Condition analysis device and condition analysis method

Abstract

One aspect of the present invention is a condition analysis device comprising: an acquisition unit that acquires rank information ranking the degree of independence or the magnitude of a disability of a patient, and biometric information about the patient; a derivation unit that derives, for each rank, range information showing a range in which the biometric information is distributed; and a superimposed display unit that displays the range information for each rank as an image, and displays a prescribed image in the range indicated by the image in a superimposed manner on the image at a position corresponding to the biometric information of one patient.

Description

[Technical Field]

[0001] The present invention relates to a condition analyzing device and a condition analyzing method. [Background technology]

[0002] In rehabilitation medicine, when a patient suffers from paralysis due to cerebrovascular disease or other conditions, the Functional Independence Measure (FIM) and Stroke Impairment Assessment Set (SIAS) are used as scales to evaluate the patient. These are ranking information obtained by medical professionals using predetermined scales to score the patient's level of independence and the severity of their disability. In addition, the use of wearable devices has made it easier to obtain information on the amount of activity in daily life. Obtaining such activity data can give an idea of ​​how active a person is in their daily life. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Technical Information Association, Biometric Sensing and Its Application to Human State Estimation, 2020. [Non-patent document 2] How to exclude outliers: https: / / toukeigaku-jouhou.info / 2018 / 04 / 25 / determine-outlier / [Non-patent document 3] Missing data handling using multiple imputation https: / / qiita.com / saltcooky / items / 2e39c4bc099d20f20d59 Summary of the Invention [Problem to be solved by the invention]

[0004] One of the goals of rehabilitation is for patients to return to their daily lives, and restoring normal activity levels is an important aspect. However, no effective methods have been proposed for visualizing or quantifying the correlation between activity levels implied from acquired rank information and biometric data. This has made it difficult to provide appropriate support for rehabilitation, such as supporting the creation of rehabilitation treatment plans or helping to increase patients' motivation for rehabilitation training.

[0005] In view of the above circumstances, an object of the present invention is to provide a technique that can provide suitable support for rehabilitation. [Means for solving the problem]

[0006] One aspect of the present invention is a condition analysis device that includes an acquisition unit that acquires rank information that ranks a patient's degree of independence or the severity of their disability and the patient's biostatistical information; a derivation unit that derives range information that indicates the range in which the biostatistical information is distributed for each rank; and a superimposition display unit that displays the range information as an image for each rank and superimposes a specified image on the image at a position corresponding to the biostatistical information of a patient within the range indicated by the image.

[0007] One aspect of the present invention is a condition analysis method comprising: an acquisition step of acquiring rank information that ranks the patient's degree of independence or the severity of their disability and the patient's biostatistical information; a derivation step of deriving range information that indicates the range in which the biostatistical information is distributed for each rank; and a superimposition display step of displaying the range information as an image for each rank and superimposing a predetermined image on the image at a position corresponding to the biostatistical information of a patient within the range indicated by the image. [Effects of the Invention]

[0008] The present invention makes it possible to provide suitable support for rehabilitation. [Brief explanation of the drawings]

[0009] [Figure 1]1 is a functional block diagram showing the functional configuration of a condition analyzing device according to a first embodiment. FIG. [Figure 2] FIG. 10 is a diagram showing a specific example of a patient database. [Figure 3] FIG. 10 is a diagram illustrating a specific example of a statistical database. [Figure 4] FIG. 10 is a diagram illustrating an example of a graph. [Figure 5] 10 is a flowchart showing the flow of processing by the condition analyzer. [Figure 6] FIG. 10 is a functional block diagram showing the functional configuration of a condition analyzing device according to a second embodiment. [Figure 7] FIG. 10 is a diagram illustrating an example using quartile values. [Figure 8] FIG. 10 is a graph showing the results after the exclusion. [Figure 9] FIG. 10 is a functional block diagram showing the functional configuration of a condition analyzing device according to a third embodiment. [Figure 10] FIG. 10 is a diagram showing an example of display by a trend display unit. [Figure 11] FIG. 10 is a functional block diagram showing the functional configuration of a condition analyzing device according to a fourth embodiment. [Figure 12] FIG. 10 is a diagram showing an example of a graph in which predicted rank statistics and vital statistics are superimposed. [Figure 13] FIG. 10 shows the correlation coefficient of predicted values. [Figure 14] FIG. 10 is a functional block diagram showing the functional configuration of a condition analyzing device according to a fifth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] (First embodiment) FIG. 1 is a functional block diagram showing the functional configuration of a condition analyzer 100. The condition analyzer 100 includes a central processing unit (CPU), memory, auxiliary storage device, and other components connected via a bus. By executing a condition analysis program, the condition analyzer 100 functions as a device including a display unit 110, a patient information storage unit 141, a statistical information storage unit 142, and a control unit 120. Note that all or part of the functions of the display unit 110, the patient information storage unit 141, and the control unit 120 may be realized using hardware such as an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a field-programmable gate array (FPGA). The condition analysis program may be recorded on a computer-readable recording medium. Examples of the computer-readable recording medium include portable media such as a flexible disk, a magneto-optical disk, a read-only memory (ROM), and a CD-ROM, and storage devices such as a hard disk built into a computer system. The condition analysis program may be transmitted via a telecommunications line.

[0011] The display unit 110 is a display device that displays various types of information using liquid crystal, organic EL (Electro Luminescence), or the like.

[0012] The patient information storage unit 141 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The patient information storage unit 141 stores a patient database. Figure 2 is a diagram showing a specific example of the patient database. The patient database is configured with a patient identifier, date information, acquisition rank information, and acquired biological information.

[0013] The patient identifier is an identifier for uniquely identifying a patient. The date information indicates the date on which the acquired rank information and the acquired biological information were acquired. The acquired rank information is information that ranks the patient's degree of independence or the severity of their disability. The patient's degree of independence is ranked, for example, by the Functional Independence Measure (FIM). The severity of their disability is ranked by the Stroke Impairment Assessment Set (SIAS). For example, when the FIM is used, the acquired rank information indicates one of seven levels, rank 1 to rank 7, for each of the 13 movement items. The acquired biological information is information determined based on heart rate or body movement. For example, the acquired biological information is heart rate, body movement, or exercise intensity.

[0014] The statistical information storage unit 142 is configured using a storage device such as a magnetic hard disk drive or a semiconductor storage device. The statistical information storage unit 142 stores a statistical database. FIG. 3 is a diagram showing a specific example of the statistical database. The statistical database is configured with a patient identifier, date information, rank statistical information, and vital statistics information. As described above, the patient identifier is an identifier for uniquely identifying a patient. The date information indicates the date on which the rank statistical information and vital statistics information were acquired.

[0015] In this embodiment, the rank statistical information indicates an integer value obtained by dividing the total score (91 points) of the FIM motor items (13 items) by 13, and rounding down the decimal point of the average value per item. Because 91 points are divided by 13, the acquired rank information indicates any of ranks 1 to 7. The vital statistics information indicates the average and standard deviation of the above-mentioned heart rate, body movement, or exercise intensity. The vital statistics information is, for example, the average heart rate per minute. In this embodiment, the rank statistical information will be used as an example of "rank information." Furthermore, the vital statistics information will be used as an example of "biological information."

[0016] 1 controls the operation of each part of the condition analyzing device 100. The control unit 120 is executed by a device including a processor such as a CPU and RAM. The control unit 120 executes a condition analyzing program to function as an acquisition unit 121, a derivation unit 122, and a superimposed display unit 123.

[0017] The acquisition unit 121 acquires acquired rank information and acquired biometric information. The acquisition method may be, for example, acquiring information input by the user as text, or acquiring information detected by another device such as a wearable device. The acquisition unit 121 acquires the above-mentioned rank statistical information from the acquired acquired rank information. The acquisition unit 121 acquires the above-mentioned biometric statistical information from the acquired acquired biometric information.

[0018] The derivation unit 122 derives range information indicating the range in which the vital statistics information is distributed for each rank. The superimposed display unit 123 displays the range information as an image (graph) for each rank, and also displays a predetermined image superimposed on the graph at a position corresponding to the vital statistics information of one patient within the range indicated by the graph. In this embodiment, a graph is used as an example of an image, but the image is not limited to this and may be any image that can indicate a range.

[0019] FIG. 4 is a diagram showing an example of a graph. The vertical axis of the graph shown in FIG. 4 indicates rank statistical information, and the horizontal axis indicates the range in which the vital statistics information is distributed. The bar graph indicates the range in which the vital statistics information exists. The derivation unit 122 derives the upper and lower limits of the vital statistics information for each rank, as well as quartile values ​​(first quartile, median, third quartile) as range information. If the vital statistics information is smaller than the median, it indicates that the vital statistics information is smaller than the vital information expected from the rank. If the vital statistics information is larger than the median, it indicates that the vital statistics information is larger than the vital information expected from the rank.

[0020] The superimposed display unit 123 displays a predetermined image (a black circle in FIG. 4) superimposed on the graph at a position corresponding to the vital statistics information of one patient within the range indicated by the graph. Here, the "one patient" refers to a patient designated in advance by the operator of the condition analyzing device 100, and the patient identifier identifying this patient is stored in a non-volatile memory (not shown) or the like.

[0021] Note that Figure 4 shows multiple black circles and arrows, which shows an example in which the history of the black circles is also displayed. In Figure 4, it is shown that the biometric information has increased as the rank has increased. In other words, it shows that there has been an improvement. Black circles are used as the predetermined images, but this is not limitative. The predetermined image may be any image that indicates that it corresponds to the biometric information of a patient.

[0022] As shown in Figure 4, by superimposing a specific image on the graph at a position corresponding to the vital statistics information of a patient within the range indicated by the graph, the relationship between the rank statistical information and the vital statistics information can be quantitatively visualized and the patient's position can be indicated. In the past, such quantification was not possible, and patients often underwent rehabilitation without fully understanding the situation and felt anxious. However, this embodiment allows patients to know their own position, which can alleviate anxiety and provide appropriate support for rehabilitation. Furthermore, displaying the history can contribute to improving patients' motivation for rehabilitation training.

[0023] 5 is a flowchart showing the processing flow of the condition analyzer 100. The acquisition unit 121 acquires acquired rank information and acquired biological information (step S101). The acquisition unit 121 acquires rank statistical information and biological statistical information from the acquired acquired rank information and acquired biological information (step S102). The derivation unit 122 derives range information (step S103), and further derives a quadrant position (step S104).

[0024] The superimposed display unit 123 displays the range information as a graph for each rank (step S105), and also displays a black circle superimposed on the graph at a position corresponding to the vital statistics information of one patient within the range indicated by the graph (step S106).

[0025] This embodiment makes it possible to get an idea of ​​the relative state of activity of a patient, such as whether their biometric data is located near the median or at the edge of a quartile, and therefore to determine whether patients of each rank are performing activities beyond those expected.

[0026] (Second embodiment) The second embodiment is configured to include an outlier removal unit that removes vital statistics information using a threshold value according to the distribution of the vital statistics information in addition to the configuration of the first embodiment. Figure 6 is a functional block diagram showing the functional configuration of a condition analyzer 100 according to the second embodiment. In the functional configuration of the condition analyzer 100 according to the second embodiment, the same functional components as those of the condition analyzer 100 of the first embodiment are assigned the same reference numerals.

[0027] The outlier exclusion unit 124 of the control unit 120 of the condition analyzer 100 excludes vital statistics information using a threshold value corresponding to the distribution of vital statistics information. For example, as described in Non-Patent Document 2, threshold values ​​are calculated using quartile values ​​(first quartile, median, third quartile) of the vital statistics data, and values ​​outside the threshold value are excluded. FIG. 7 is a diagram showing an example in which quartile values ​​are used as threshold values ​​corresponding to the distribution of vital statistics information. The vertical axis of the graph shown in FIG. 7 indicates rank statistical information, and the horizontal axis indicates the range in which the vital statistics information is distributed. The rank statistical information shown in FIG. 7 indicates the total score (91 points) of the FIM motor items (13 items).

[0028] As shown in Figure 7, the outlier removal unit 124 derives the first quartile, median, and third quartile from the distribution of vital statistics information. Then, a threshold is determined using the width w between the first and third quartiles. As an example of the threshold, the value obtained by multiplying the width w by 1.5 is set as the third quartile, or 1.5w. The outlier removal unit 124 removes vital statistics information that does not fall within the range from below this threshold to 0.

[0029] FIG. 8 shows the graph after the exclusion. FIG. 8 shows that vital statistics information not falling within the above range has been excluded. In this embodiment, the exclusion is performed using a method using quartile values, but other methods other than quartile positions may be used if other methods are available for excluding outliers. Since the derivation unit 122 derives range information using the excluded vital statistics information, it is possible to suppress variability and reduce the impact on the analysis.

[0030] (Third embodiment) The third embodiment has a configuration that includes a trend display unit 125 in addition to the configuration of the second embodiment. Fig. 9 is a functional block diagram showing the functional configuration of the condition analyzing device 100 according to the third embodiment. In the functional configuration of the condition analyzing device 100 according to the third embodiment, the same functional configuration as that of the condition analyzing device 100 of the second embodiment is assigned the same reference numerals.

[0031] The control unit 120 according to the third embodiment includes a trend display unit 125. The trend display unit 125 detects one or more similar patients who have rank statistical information and vital statistics information at a certain time in the past that are similar to the rank statistical information and vital statistics information of a certain patient. The trend display unit 125 acquires rank statistical information of the detected similar patients after a predetermined period has elapsed since the certain time period, and displays the trend of the rank indicated by the acquired rank statistical information.

[0032] The trend display unit 125 first acquires the rank statistical information and vital statistics information of a patient. Then, the trend display unit 125 detects, in the statistical database, one or more similar patients who have rank statistical information and vital statistics information at a certain time in the past that are similar to the rank statistical information and vital statistics information of the patient. Here, rank statistical information that is "similar" to the rank statistical information of a patient is rank statistical information that indicates a value near the value indicated by the rank statistical information of the patient. Similarly, vital statistics information that is "similar" to the vital statistics information of a patient is vital statistics information that indicates a value near the value indicated by the vital statistics information of the patient.

[0033] For example, when the value indicated by the rank statistical information of one patient is a and b is a positive number, the neighborhood is the interval [ab, a+b]. The value of b may be set by the operator or may be determined by default. Similarly, when the value indicated by the rank statistical information of one patient is c and d is a positive number, the neighborhood in vital statistics information is the interval [cd, c+d]. The value of d may be set by the operator or may be determined by default.

[0034] As shown in Fig. 3, the statistical database includes dates. The trend display unit 125 then detects one or more similar patients who have rank statistical information and vital statistics information from a certain period in the past (for example, a predetermined period from the present (for example, two weeks ago)). Next, the trend display unit 125 obtains rank statistical information from the detected similar patient a predetermined period after the period (for example, the present). This allows, for example, 100 similar patients to be detected and the current rank statistical information of these similar patients to be obtained, allowing the trend display unit 125 to obtain the rank trend (for example, percentage) of similar patients after a predetermined period.

[0035] Fig. 10 is a diagram showing an example of display by the trend display unit 125. Fig. 10 is a diagram showing the rank ratio of similar patients after a predetermined period in the form of a pie chart. In the example of Fig. 10, 67% are rank 2, 28% are rank 3, and 5% are rank 4, which shows that the rank of a certain patient has a strong tendency to be rank 2. In this way, it is possible to determine whether a certain patient's condition is in a group where it is on an upward trend or a group where it is on a stagnant trend.

[0036] (Fourth embodiment) The fourth embodiment has a configuration that includes a prediction unit 126 in addition to the configuration of the third embodiment. The prediction unit 126 predicts rank statistical information and vital statistics information after a predetermined period has elapsed from the rank statistical information and vital statistics information by learning using rank statistical information and vital statistics information at a certain time in the past and rank statistical information and vital statistics information after a predetermined period has elapsed from that time.

[0037] 11 is a functional block diagram showing the functional configuration of the condition analyzing device 100 according to the fourth embodiment. In the functional configuration of the condition analyzing device 100 according to the fourth embodiment, the same functional components as those in the condition analyzing device 100 according to the third embodiment are denoted by the same reference numerals.

[0038] The control unit 120 according to the fourth embodiment includes a prediction unit 126. The prediction unit 126 includes a learning device. For example, random forest or gradient boosting may be used as the learning device. Biostatistical information of other patients may be used as the explanatory variable, and rank statistical information of other patients acquired after the date of acquisition of the biostatistical information may be used as the objective variable. For example, by performing learning using rank statistical information two weeks after the acquisition of the biostatistical information as the objective variable, a learning model (hereinafter referred to as "learning model A") that predicts rank statistical information about two weeks in the future from the biostatistical information can be acquired.

[0039] If a new learning model (hereinafter referred to as "learning model B") is obtained by changing the objective variable to the rank statistical information four weeks later and performing new learning, it will be possible to use the biostatistical information of any patient to predict the rank statistical information two weeks later from the time of measurement from learning model A. In addition, it will be possible to predict the rank statistical information four weeks later from learning model B.

[0040] It should be noted that by using rank statistical information as an explanatory variable and biological statistical information as a response variable for learning, biological statistical information can be predicted from rank statistical information.

[0041] Images showing the predicted rank statistical information and vital statistics information two and four weeks later may be superimposed on the graph. FIG. 12 is a diagram showing an example of a graph on which predicted rank statistical information and vital statistics information are superimposed. The vertical axis of the graph shown in FIG. 12 indicates the rank statistical information, and the horizontal axis indicates the range in which the vital statistics information is distributed. White circles indicate the patient's current rank statistical information and vital statistics information, and black circles indicate the predicted rank statistical information and vital statistics information. FIG. 12 shows, as an example, prediction results for two weeks, four weeks, and six weeks later. In the case of FIG. 12, both the rank statistical information and vital statistics information are predicted to increase from the current values.

[0042] FIG. 13 is a diagram showing the correlation coefficients of predicted values. FIG. 12 shows the correlation coefficients for the rank statistical information and vital statistics information after two weeks, four weeks, and six weeks, respectively. The correlation coefficient for the rank statistical information after two weeks was approximately 0.8, and the correlation coefficient for the vital statistics information after six weeks, the furthest time from the time of acquisition, was approximately 0.72, indicating high reliability. This indicates that the prediction unit 126 can predict the condition of a patient up to six weeks into the future with high reliability.

[0043] In this way, according to this embodiment, by making it possible to predict rank statistical information and biostatistical information, it is possible to provide suitable support for rehabilitation, such as assisting in the creation of rehabilitation treatment plans and contributing to improving patients' motivation for rehabilitation training.

[0044] (Fifth embodiment) The fifth embodiment has a configuration that includes a complementing unit 127 in addition to the configuration of the fourth embodiment. The complementing unit 127 complements the biostatistical information from the rank statistical information or complements the rank statistical information from the biostatistical information by learning the rank statistical information and biostatistical information for each patient.

[0045] 14 is a functional block diagram showing the functional configuration of the condition analyzing device 100 according to the fifth embodiment. In the functional configuration of the condition analyzing device 100 according to the fifth embodiment, the same functional components as those in the condition analyzing device 100 according to the fourth embodiment are denoted by the same reference numerals.

[0046] The control unit 120 according to the fifth embodiment includes a complementing unit 127. When there is a loss of data, such as the absence of either or both of the patient's rank statistical information and biological data, the complementing unit 127 complements the missing information using a learning device. The complementing unit 127 may complement the missing information by using, for example, single imputation or multiple imputation, as described in Reference 3.

[0047] If rank statistical information is missing, the complementing unit 127 can complement the rank statistical information. Such complementing can be performed by using the method of Reference 3, with the rank statistical information as the objective variable and the vital signs statistical information as the explanatory variable.

[0048] Similarly, if vital statistics information is missing, the complementing unit 127 may complement the vital statistics information. In this case, the complementing unit 127 may use one variable of the vital statistics information (one of heart rate, body movement, exercise intensity, etc.) as a dependent variable, and may train the other vital statistics information and rank statistics information as explanatory variables, and may repeatedly complement the vital statistics information by changing the dependent variable so that all missing vital statistics information is complemented.

[0049] Even when both vital statistics information and rank statistical information are missing, the completion unit 127 can complete all of the missing information by completing them in order. Furthermore, when it is considered appropriate to leave missing information of rank statistical information or vital statistics information missing, completion may be performed only on specific vital statistics information or rank statistical information.

[0050] In this embodiment, the ranks of each FIM item used in the rank statistical information may not be entered by the operator for some reason, in which case the rank statistical information becomes missing information. Furthermore, when biostatistical information uses data from a wearable device, data may not be acquired due to the wearable device not being worn or a measurement error, in which case the biostatistical information becomes missing information. Even if missing information occurs due to such circumstances, this embodiment makes it possible to complement the missing information, thereby providing a highly reliable analysis of the condition.

[0051] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Industrial Applicability]

[0052] The present invention is applicable to rehabilitation analysis and diagnosis. [Explanation of symbols]

[0053] 100: condition analyzer, 110: display unit, 120: control unit, 121: acquisition unit, 122: derivation unit, 123: superimposition display unit, 124: outlier removal unit, 125: trend display unit, 126: prediction unit, 127: complementation unit, 141: patient information storage unit, 142: statistical information storage unit

Claims

1. an acquisition unit that acquires ranking information that ranks the degree of independence or the severity of disability of a patient and biological information of the patient; a derivation unit that derives range information indicating a range in which the biometric information is distributed for each rank; a superimposition display unit that displays the range information as an image for each rank, and that displays a predetermined image superimposed on the image at a position corresponding to the biological information of one patient within the range indicated by the image; A condition analysis device equipped with the above.

2. The condition analyzing device according to claim 1 , further comprising an outlier removing unit that removes biological information using a threshold value according to the distribution of the biological information.

3. 3. The condition analysis device according to claim 1, further comprising a trend display unit that detects one or more similar patients who have rank information and biometric information at a certain time point in the past that are similar to the rank information and biometric information of a patient, acquires rank information of the detected similar patients after a predetermined period of time has elapsed since the certain time point, and displays the trend of the rank indicated by the acquired rank information.

4. The condition analyzer according to claim 1, further comprising a prediction unit that predicts rank information and biological information after a predetermined period of time has elapsed using rank information and biological information from a certain period of time in the past and rank information and biological information from a certain period of time after the certain period of time.

5. 2. The condition analyzer according to claim 1, further comprising a complementing unit that complements the biological information from the rank information or complements the rank information from the biological information.

6. 2. The condition analyzer according to claim 1, wherein the rank information is determined based on the Functional Independence Measure (FIM) or the Stroke Impairment Assessment Set (SIAS), and the biological information is determined based on heart rate or body movement.

7. an acquiring step of acquiring rank information that ranks the patient's independence or the severity of the disability and the patient's biological information; a derivation step of deriving range information indicating a range in which the biometric information is distributed for each rank; a superimposed display step of displaying the range information as an image for each rank, and displaying a predetermined image superimposed on the image at a position corresponding to the biological information of one patient within the range indicated by the image; A condition analysis method comprising:

Images