. 5 and searching the interpretation report database 6 using the examination number as a search query, first, an interpretation report of the image is obtained and further recorded in the interpretation report. The medical record information of the patient is acquired by referring to the patient number and searching the medical record information database 7 using the patient number as a search query. Then, the image interpretation report database 6 is searched again using another examination number recorded in the medical record information as a search query, and another image interpretation report associated with the medical record information is acquired. Further, the image information database 5 is searched to acquire image information of the image to be interpreted in other image interpretation reports. Thereby, the similar case search means 23 acquires the case information to which the image information extracted in step S302 belongs.

  Subsequently, the similar case search means 23 searches for an image using the case information collected and acquired in step S303 as a search range and using the second query image and the time interval as a search query (S304). When relevant image information is extracted by this search (S305), it is determined again whether another query image is set as a search condition (S306). If there is no query image to be used as a search query, it means that image information corresponding to all the query images set in the search condition information is extracted. For this reason, the similar case search means 23 refers to the patient number included in the supplementary information of the image information extracted at this time, and records it in the memory of the DBMS 8 or the like (S307). For example, in the example of FIG. 14, three pieces of image information surrounded by a solid line are extracted in the processes of steps S304 to S306 of FIG.

  The similar case search means 23 determines whether there is any other image information extracted in step S301 (S308), and repeats the processing of S304 to S307 if there is a next image. For example, in the example of FIG. 14, image search processing for image information included in the case information of the patient 3254 is performed next. However, in the case of the patient 3254, the acquisition time interval between the image corresponding to the first query image (captured on December 13, 2001) and the image corresponding to the second query image (captured on February 14, 2006). Is not three months, image information generated on February 14, 2006 is not extracted. For this reason, it is determined that there is no corresponding image information in step S305, and no patient number is recorded in step S307.

  When the similar case search means 23 determines that there is no other image information extracted in step S301 (S308), it outputs the case information extracted in the case search process so far as a search result (S309). The outputted case information is transferred to the radiologist workstation 3 and supplied to the case information display means 22.

  Next, screen control performed by the case information display unit 22 will be described with reference to screens to be displayed. In the present embodiment, the case information display means 22 divides a window controlled by the operating system into a plurality of frames, and displays the case information supplied from the similar case search means 23 in each frame in a predetermined layout. . As a display method of the case information, a method of displaying all the information supplied from the similar case search means 23 is also conceivable. However, the case information display means 22 of the present embodiment includes a list display screen showing an outline of the search results, Display control is performed so that the acquired case information is displayed in stages by switching between two types of screens, a detail display screen showing details of the search results.

  FIG. 16 shows a layout example of the list display screen 17. In the example shown in the figure, the window is divided into four frames, three in the upper row and one in the lower row. In the upper left frame, the subject information 34 represented by the image used as the search query is searched, and the upper frame in the middle is searched. The medical image 31 with the latest examination date among the images used as the query, the past images 32 and 33 with the old examination date among the images used as the search query in the upper right frame, and the lower frame are searched. A list 35 of similar cases extracted by the above is arranged. In the present embodiment, an operation button 37 for calling up a treatment plan planning support screen is arranged in the lower frame.

  In the present embodiment, since the information displayed in the upper frames of the list display screen is information set by the region-of-interest setting unit 21, the case information display control unit 22 directly receives the information from the region-of-interest setting unit 21. Is acquired and arranged at a predetermined position of each frame. The case information display means 22 controls the upper center frame and the upper left frame so that image information constituting the case information is arranged and displayed at a predefined position. For the upper right frame, control is performed so that the past image is arranged at a predetermined position in the initial stage, and then a selection operation for each past image being displayed is accepted. When a selection operation with a pointing device or the like is detected, each selected image is enlarged and displayed in the upper middle frame, and the image currently displayed in the upper middle frame is reduced and displayed in the upper right frame. Change the display position and image size.

  For the lower frame, control is performed so that case numbers that can be used as case identifiers are extracted from the chart information of each case information supplied from the similar case search means 23, and the case numbers are displayed in a list in the frame. I do. In this list, only the case number is displayed as a selection item of the list box or combo box, but in the present embodiment, as shown in the figure, the patient number extracted from the chart information is also displayed, and further, the chart information or A reduced image of the medical image recorded as the representative image in the interpretation report is displayed together with a frame indicating the ROI. At this time, if the number of cases to be displayed is large, only a part of cases are displayed, and a case list can be browsed by installing a horizontal scroll bar in the lower frame.

  Then, a selection operation in units of a case number, a patient number, and a reduced image enclosed by a frame 36 in the figure is accepted, and when a selection operation by a pointing device or the like is detected, a second window is opened. The detailed display screen is controlled at.

  For the lower frame, the case information display means 22 also accepts a selection operation for the operation button 37, opens a third window when a selection operation is detected, and provides a treatment plan support function in the third window. To do.

  Next, FIG. 17 shows a layout example of the detail display screen 18. In the example shown in the figure, the window is divided into two left and right frames. In the right frame, subject information represented by an image used as a search query and images 31, 32, and 33 used as a search query are arranged. The left frame shows an example in which detailed information of cases selected on the list display screen 17 is arranged. The figure exemplifies a detailed display screen when a case number 1005 surrounded by a frame 36 is selected on the list display screen 17, and is not limited to an image used as a search query, but case information of the case number 1005. All the inspection information recorded in is arranged in a time series in order from the top of the screen.

  The case information display means 22 extracts all image information and an interpretation report related to the image information from the case information specified by the selected case number for the left frame, and at least a part of the entire image information. Control is performed so that the items are displayed on the screen in time series together with at least a part of the interpretation report corresponding to the image information. In the example of the figure, three items of image data, examination date, and modality are displayed for image information, and two items of examination purpose and findings are displayed for an interpretation report.

  The arrangement order of the inspection result information 38a to 38e including the set of image data, inspection date, modality, inspection purpose, and findings is determined based on the inspection date information so that the inspection date is in ascending order or descending order. In addition, when there is a lot of examination information to be displayed, only a part of the examination information is displayed, and a vertical scroll bar is installed as shown in the figure to view all examination information included in the case information. Display control is performed so that

  Further, the case information display means 22 highlights the examination result information including the image used for similarity determination by the similar case retrieval means 23 when displaying the examination result information in time series. For example, the figure exemplifies a screen that emphasizes the inspection result information by displaying a frame 39 surrounding the inspection result information for the inspection result information 38b, 38c, and 38d including the image contrasted with the query image. ing.

  FIG. 18 is a diagram illustrating a treatment plan planning support screen 19 that is called when the operation button 37 of the list display screen 17 is operated in the present embodiment. As shown in the figure, the examination history and treatment history of similar cases extracted by the search are displayed as a chart on this screen. The case display means 22 extracts examination history and treatment history information from the chart information included in each case information supplied from the similar case search means 23, and arranges them in a time series in the horizontal direction on the screen. The chart 40 is configured. In the present embodiment, the case display means 22 displays a chart 41 of the examination and treatment history of the patient under diagnosis at the bottom of the screen, thereby enabling comparison with the examination and treatment history of the case under diagnosis. ing.

  Furthermore, the case display means 22 accepts designation operations for the examination items and treatment items displayed on the chart 40. When a selection operation by the pointing device is accepted, detailed items 42 relating to the designated examination or treatment are displayed in a balloon as shown in the figure. The detailed item 42 extracts image information and an interpretation report corresponding to the specified examination from the case information, and extracts and displays the incidental information of the image information and the contents of the items recorded in the interpretation report.

  In addition, the figure shows a state in which information useful for planning a treatment plan is simply displayed on the screen from case information extracted by the search. In this embodiment, the information is displayed on this screen. When one case is selected from the charts 40, the names of examinations and treatments that are recommended to be performed in the future are displayed on the chart 41 at positions corresponding to the recommended times of the examination or treatment. Is done. The recommended examination item and the recommended treatment item are displayed by characters having a color different from that of the examination and treatment already performed.

  FIG. 19 is a flowchart showing an outline of the planning support process executed in the process in which the case information display means 22 controls the screen illustrated in FIG. As shown in the figure, when the case information display means 22 accepts the selection of the case by the designation of the chart 40 (S401), the examination items and treatment items that are being displayed on the chart 40 of the patient being diagnosed, that is, have already been performed are displayed. The medical record information of the selected case information and the examination items and treatment items recorded in the interpretation report are correlated with each other to associate the examination and treatment execution times (S402), and the examination performed on the selected case or Examinations and treatments that are treatments and have not yet been performed on patients being diagnosed are extracted, and the recommended implementation time is determined (S403). For example, if a patient under diagnosis has just undergone a first CT scan and the history of CT scans performed regularly every three months is recorded in the selected case information, this is recommended. The next test is a CT test, and the recommended timing is 3 months later.

  Subsequently, the case information display means 22 refers to the medical record information of the patient under diagnosis, and acquires information such as the patient's age, past illness, allergies, and whether or not a pacemaker is attached (S404). Then, the case information display means 22 determines whether there is no problem even if the examination and treatment extracted in step S403 are performed on the patient (S405).

  This determination is made in advance by creating a table that associates the data that can be recorded in the chart information with the prohibited items and the restricted items when such data is recorded, and stores it in the memory of the radiologist workstation 3. This is realized by determining the presence or absence of prohibition and restriction items by storing them and referring to this table in step S405. For example, in the association table, MR examination prohibition for a heart pacemaker wearer, prohibition of prescribed medicine for a predetermined allergy carrier, allowable upper limit of radiation dose when irradiating a patient under a predetermined age Etc. are defined in advance.

  For example, in the case where symbols, characters, and the like indicating wearing of the cardiac pacemaker are recorded in the medical record information, and the MR examination is included in the examination extracted in step S403, in step S407 Alerts with message display, warning sound, voice, etc.

  If there is no problem even if the extracted inspection is performed by adjusting the value, such as the allowable upper limit value of the radiation dose, the inspection content is optimized in step S406. For example, it is assumed that the case information extracted as a similar case records that an examination or treatment for irradiating an adult patient was performed. On the other hand, if the patient under diagnosis is an elderly person or a child, a small amount or number of times is set as the radiation dose or the number of times of radiation. When there is no need for optimization, the process of step S406 is substantially omitted.

  When the recommended examination items and treatment items and their implementation timing are determined by the above processing, the case information display means 22 has already implemented the recommended examination items and the recommended treatment items on the chart 41 of the patient under diagnosis. Items are displayed by color (S408).

  Instead of accepting the designation of the chart 40 in step S401, the case information of the case determined to have the highest degree of similarity at the time of similar case search is automatically selected, and the processing after step S402 is performed on the case information. It may be.

  As described above, the system of the present embodiment sets a plurality of images acquired in time series as a search query, and searches for cases in which the temporal change of image characteristics, that is, the course of disease is similar. can do. At this time, unlike the findings input by the doctor, the image characteristics are not influenced by the subjectivity or experience of the doctor, so that cases with similar progress can be searched based on objective facts. .

  As in the conventional system, providing information on cases having similar image characteristics acquired at a certain point in time is naturally useful in supporting diagnosis. However, there are many cases in which the same shadow is observed at a certain point in time, followed by a different course, and finally diagnosed with a different disease. For this reason, it may not be confident in the diagnosis that the disease is the same just because the features of the image at the point in time are similar. However, if similar shadows are observed at multiple time points, it is easy to be confident in the diagnosis of the same disease. That is, the system according to the present embodiment can provide more useful information than before when making a diagnosis to a doctor, thereby further contributing to the diagnosis of the doctor.

  Information on cases with similar courses is not only useful for diagnosis, but also for planning the next examination to be performed. For example, when examining what kind of examination should be performed next, as information on cases with similar courses, a case where cancer was diagnosed after repeated examinations by X-ray simple radiography If information about cases diagnosed with cancer at an early stage by providing a PET examination in the second examination is provided, doctors will consider conducting a PET examination at an early stage. In addition, even when the necessity of the examination is explained to the patient and the consent is obtained, providing the patient with such case information makes it easier to obtain the consent of the patient.

  Similarly, information on cases with similar courses is also useful when planning treatment plans. For example, there are two known cancer treatment methods: removing the lesion by surgery and eliminating the cancer cells by radiation therapy without surgery, but the size, location, and age of the patient (physical fitness) ), It may be difficult to decide which treatment to choose. In such a case, if information on cases with similar courses is provided, doctors will dispel their doubts about their diagnosis by referring to information on cases that have been treated successfully in cases where the patient's age is near be able to.

  Further, in the system according to the present embodiment, an examination or treatment plan that is recommended to be performed in the future is presented by the planning support function illustrated in FIG. For this reason, even a doctor who has little treatment experience can make a treatment plan with reference to treatment performed by a veteran doctor for cases with similar courses. In addition, when there is a case in which the effect of treatment appears remarkably, the patient's anxiety can be resolved by introducing such a case to the patient.

  At this time, the case information that is the basis for creating the recommendation is not extracted by similarity determination based on information input by the doctor, but is extracted by similarity determination based on image characteristics. Therefore, an examination plan or a treatment plan can be made based on information of cases that are truly similar in course.

  Further, in the system of the present embodiment, when examinations and treatments performed in cases with similar progress are not suitable for the patient being diagnosed, the fact is warned or adjusted to suit the patient being diagnosed. Since a recommendation is made, it is possible to prevent medical accidents due to misjudgment.

  In the above embodiment, as described with reference to FIG. 1, the registered image information, interpretation report, and medical chart information are used as they are as case information by associating them with each other. For example, when treatment is completed, the image information, the interpretation report, and the medical record information registered between the first diagnosis and the complete cure may be separately registered in the database as one piece of case information. The latter method has a demerit that the information of the case under treatment cannot be referred to as case information, but the search range at the time of similar case search is reduced by registering the image information, the interpretation report, and the medical record information together in one place. By narrowing the format of similar case information to a format suitable for search, high-speed search becomes possible.

  In addition, the format of information registered in the database and the search procedure by the similar case search means can be arbitrarily determined in consideration of the search speed or the restrictions of the existing database or system. The user interface of the region-of-interest setting means 21 and the screen layout controlled by the case information display means 22 can be arbitrarily determined. That is, the illustrations in the above embodiment are merely examples, and do not limit the scope of the present invention.

The figure which shows the structural example of the planning support system provided with the search function of similar cases The figure which shows an example of medical image information Figure showing an example of an interpretation report The figure which shows an example of medical chart information The figure which shows correlation with medical image information, an interpretation report, and medical chart information The figure which shows an example of the structure for implement | achieving the search function of a similar case The figure which shows the other example of a structure for implement | achieving the search function of a similar case The flowchart which shows an example of a process of a region-of-interest setting means Conceptual diagram showing a method for estimating ROI The figure which shows an example of ROI set by the region-of-interest setting means The figure which shows the example of the search condition information produced | generated by the region-of-interest setting means Conceptual diagram showing a first search procedure for searching for similar cases The flowchart which shows the similar case search process based on a 1st search procedure Conceptual diagram showing a second search procedure for searching for similar cases The flowchart which shows the similar case search process based on a 2nd search procedure The figure which shows an example of the list display screen of a search result The figure which shows an example of the display screen according to the case of a search result A figure showing an example of a treatment planning support screen Flow chart showing planning support processing

Explanation of symbols

1 modality, 2 laboratory workstation,
3 Radiologist workstation, 4 Physician workstation,
5 Image information database, 6 Interpretation report database,
7 chart information database, 8 DBMS, 9 network,
10 medical image information, 11 interpretation report, 12 medical record information, 13 case information,
16 List display screen, 17 Detailed display screen, 18 Planning support screen,
31-33 Image as a search query, 34 Subject information,
35 List of similar case identifiers, 36 selection frames, 37 operation buttons,
38 Test result information, 39 Highlight display frame, 40, 41 chart, 42 Detailed items

Claims (6)

A medical planning support system for supporting examination planning and / or treatment planning for a predetermined patient by displaying case information registered in a database,
A region of interest that sets a region of interest for each of the medical images by displaying a plurality of medical images acquired in time series on a monitor screen and accepting an area designating operation for at least one of the plurality of medical images. Setting means;
By searching the database, case information including case images each having a region having an image feature similar to the image feature in the region of interest is extracted for all the regions of interest set by the region of interest setting means. Similar case search means,
Case information display means for supporting the planning of the plan by extracting and displaying the history information of the examination and / or treatment from the case information extracted by the similar case search means, Planning support system to do. The similar case search means includes
For each region of interest set by the region-of-interest setting means, a process for searching for a case image having a region having an image feature similar to the image feature in the region of interest,
A set of case images related to the same patient from the case images detected by the search of the case image, and for each of the regions of interest, image features similar to the image features in the region of interest, respectively. A process of identifying a set including a case image having a prepared area;
The plan according to claim 1, wherein a process of supplying case information including case images belonging to the identified set to the case information display unit as case information extracted by searching the database is executed. Planning support system. The similar case search means includes
A process of searching for a case image having a region having an image characteristic similar to the image characteristic in the region of interest for the region of interest of one medical image set by the region of interest setting unit;
A case image having a region having an image characteristic similar to the image characteristic in the region of interest of the medical image other than the one medical image, with the case information including the case image extracted by the search of the case image as a search range Processing to further search
The search for case information in which case images having regions having image features similar to the image features in the region of interest of the medical image are extracted from all the medical images other than the one medical image by the search. The planning support system according to claim 1, wherein processing for supplying the case information display means as the case information extracted by the step is executed. Further comprising time interval setting means for setting an acquisition time interval of the plurality of medical images,
Cases in which the similar case retrieval unit has regions having image features similar to the image features in the region of interest for all the regions of interest set by the region of interest setting unit by searching the database. 4. The planning according to any one of claims 1 to 3, wherein case information including an image and the acquisition time interval of the case image is equal to the time interval set by the time interval setting means is searched. Support system. A process in which the case information display means extracts history information of examination and / or treatment from one case information in the case information extracted by the similar case search means;
A process of associating the timing of the examination and / or treatment by comparing the examination item and / or treatment item included in the extracted history information with the examination item and / or treatment item that has been performed on the patient;
A process of determining a recommended execution time of an unimplemented examination item and / or treatment item based on the association of the execution time;
The planning support according to any one of claims 1 to 4, wherein a process of displaying the unexecuted examination item and / or treatment item together with information of the recommended execution time as an execution recommended item is executed. system. Corresponding to estimate a region representing the same object as the object represented by the region specified on the first medical image on the second medical image from the coordinate information specifying the region specified on the first medical image. Further comprising region estimation means,
The region-of-interest setting unit accepts designation of a region of interest by the region designation operation for only one medical image, and supplies coordinate information for specifying the region of interest to the corresponding region estimation unit, thereby interested regions of other medical images 6. The planning support system according to claim 1, wherein the region of interest of each medical image is set by estimating.

Images