JP5557687B2 - Inspection plan creation device - Google Patents

Description

  The present invention relates to an examination plan creation apparatus that supports creation of an examination plan for diagnostic imaging examination.

  Conventionally, examinations using an X-ray CT apparatus, MR apparatus, ultrasonic apparatus, and other diagnostic imaging apparatuses have been performed, but since examinations may be a burden on patients, an appropriate examination plan is created. There is a need. For example, Patent Document 1 describes a mechanism for determining an examination schedule in time series according to patient characteristics, pathological conditions, disease characteristics, characteristics of various examinations, examination characteristics of examination equipment, and management status of examination equipment operating hours. Has been.

JP 2001-60230 A

  However, Patent Literature 1 does not describe a specific method that fully utilizes the patient's pathological condition or disease characteristics. For example, the growth process of the tumor varies from patient to patient or from tumor to tumor. If an inspection plan is not prepared so as to suit such individual cases, it will be too late if the period until the next inspection is long, and excessive inspection will occur if the period is shortened.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an examination plan creation device capable of obtaining an appropriate examination date according to tumor growth.

  In order to achieve the above-described object, the present invention provides a tumor data acquisition unit that acquires, in time series, tumor data that is data relating to the size of a tumor from examination information obtained in a past examination, and the tumor data acquisition means. Based on the acquired time-series tumor data, a time-series change of future tumor growth is obtained, an estimation means for estimating the arrival date and time when the tumor reaches a predetermined state, and the arrival estimated by the estimation means An inspection plan comprising: an inspection date and time calculating means that sets a date that is a predetermined grace period after the date and time as the next inspection date and time, and an output means that outputs the inspection date and time calculated by the inspection date and time calculating means It is a creation device.

  According to the present invention, it is possible to provide an examination plan creation device capable of obtaining an appropriate examination date according to tumor growth.

Hardware block diagram of inspection management system 1 including inspection plan creation device 100 The flowchart explaining the flow of the inspection date calculation process by this invention An example of the operation date calculation operation screen 200 The flowchart explaining the flow of the inspection date calculation process in 2nd Embodiment The flowchart explaining the flow of the inspection date calculation process in 3rd Embodiment The figure explaining the discriminator 5 used in 4th Embodiment The flowchart explaining the flow of the learning process in 4th Embodiment The flowchart explaining the flow of the examination date calculation process using the discriminator 5 in the fourth embodiment. The figure explaining calculation of the inspection date when there are a plurality of tumors (fifth embodiment) The flowchart explaining the flow of the inspection date calculation process in 5th Embodiment The figure explaining the example which calculates examination date and time in consideration of the severity of the tumor when there are a plurality of tumors (sixth embodiment) The flowchart explaining the flow of the inspection date calculation process in 6th Embodiment

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[First Embodiment]
First, with reference to FIG. 1, the structure of the test | inspection management system 1 which has the test | inspection plan preparation apparatus 100 of this invention is demonstrated.

  As shown in FIG. 1, an inspection plan creation apparatus 100 according to the present invention includes a display device 107 and an input device 109, and is connected to an image database 111, an image capturing device 112, and the like via a network 110.

The inspection plan creation apparatus 100 is a computer that performs processing for calculating an optimal inspection plan based on the image information of the subject.
As shown in FIG. 1, the inspection plan creation apparatus 100 includes an external device such as a CPU (Central Processing Unit) 101, a main memory 102, a storage device 103, a communication interface (communication I / F) 104, a display memory 105, and a mouse 108. Interface (I / F) 106, and each part is connected via a bus 113.

  The CPU 101 calls a program stored in the main memory 102 or the storage device 103 to the work memory area on the RAM of the main memory 102 and executes it, drives and controls each unit connected via the bus 113, and creates an inspection plan. Various processes performed by the apparatus 100 are realized.

  In addition, the CPU 101 executes an examination date calculation process for calculating an optimum next examination date using time-series data regarding the tumor acquired in the past examination of the patient. Details of the inspection date calculation process will be described later.

  The main memory 102 includes a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The ROM permanently holds a computer boot program, a program such as BIOS, data, and the like. The RAM temporarily stores programs, data, and the like loaded from the ROM, the storage device 103, and the like, and includes a work area used by the CPU 101 for performing various processes.

  The storage device 103 is a storage device that transfers data to and from a portable recording medium such as an HDD (hard disk drive), a flexible disk, an optical (magnetic) disk, a ZIP memory, a USB memory, and the like. Necessary data, OS (operating system) and the like are stored. As for the program, a control program corresponding to the OS and an application program are stored. Each of these program codes is read by the CPU 101 as necessary, transferred to the RAM of the main memory 102, and executed as various means.

The communication I / F 104 includes a communication control device, a communication port, and the like, and mediates communication between the inspection plan creation device 100 and the network 110. The communication I / F 104 performs communication control with the image database 111, another computer, or an image capturing apparatus 112 such as an X-ray CT apparatus or an MRI apparatus via the network 110.
The I / F 106 is a port for connecting a peripheral device, and transmits / receives data to / from the peripheral device. For example, a pointing device such as a mouse 108 or a stylus pen may be connected via the I / F 106.

  The display memory 105 is a buffer that temporarily accumulates display data input from the CPU 101. The accumulated display data is output to the display device 107 at a predetermined timing.

  The display device 107 includes a display device such as a liquid crystal panel and a CRT monitor, and a logic circuit for executing display processing in cooperation with the display device, and is connected to the CPU 101 via the display memory 105. The display device 107 displays display data stored in the display memory 105 under the control of the CPU 101.

  The input device 109 is an input device such as a keyboard, for example, and outputs various instructions and information input by the operator to the CPU 101. The operator interactively operates the examination plan creation apparatus 100 using external devices such as the display device 107, the input device 109, and the mouse 108. The mouse 108 may be another pointing device such as a trackpad or a trackball.

  The network 110 includes various communication networks such as a LAN (Local Area Network), a WAN (Wide Area Network), an intranet, the Internet, and the like, and communication connection between the image database 111, a server, other information devices, and the inspection plan creation apparatus 100. Mediate.

  The image database 111 accumulates and stores image data photographed by the medical image photographing device 112. In the inspection management system 1 shown in FIG. 1, the image database 111 is connected to the inspection plan creation apparatus 100 via the network 110, but the image database 111 is provided in, for example, the storage device 103 in the inspection plan creation apparatus 100. You may do it.

  Next, the operation of the inspection plan creation apparatus 100 will be described with reference to FIGS.

  The CPU 101 of the inspection plan creation apparatus 100 reads out the program and data related to the inspection date calculation processing in FIG. 2 from the main memory 102, and executes processing based on the program and data.

  In the examination date calculation process of FIG. 2, first, the CPU 101 of the examination plan creation apparatus 100 uses time series of data relating to tumors (hereinafter referred to as tumor data) obtained from past examinations on patients to be observed from the image database 111. (Step S101). The tumor data includes tumor size information (for example, volume, area, diameter, etc.) and date / time information such as examination date. It is assumed that the medical image information stored in the image database 111 includes information on tumor size and date / time information such as examination date in addition to image data.

Next, the CPU 101 obtains a time-series change in future tumor growth based on the time-series tumor data obtained in step S101. Therefore, first, the CPU 101 sets two sets of old and new tumor data adjacent in time series as one set, and calculates a doubling time DT according to the time series from each set of tumor data (step S102). For example, as tumor data of a patient in the image database 111, tumor size information at times t _A , t _B , t _C , and t _D is accumulated as V _A , V _B , V _C , and V _D , respectively. Shall. In this case, the CPU 101 extracts three sets of tumor data at adjacent times (t _A , t _B ), (t _B , t _C ), and (t _C , t _D ). First, the doubling time DT _{A, B} is obtained from the extracted tumor data of the first set (t _A , t _B ). Similarly, the doubling times DT _{B, C} , DT _{C, D} are obtained for each set of (t _B , t _C ) and (t _C , t _D ). Tumor doubling time DT can be calculated from the following equation (1).

In Expression (1), t ₁ and t ₂ are examination times, and V ₁ and V ₂ are tumor sizes at times t ₁ and t ₂ , respectively.

Next, the CPU 101 converts the doubling times DT _{A, B} , DT _{B, C} , DT _{C, D} of each set obtained in step S102 to tumor growth rates r _{A, B} , r _{B, C} , r _{C, D.} Conversion is performed (step S103). Since the doubling time DT and the growth rate r of the tumor are expressed by the relationship of the following equation (2), the growth rate r _{A, B} , r _{B, C} , r _{C of} each set of tumors is expressed by the equation (3). _{, D} can be calculated.

exp (growth rate r * doubling time DT) = 2 (2)
Growth rate r = log2 / doubling time DT (3)

For example, in step S102, the doubling times DT _{A, B} , DT _{B, C} , DT _C of the three sets of tumor data at times (t _A , t _B ), (t _B , t _C ), (t _C , t _D ) are obtained. _{, D} are obtained respectively. In step S103, using equation (3), the growth rate r _{A, B} , r is obtained for each set of tumor data at time (t _A , t _B ), (t _B , t _C ), (t _C , t _D ). _{B, C} , r _{C, and D} are calculated.

Next, the CPU 101 calculates an interpolation function for interpolating the tumor growth rate r obtained at step S103 at each time, and obtains a time-series change in the future growth rate using the calculated interpolation function (step S104). .
For example, the growth rate r _{A, B} , r _{B, C} of each set of tumor data at time (t _A , t _B ), (t _B , t _C ), (t _C , t _D ) obtained in step S103 , R _{C, D} , a function (interpolation function) indicating a change in growth rate is obtained. The obtained interpolation function is represented by a function having the date D (date information t) as a variable, and the future growth rate can be estimated by setting a future date for this interpolation function.

Next, the CPU 101 calculates the arrival date and time when the tumor reaches a predetermined state according to the interpolation function obtained by the calculation of step S104 (step S105). Examples of the above-mentioned “predetermined state” include, for example, a stage where a tumor transitions to a certain state, a stage where it becomes dangerous if left unattended, and a stage where the stage indicating the tumor state changes. If this “predetermined state” is given as a threshold that can be arbitrarily set and applied to the above-described interpolation function, the arrival date and time can be calculated. The threshold value is given as, for example, a doubling time DT ^* or a growth rate r ^* .
In addition, although an arbitrary numerical value can be set as the threshold, it is desirable to set a value estimated based on statistical, medical knowledge, or experience.
In the following description, a doubling time DT ^* or a growth rate r ^* that represents a stage that becomes dangerous if left untreated as a threshold value is used.

  As the threshold value, a value preset in the storage device 103 may be used, or input to each input column of the optimization parameter setting area 34 displayed on the examination date calculation operation screen 200 (FIG. 3) described later. Direct input may be performed using the device 109. Further, a graph 33 (FIG. 3) showing the transition of tumor growth displayed on the examination date calculation operation screen 200 in time series is displayed, and a threshold handle (horizontal double dotted line 41 in FIG. 3) is displayed on the graph 33. ), And the threshold value can be set by dragging the threshold value handle 41 by the operation of the mouse 108 and changing the position to a position indicating an arbitrary value.

  FIG. 3 is an example of the examination date calculation operation screen 200 displayed on the display device 107. The operation screen 200 for calculating the examination date includes a graph area 31 that displays a graph 33 showing the transition of tumor growth in time series, and the vertical axis component of the graph is one of “volume”, “speed”, and “acceleration”. A graph display selection button 32 for selecting from, an optimization parameter setting area 34 for setting various threshold values, an interpolation function, and the like, a next inspection date estimation parameter display area 36 for displaying a calculation result of the inspection date, and the like When displaying a nodule number to be displayed on the graph 33, a nodule number display input field 37 for inputting, a feed button 38, 39 for sending and displaying a nodule of the next number, and a report output instruction And an output button 40 to be operated.

  When “speed” is selected by the graph display selection button 32 on the operation screen 200 for calculating the examination date, a graph 33 showing a time-series change in the growth rate of the tumor with the vertical axis representing the growth rate and the horizontal axis representing time. 31. Similarly, when “volume” or “acceleration” is selected, each graph is displayed in the graph area 31 with the vertical axis representing volume or acceleration and the horizontal axis representing time.

  In the optimization parameter setting area 34 of the operation screen 200 for calculating the inspection date, input fields for setting threshold values of a grace period, a growth rate, a doubling time, a volume, and a maximum diameter, and an interpolation function are set. An input field is provided. In addition, a slider bar for adjusting the value is provided in the threshold value input field. In the input field for setting the interpolation function, a pull-down list or the like indicating that the type of the interpolation function can be selected is provided. Examples of types of interpolation functions that can be selected include linear equations and higher-order polynomials. Using a function selected from the pull-down list, an interpolation function representing a transition of tumor growth is calculated. In the example of FIG. 3, a linear function interpolation function (r = 20t + 10) is calculated, and a straight line 45 indicating the interpolation function is also displayed in the graph area.

The next inspection date (Date ^* ) calculated by the inspection date calculation process is calculated in the next inspection date estimation parameter display area 36. In step S105, the growth rate on the next examination date, the doubling time of the tumor from this time to the next examination date, the volume of the tumor, and the like may be calculated and displayed in the next examination date estimation parameter display area 36.
The horizontal axis position of the intersection of the straight line (or curve) 45 indicating the interpolation function and the threshold handle 41 represents the time (arrival date D _th ) at which the tumor growth rate reaches the threshold.

Here, a threshold setting method using the examination date calculation operation screen 200 will be described.
In FIG. 3, a graph 33 displayed in the graph area 31 is a graph showing the relationship between the growth rate of the tumor and time.
A threshold handle (horizontal double dotted line) 41 on the graph 33 represents a threshold at which the tumor is judged to be in a dangerous state, and the position (value) can be arbitrarily changed. For example, the horizontal double line 41 may be moved directly by operating the mouse 108 or the like, or may be handled by sliding the “growth speed” or “doubling time” slider bar in the optimization parameter setting area 34. It may be possible to move by the amount of movement. Alternatively, the threshold value may be set by inputting a numerical value in an input field of “growth rate” or “doubling time” in the optimization parameter setting area 34. When the threshold value is changed by any of these operations, the CPU 101 recalculates the next inspection date Date ^* in real time using the changed threshold value, and displays each graph in the graph 33 and the next inspection date estimation parameter display area 36. Update the display contents of the column.

Next, the CPU 101 sets, as the inspection date / time Date ^* , the date and time that is traced back from the arrival date and time by a predetermined grace time from the arrival date and time _Dth obtained by the calculation of step S105 and the preset grace time.
The inspection date Date ^* is obtained by the following equation (4).

Inspection date and time Date ^* = Arrival date and time D _th -Grace time (4)

The calculated examination date Date ^* is output (displayed) to the graph area 31 and the next examination date estimation parameter display area 36 in FIG. 3 (step S106).
In step S106, the growth rate on the next examination date Date ^* , the doubling time of the tumor from this time to the next examination date, the tumor volume, and the like are calculated and displayed in the next examination date estimation parameter display area 36. Good.

Regarding the grace time, a value preset in the storage device 103 may be used, a value numerically input in the grace time input field of the optimization parameter setting area 34 may be used, or “grace time” may be used. The value adjusted by the sliding operation of the slider bar may be used. Further, the grace time handle 42 indicated by the vertical double dotted line on the graph 33 may be moved and changed directly by a drag operation of the mouse 108.
When the grace time is changed by any of these operations, the CPU 101 recalculates the next inspection date Date ^* in real time using the changed grace time, and displays the graph 33 or the next inspection date estimation parameter display area 36. Update the display contents in each column.

As described above, according to the test plan creation apparatus 100 of the first embodiment, tumor data is acquired in time series, the doubling time of the tumor is calculated, and the growth rate of the tumor is further calculated. Further, the future tumor growth is obtained by interpolation processing based on the calculated tumor growth rate, the arrival date and time to reach the predetermined state is estimated, and the date that is back from the estimated arrival date by the predetermined grace time is next time Calculate and output as the inspection date.
Thereby, an appropriate examination date according to the tumor can be obtained. Therefore, it is possible to reduce the burden on the patient due to excessive examinations and to obtain an optimum examination date that is not too late.
In addition, in order to set a grace period and the day after the specified grace time from the arrival date and time to reach a state where the tumor is present as the next examination date, in response to sudden changes in the tumor exceeding the prediction that can sometimes occur in living tissue The inspection date can be set more safely.

In the example of FIG. 3, the graph 33 representing the time-series change of the growth rate is displayed on the inspection date calculation operation screen 200, and the interpolation function (eg, r = 20t + 10 etc.) was calculated, and the date and time when the tumor reached a certain state (threshold value) was calculated according to this interpolation function. The interpolation function indicating the growth of the tumor is expressed by the relationship between “speed” and time as in this example. It is not limited. For example, when it is preferable to express the relationship between “volume” and time, “volume” is selected by operating the graph display selection button 32 on the operation screen 200 for examination date calculation, and the relationship between tumor volume and time is expressed. A graph is displayed in the graph area 31 and an interpolation function (for example, V (t) = pt + q, etc .; p and q are constants) representing the relationship between volume and time is obtained. The date and time when the threshold value is reached may be calculated.
Further, when it is preferable to express the tumor growth in terms of “acceleration” and time, “acceleration” is selected by operating the graph display selection button 32 on the examination date calculation operation screen 200, and the acceleration of the tumor growth and A graph representing the relationship between time is displayed in the graph area 31, and an interpolation function (for example, a (t) = pt + q, etc .; p and q are constants) representing the relationship between acceleration and time is obtained. The date and time when a certain state (speed or acceleration threshold) is reached may be calculated. An example in which tumor growth is expressed by the relationship between “acceleration” and time and the arrival time is calculated will be described in the second and third embodiments.

[Second Embodiment]
Next, an inspection plan creation apparatus 100 according to the second embodiment of this invention will be described with reference to FIG. The hardware configuration of the inspection plan creation apparatus 100 according to the second embodiment is the same as that of the inspection plan creation apparatus 100 according to the first embodiment. I will explain.

  In the second embodiment, “acceleration” is selected by the graph display selection button 32 on the inspection date calculation operation screen 200 in FIG. 3, and the interpolation function type is selected from the interpolation function pull-down menu in the optimization parameter setting area 34. It is assumed that “primary expression” is selected.

  First, the CPU 101 of the examination plan creating apparatus 100 acquires tumor data in time series from the past examination information of the patient to be observed from the image database 111, as in the examination date calculation process of the first embodiment (step). S201), based on the time-series tumor data obtained in step S201, the doubling time DT according to the time series is calculated from each set of tumor data including two sets of old and new tumor data adjacent in time series. (Step S202). Next, the CPU 101 converts the doubling time of each group obtained in step S202 into a tumor growth rate (step S203).

Next, the CPU 101 further converts the growth rate obtained in step S203 into acceleration (step S204).
For example, the growth rate of each set of tumor data at the times (t _A , t _B ), (t _B , t _C ), and (t _C , t _D ) obtained in step S203 is expressed as r _{A, B} , r _{B ,} respectively. _{, C 1} , r _{C, D} , the acceleration can be calculated from the gradient of each set of growth rates (r _{A, B} , r _{B, C} ) or (r _{B, C} , r _{C, D} ).

  Based on the acceleration time-series data obtained in step S204, the CPU 101 calculates a linear interpolation function (step S205). By using a linear interpolation function, when the tumor growth rate is monotonously increasing or monotonically decreasing, the date and time when the predetermined state (target growth rate) is reached can be easily calculated.

Next, the CPU 101 calculates the date and time (arrival date and time D _th ) when the growth acceleration reaches a preset threshold according to the linear interpolation function obtained in step S205 (step S206). About a threshold value, it can set arbitrarily similarly to 1st Embodiment.

An inspection date and time Date ^* is defined as a date and time that is back by a predetermined grace time from the arrival date and time _Dth from the arrival date and time _Dth obtained by the calculation in step S206 and a preset grace time (step S207). . The inspection date Date ^* can be expressed by the above formula (4). In addition, the grace time can be arbitrarily set similarly to the first embodiment.

The calculated examination date Date ^{* and the} like are output (displayed) to the graph area 31 and the next examination date estimation parameter display area 36 in FIG.
In step S207, the growth rate on the next examination date Date ^* , the doubling time of the tumor from this time to the next examination date, the tumor volume, and the like are calculated and displayed in the next examination date estimation parameter display area 36. Also good.

When the threshold value r ^* and the grace time are changed by the operator, the CPU 101 recalculates the inspection date / time Date ^{* and the} like in real time based on the changed threshold value r ^* and the grace time, and the graph 33 and the next inspection date estimation parameter are calculated. The display contents of each column in the display area 36 are updated.

  As described above, according to the test plan creation apparatus 100 of the second embodiment, the arrival date and time when the tumor reaches a predetermined state is determined using the linear function indicating the time-series change in the acceleration of tumor growth as an interpolation function. The date and time that is calculated and the predetermined grace time is traced back from the arrival date and time is taken as the inspection date. For this reason, when the growth rate of the tumor monotonously increases or monotonously decreases, the date and time when the tumor reaches a predetermined state can be easily calculated. Further, since the acceleration of tumor growth is used, it is possible to calculate the change tendency when there is a change in the growth rate.

[Third Embodiment]
Next, an inspection plan creation apparatus 100 according to the third embodiment of the present invention will be described with reference to FIG. Note that the hardware configuration of the inspection plan creation apparatus 100 according to the third embodiment is the same as that of the inspection plan creation apparatus 100 according to the first embodiment. I will explain.

  In the third embodiment, “acceleration” is selected by the graph display selection button 32 on the inspection date calculation operation screen 200 in FIG. 3, and the interpolation function type is selected from the interpolation function pull-down menu in the optimization parameter setting area 34. It is assumed that “higher order polynomial” is selected.

  First, the CPU 101 of the examination plan creation apparatus 100 obtains tumor data in time series from the examination information of the patient to be observed from the image database 111 in the same manner as the examination date calculation processing of the first and second embodiments. (Step S301), and based on the time-series tumor data obtained in Step S301, the doubling time according to the time series from each set of tumor data including two sets of old and new tumor data adjacent in time series DT is calculated (step S302). Next, the CPU 101 converts each group doubling time obtained in step S302 into a tumor growth rate (step S303).

  Next, the CPU 101 further converts the growth rate obtained in step S303 into acceleration (step S304). The calculation of the acceleration is the same as step S204 (FIG. 4) in the examination date calculation process of the second embodiment.

  Based on the acceleration time-series data obtained in step S304, the CPU 101 calculates a high-order polynomial interpolation function (step S305). By using a high-order polynomial as an interpolation function, it is possible to easily calculate the arrival date and time to reach a predetermined state even when the growth rate of the tumor is accompanied by a complicated change.

Next, the CPU 101 calculates the date and time (arrival date and time D _th ) when the growth acceleration reaches a preset threshold according to the interpolation function of the higher-order polynomial obtained in step S305 (step S306). About a threshold value, it can set arbitrarily similarly to 1st Embodiment.

An inspection date and time Date ^* is a date and time that is a predetermined grace period from the arrival date and time _Dth as a starting point from the arrival date and time _Dth obtained by the calculation of step S306 and a preset grace time (step S307). . The inspection date Date ^* can be expressed by the above formula (4). In addition, the grace time can be arbitrarily set similarly to the first embodiment.

The calculated examination date and time is output (displayed) to the graph area 31 and the next examination date estimation parameter 36 in FIG.
Further, in step S307, the growth rate on the next examination date Date ^* , the doubling time of the tumor from this time to the next examination date, the tumor volume, etc. are calculated and displayed in the next examination date estimation parameter display area 36. Also good.

When the threshold value r ^* and the grace time are changed by the operator, the CPU 101 recalculates the inspection date / time Date ^{* and the} like in real time based on the changed threshold value r ^* and the grace time, and the graph 33 and the next inspection date estimation parameter are calculated. The display contents of each column in the display area 36 are updated.

  As described above, according to the test plan creation apparatus 100 of the third embodiment, the time-series change in the acceleration of tumor growth is represented by a high-order polynomial, and this is used as an interpolation function to bring the tumor into a predetermined state. The arrival date and time is calculated, and the date and time that goes back a predetermined grace time from the arrival date and time is taken as the inspection date. For this reason, even when the growth rate of the tumor changes in a complicated manner, the date and time when the tumor reaches a predetermined limit state can be calculated.

[Fourth Embodiment]
Next, with reference to FIGS. 6-8, the test | inspection plan preparation apparatus 100 of the 4th Embodiment of this invention is demonstrated. The hardware configuration of the inspection plan creation apparatus 100 according to the fourth embodiment is the same as that of the inspection plan creation apparatus 100 according to the first embodiment. I will explain.

  In addition to the function of the test plan creation device 100 of the first embodiment, the test plan creation device 100 of the fourth embodiment has an optimum threshold value according to the state of the tumor, as shown in FIG. A discriminator (learning device) 5 for setting (a value representing a stage at which the tumor is in a predetermined state) is provided. When a certain tumor data is input, the discriminator 5 analyzes the state of the tumor data and performs a calculation for obtaining a threshold value that matches the state of the tumor data. The discriminator 5 also functions as a learning device as shown in FIG. 6A, and can learn the relationship between the state of the tumor and the optimum threshold using a plurality of tumor data for learning.

As shown in FIG. 7, first, the CPU 101 of the inspection plan creation apparatus 100 performs a learning process for the classifier 5.
In the learning process, the CPU 101 of the examination plan creation apparatus 100 collects various tumor data m1, m2,... For various patients from the past to the present accumulated in the image database 111 and the storage device 103 as learning data. Then, an identification function for classifying the group of learning data into a plurality of groups (state A, state B, state C,...) According to a predetermined classification method is calculated (step S402).

  The learning data to be collected includes time series data (volume, area, examination date, etc.) of tumor size and a threshold value set in the past for the tumor (the tumor is in a “predetermined state”) Value). If the threshold value is set by the operator, the threshold value is collected as learning data. If the examination date calculated by the CPU 101 is not valid, the threshold value that should have been set is the learning data. Collect as. The classifier 5 classifies the states of the tumors m1, m2,... Using these learning data, and associates a threshold having a large correlation with each classification. When the learning process is performed in this way, the state of the tumor data (target tumor) input using the discriminator 5 can be analyzed to obtain an optimum threshold value.

  The classifier 5 may learn using medical knowledge in addition to the learning data described above. The classifier 5 may use a multi-class classifier that classifies a group of data into multiple classes, or may combine two classes of classifiers to classify into multiple classes. As a classification method, for example, a support vector machine, a supervised classifier such as a linear classifier, an unsupervised classifier such as clustering or a neural network, or other classifiers may be combined.

  When learning is performed in step S402, it is desirable to confirm by using cross-validation or the like that the group classified in step S401 can be correctly classified by the identification function obtained in step S402.

When the above learning process is performed, it is possible to automatically set an optimum threshold for the target tumor using the classifier 5.
As shown in FIG. 8, when data (tumor data) regarding a patient's tumor to be observed is input from the image database 111 to the examination plan creation apparatus 100 (step S411), the CPU 101 performs the identification calculated in step S402. The state of the tumor data is analyzed using a function, an optimum threshold value corresponding to the state is calculated, and this threshold value is set as a threshold value representing a “predetermined state”. Thereafter, the CPU 101 calculates the inspection date Date ^* using any one of the methods of the first, second, and third embodiments (step S412).

  As described above, according to the test plan creation apparatus 100 of the fourth embodiment, the state and threshold value of each tumor data using various past tumor data and threshold information applied to the tumor data. It is possible to set the optimum threshold value to be set for the input tumor data based on the learning result. As a result, the optimum threshold value corresponding to the target tumor can be automatically set based on the past data, so that the operator can easily obtain the next examination date without hesitation in setting the threshold value.

[Fifth Embodiment]
Next, with reference to FIG.9 and FIG.10, the test | inspection plan preparation apparatus 100 of the 5th Embodiment of this invention is demonstrated. The hardware configuration of the inspection plan creation apparatus 100 according to the fifth embodiment is the same as that of the inspection plan creation apparatus 100 according to the first embodiment. I will explain.

  In the examination plan creation apparatus 100 according to the fifth embodiment, an examination date calculation process when a plurality of tumors m1, m2, m3, and m4 exist in the same patient as shown in FIG. 9 will be described.

As shown in FIG. 10, first, when time-series tumor data for each of the plurality of tumors m1, m2, m3, and m4 is input (step S501), the CPU 101 selects any one of the first to fourth for each tumor. The inspection date and time Date ^* is calculated using the above method. In the flowchart of FIG. 10, as an example, as in the fourth embodiment, an optimum threshold value is set for each tumor data using the discriminator 5, and the examination date Date ^* is calculated for each tumor (step) S502, step S503).
The CPU 101 of the examination plan creation apparatus 100 detects a tumor that requires the earliest response from the calculated examination date and time Date ^* , and outputs the examination date and time of the tumor as the next examination date and time (step S504).

  In the example shown in FIG. 9, the examination dates and times t1, t2, t3, and t4 are calculated for the tumors m1, m2, m3, and m4, respectively, and the order is t3 → t4 → t1 → t2. In this case, since the examination date / time t3 calculated for the tumor m3 is the earliest examination date / time, the examination date / time t3 calculated for the tumor m3 is output as the next examination date / time of this patient.

  As described above, according to the test plan calculation apparatus 100 of the fifth embodiment, when there are a plurality of tumors, the next test date is calculated for each tumor, and the most calculated next test date is the most. The tumor that needs to be dealt with early is determined, and the next examination date of the tumor is taken as the patient's next examination date. Therefore, even when there are a plurality of identical tumors, the next examination date can be easily calculated.

[Sixth Embodiment]
Next, with reference to FIG.11 and FIG.12, the test | inspection plan preparation apparatus 100 of the 6th Embodiment of this invention is demonstrated. Note that in the inspection plan creation apparatus 100 according to the sixth embodiment, the same components as those in the inspection plan creation apparatus 100 according to the first embodiment are denoted by the same reference numerals, and description thereof is omitted.

  In the sixth embodiment, when a plurality of tumors exist in the same patient, an examination date calculation process that reflects the severity of each tumor is performed.

First, when tumor data for a plurality of tumors is input (step S601), an examination date is calculated for each tumor using any of the first to fourth methods (step S602).
Next, the CPU 101 of the examination plan creation apparatus 100 calculates a severity coefficient for each tumor (step S603). As the severity coefficient, a value reflecting the malignancy and benignity of the tumor may be used, a value indicating the stage, or a value indicating the other severity may be used. Or the value which reflected the allowance which shows the tolerance | permissible_range which can postpone an inspection date may be sufficient.

For example, suppose that the severity coefficient of a tumor is set to three levels (1, 2, 3), and 3 is set to be the most severe. In addition, a patient has tumors m1, m2, m3, and m4, and the severity coefficient of each tumor is W (m1) = 1, W (m2) = 2, W (m3) = 1, W Consider the case where (m4) = 1.
The examination dates calculated for each tumor m1, m2, m3, and m4 by any of the methods of the first to fourth embodiments are t1, t2, t3, and t4, respectively. The total coefficient W for normalization is W = W (m1) + W (m2) + W (m3) + W (m4) = 5.

Next, the CPU 101 obtains the center of gravity of the examination date from the severity coefficient and the examination date (step S604).
In the above example, the center of gravity of the inspection date is (0.2 × t1 + 0.4 × t2 + 0.2 × t3 + 0.2 × t4).

Here, the CPU 101 compares the center of gravity of the examination date and time with the earliest danger date (early arrival date and time D _{th for} all tumors to be calculated) obtained in the calculation process of step S602 (step S605). And if the following formula | equation (5) is satisfy | filled (step S605; Yes), the gravity center of a test | inspection date will be made into the 1st test | inspection date (step S606).

          Center of gravity of inspection date <earliest danger day ... (5)

  When the above formula (5) is not satisfied, that is, when the center of gravity of the inspection date / time is after the earliest danger date obtained in the calculation process of step S602 (step S605; No), the earliest inspection date / time (in FIG. 11) In the example, t3) is set as the next inspection date and time (step S607).

  Next, the center of gravity of the inspection date and time is compared with the latest inspection date and time (t2 in the example of FIG. 11) obtained in step S602 (step S608), and if the difference is equal to or greater than a predetermined threshold (step S609; Yes) ), The latest inspection date is set as the second inspection date (step S610), and the inspection is performed twice in total. If the difference is smaller than the predetermined threshold value (step S609; No), it is considered that there is no large bias in the growth rate of the tumor, so the second examination does not need to be set.

  As described above, the optimum examination date and time and the number of examinations are calculated from the examination date and time for each tumor obtained in step S602 and the severity coefficient for each tumor obtained in step S603.

As described above, in the sixth embodiment, when there are a plurality of tumors in the same patient, the optimal examination date and time and the number of examinations can be determined in consideration of the severity of the tumor.
Note that the number of inspections is not limited to one or two, and may be three or more. For example, for each inspection date and time after the date and time of the center of gravity of the inspection date and time, if the interval between the inspection dates and times is larger than a predetermined threshold value, the third and fourth inspection dates and times may be set.

  As mentioned above, although preferred embodiment of the test | inspection plan preparation apparatus concerning this invention was described, this invention is not limited to the above-mentioned embodiment. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these are naturally within the technical scope of the present invention. Understood.

100... Inspection plan creation device 101... CPU
102... Main memory 103... Storage device 104... Communication I / F
105 ··· Display memory 106 ··· I / F
DESCRIPTION OF SYMBOLS 107 ... Display apparatus 108 ... Mouse 109 ... Input apparatus 110 ... Network 111 ... Image database 112 ... Medical imaging apparatus 200 ... Calculation of examination date Operation screen 31... Graph area 32... Graph display selection button 33... Graph showing changes in tumor growth in time series 34... Optimization parameter setting area 36 ... Next inspection date estimation parameter display area 37 ... Nodule number display input field 38, 39 ... Feed button 40 ... Output button 41 ... Threshold handle 42 ... ..Grace time handle 45 ... Line representing interpolation function

Claims (7)

Tumor data acquisition means for acquiring in a time series tumor data, which is data relating to the size of a tumor, from examination information obtained in a past examination;
Based on the time-series tumor data acquired by the tumor data acquisition means, the time-series change of future tumor growth is determined, and the estimation means for estimating the arrival date and time when the tumor reaches a predetermined state;
An inspection date and time calculating means for setting the date after the predetermined grace time from the arrival date and time estimated by the estimating means as the next inspection date and time,
Output means for outputting the inspection date and time calculated by the inspection date and time calculating means;
An inspection plan creation device comprising: The estimation means includes
A growth rate calculation means for calculating a doubling time of a tumor in a time series based on the time series tumor data, and converting the calculated doubling time into a growth rate;
Based on the growth rate of the tumor calculated by the growth rate calculation unit, an interpolation function calculation unit that calculates an interpolation function that represents a time-series change in future tumor growth;
The inspection plan creation device according to claim 1, further comprising: The estimation means includes
A growth rate calculation means for calculating a doubling time of a tumor in a time series based on the time series tumor data, and converting the calculated doubling time into a growth rate;
Growth acceleration calculating means for calculating growth acceleration from the time-series change of the growth rate;
Based on the growth acceleration of the tumor calculated by the growth acceleration calculation means, an interpolation function calculation means for calculating an interpolation function representing a time-series change in future tumor growth;
The inspection plan creation device according to claim 1, further comprising: Interpolation function selection means for selectively presenting to the user the type of interpolation function that represents a time-series change in future tumor growth,
The estimation means uses the type of interpolation function selected by the interpolation function selection means as an interpolation function representing the growth of the tumor, and uses the interpolation function to estimate the arrival date and time. Inspection plan creation device of description. Learning means for learning an optimal threshold for the tumor to transition to the predetermined state using a plurality of learning test information,
Using a learning result by the learning means, further comprising a threshold setting means for setting the optimum threshold for the tumor data acquired by the tumor data acquiring means,
2. The examination plan creation apparatus according to claim 1, wherein the estimation means estimates the arrival date and time using a time-series change in future tumor growth and the threshold value. When there are a plurality of the tumors for the same patient, the tumor data acquisition means, the estimation means, and the examination date calculation means perform a process for calculating the examination date and time for each tumor,
The inspection plan creation apparatus according to claim 1, wherein the output means outputs the earliest date and time among the calculated plurality of inspection dates and times as the next inspection date and time. When there are a plurality of the tumors for the same patient, the tumor data acquisition means, the estimation means, and the examination date calculation means perform a process for calculating the examination date and time for each tumor,
The output means includes
A severity coefficient calculating means for calculating a severity coefficient representing the severity of each tumor;
Center of gravity calculation means for calculating the center of gravity of the examination date and time using the examination date and time calculated for each tumor and the severity coefficient of each tumor;
A first output means for outputting the centroid of the inspection date as the first inspection date when the centroid of the inspection date calculated by the centroid calculating means is earlier than the earliest danger date;
A second output means for outputting the latest examination date as the second examination date when the difference between the centroid of the examination date calculated by the centroid calculating means and the latest examination date is equal to or greater than a predetermined threshold; ,
The inspection plan creation device according to claim 1, further comprising:

Images