Method and system for predicting effect of rehabilitation for improving body pain of stroke patient

Abstract

To provide a system that accumulates subjective data of characteristics and degree of pain before / after executing rehabilitation of a stroke patient, and calculates an estimated value regarding prognosis of persistent pain after stroke.SOLUTION: A system calculates an NPI (Neuropathic Pain Inventory) value (NPI denotes any value of 0 to 10) showing an index for types of pain before rehabilitation by each individual pain of a plurality of control patients, and an NRS (Numeric Rating Scale) value showing an index of overall pain regardless of types of pain after rehabilitation of new patients from similarity between a plurality of control patients and new patients on the basis of an NRS value after rehabilitation.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a method and system for predicting the effect of rehabilitation aimed at improving post-rehabilitation pain in stroke patients. More specifically, the present invention relates to a method and system for providing medical professionals such as physical therapists with a guideline for considering whether or not to implement rehabilitation that places a heavy burden on patients by calculating a predicted value of a comprehensive pain index of a new patient after rehabilitation based on the similarity between response data that subjectively indexes the bodily pain before rehabilitation, obtained through a questionnaire survey of a control patient group that had previously undergone rehabilitation, and similar response data of a new patient before rehabilitation. [Background technology]

[0002] When a stroke occurs, in addition to motor paralysis appearing as an aftereffect, "pain (= post-stroke pain)" may occur, which often becomes an obstacle to rehabilitation for motor paralysis. Therefore, rehabilitation to relieve post-stroke pain is also important.

[0003] The basic procedure for general rehabilitation involves evaluating the symptoms present in the case (patient), estimating the prognosis for each symptom, and then planning and implementing the most appropriate rehabilitation program based on this. However, in the case of rehabilitation for pain that occurs after a stroke, the prognosis varies greatly from patient to patient, so the current situation is that physical therapists and other medical professionals are formulating rehabilitation programs by trial and error, based on their past experience and the subjective complaints of pain from new patients.

[0004] In addition, therapies for improving physical pain include, first, physical therapy centered on rehabilitation, second, drug therapy using nerve-acting drugs, and third, surgical methods such as surgery, but rehabilitation places a heavy burden on the patient, so when no effect is expected, it is preferable to choose drug therapy or surgical methods. Therefore, if a certain degree of indication could be obtained in advance for new patients regarding the effect of rehabilitation and the appropriateness of its implementation, it would be extremely beneficial for both patients and medical institutions.

[0005] Here, as a prior art for grasping and analyzing the physical pain felt by humans, a method for generating a model for monitoring and discriminating pain based on the patient's brain waves, such as "Pain discrimination using trend analysis" disclosed in Patent Document 1, has been proposed. In addition, as a prior art for providing an appropriate rehabilitation plan according to the patient's symptoms and condition, a "Rehabilitation approach suggestion method" disclosed in Patent Document 2 has been proposed. However, although the former makes it possible to objectively discriminate the quality and quantity of pain using trend analysis of the target patient's biosignals, it does not provide a prediction of the effectiveness or suitability of rehabilitation. Also, although the latter makes it possible to provide an appropriate rehabilitation plan by inputting information on the current physical condition and rehabilitation wishes of the target patient, it does not provide a prediction of the effectiveness or suitability of rehabilitation itself. [Patent Document 1] Republished Patent Publication No. 2019 / 009420 [Patent Document 2] JP 2022-87965 A

[0006] The prognosis of post-stroke pain depends on the individual pathology behind the pain. Previous studies have suggested that brain image analysis data identifying the damaged area of ​​the brain and electrophysiological data clarifying abnormal findings in pain pathways can be used to estimate the prognosis of post-stroke pain. For example, it has been reported that when damage to the brain area called the pulvinar is combined with a decrease in the amplitude of brain potentials caused by pain stimuli, there is an 87% probability of a poor prognosis of post-stroke pain (Vartiainen N et al. Brain. 2016).

[0007] Although such clinical evaluation and data analysis may be able to predict the prognosis of post-stroke pain, it is not realistic to carry out such a large-scale clinical evaluation in a busy rehabilitation setting. DISCLOSURE OF THEINVENTION [Problem to be solved by the invention]

[0008] In this way, since it is difficult to identify the affected area and cause of physical pain after a stroke, attempts to distinguish and analyze it have so far focused on objective physiological data. However, it is not easy to identify the cause, and in reality it is difficult to even provide a guideline for predicting the effectiveness or appropriateness of rehabilitation for new patients.

[0009] On the other hand, after stroke, patients routinely complain of a wide variety of pain characteristics, such as "stinging" or "burning." However, in the field of rehabilitation, such complaints have been limited to listening to only in order to establish communication with patients, on the grounds that "subjective opinions are unreliable."

[0010] However, looking back at past research, it appears that the nature of pain differs depending on the damaged area of ​​the brain (Klit H et al. Pain.2011), and that there are specific pain characteristics associated with muscle and joint problems (Choi-Kwon S et al. Acta Neurol Scand. 2017 Apr;135(4):419-425.). If there is a correlation between subjective data and pathological mechanisms, it may be possible to use this subjective data to estimate the prognosis of post-stroke pain.

[0011] From this perspective, the present invention has an objective of providing an information processing method and system using software that can provide patients and medical professionals with an indication of the effectiveness and appropriateness of rehabilitation by accumulating subjective data on the nature and severity of pain reported by patients who have already undergone rehabilitation, and calculating a predicted value for the prognosis of post-stroke pain through statistical processing of the similarity with the subjective data of new patients before rehabilitation.

[0012] In order to solve the above problem, the invention described in claim 1 of the present application is as follows: To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) which indicates the index for each type of pain, The NRS (Numeric Rating Scale) value indicates an overall pain index regardless of the type of pain, Based on the similarity with the NPI value of new patient N (new patient) before rehabilitation, To calculate the predicted NRS value after rehabilitation for new patient N Software-based information processing method And, In the software implemented in an information processing device connected to at least a database and an input / output terminal, The control patient C's For each x type of pain, NPI value before rehabilitation is b(before)1~bx and NRS score before rehabilitation is Pb (Before Pain) and NRS score after rehabilitation is Pa (Pain after) From , For the i-th control patient C(i), where the total number of control patients C is n, The x-dimensional vector consisting of the sequence of b1 to bx is defined as CVb(i), The Pb is designated as CPb(i), The Pa is CPa(i), Input from an I / O terminal New patient N, Let NVb be an x-dimensional vector consisting of a sequence of b1 to bx. The NRS value before rehabilitation was defined as NPb. The cosine similarity between the CVb(i) and the NVb is Sim(i). After carrying out the above process, From the Sim(i) and NPb of all control patients C, the NRS value after rehabilitation of new patient N, NPa(all), is calculated. NPa(all)=NPb×{(Σ(i=1 to n)[(Sim(i) / Σ(i=1 to n)Sim(i))×(CPa(i) / CPb(i))]} By the formula The present invention is characterized in that the calculation can be performed by inputting and outputting the calculated value from an input / output terminal. .

[0013] The Neuropathic Pain Inventory (NPI) value, which indicates an index by type of pain, is an index obtained from a questionnaire survey of patients using the Neuropathic Pain Symptom Inventory, a tool used to evaluate the severity of neuropathic pain (NP). For example, as shown in Figure 1, for each of 10 types of pain, from "Q1 burning pain" to "Q10 tingling and numbing pain," patients are asked to answer on an 11-point scale from "0 no pain" to "10 worst pain imaginable." Note that there are several types of neuropathic pain evaluation tools, and the types and number of pains asked vary depending on the tool, so the number of types of pain is not necessarily limited to 10.

[0014] On the other hand, the NRS value is an index obtained by having the patient subjectively select the overall level of pain on an 11-point scale (0 to 10) without regard to the type of pain. In the present invention, a predicted NRS value after rehabilitation for new patient N is calculated from the similarity between the NPI values ​​indicating the index for each type of pain before rehabilitation for control patient C who has undergone rehabilitation and new patient N who has not yet undergone rehabilitation, and the ratio (degree of improvement) of the NRS values ​​before and after rehabilitation for control patient C, and this is compared with the NRS value before rehabilitation to provide a guideline (criteria) for judging the appropriateness of implementing rehabilitation and the expected degree of improvement.

[0015] As mentioned above, since a correlation has been found between the area of ​​damaged brain and the type of pain, it can be assumed that stroke patients who show similar trends in NPI values ​​for each type of pain also have similar causes of symptoms, and if the causes of symptoms are similar, it is expected that a certain correlation will also exist in the effectiveness of rehabilitation (the ratio of NRS values ​​before and after rehabilitation).

[0016] On the other hand, individual types of pain do not correlate with the overall level of pain, and the sum of multiple types of pain does not necessarily reflect the overall level of pain. Depending on the patient, the NRS value may be high even though the NPI values ​​of each type of pain are small, and conversely, there are cases where the NRS value is not very high even though there is pain that shows an extremely high NPI value, and cases where the NRS value is 1 or more even though the NPI value is not 0. This is natural since both the NPI value and the NRS value are indices based on the patient's subjective perception.

[0017] Therefore, the present invention relates to claim 1. Software-based information processing method In this system, the pre-rehabilitation NPI values ​​of patients in control patient group C and the pre-rehabilitation NPI value of new patient N are replaced with vectors with the dimensionality equal to the number of types of pain, and a predicted post-rehabilitation NRS value for new patient N is calculated by multiplying the pre-rehabilitation NRS value of new patient N by a coefficient calculated from the similarity between the pre-rehabilitation vectors of all control patients C and new patient N and the rate of change in the NRS value of control patient C before and after rehabilitation.

[0018] Cosine similarity is used to calculate the similarity of NPI values ​​as vectors. Cosine similarity is calculated using the formula cos(θ) = A·B / (||A|| ||B||), where A·B is the dot product (inner product) of vector A and vector B, and ||A|| and ||B|| are the magnitudes (norms) of vector A and vector B, respectively.

[0019] Cosine similarity is a similarity calculation method mainly used for handling high-dimensional data, and has features such as being suitable for natural language processing such as calculating similarity between documents, and the calculation result is not affected by the size of the vector. Therefore, it is considered to be suitable for calculating the similarity of data with large numerical variation depending on the person, such as NPI values, which are indexes based on the subjective opinion of the respondent (patient). However, the method of calculating similarity is not limited to cosine similarity, and the adoption of other calculation methods with similar characteristics is not excluded.

[0020] Next, the invention described in claim 2 of the present application is as follows: To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) which indicates the index for each type of pain, The NRS (Numeric Rating Scale) value indicates an overall pain index regardless of the type of pain, Based on the similarity with the NPI value of new patient N (new patient) before rehabilitation, To calculate the predicted NRS value after rehabilitation for new patient N Software-based information processing method And, In the software implemented in an information processing device connected to at least a database and an input / output terminal, The control patient C's For each x type of pain, NPI value before rehabilitation is b(before)1~bx and NRS score before rehabilitation is Pb (Before Pain) and NRS score after rehabilitation is Pa (Pain after) From , For the i-th control patient C(i), where the total number of control patients C is n, The x-dimensional vector consisting of the sequence of b1 to bx is defined as CVb(i), The Pb is designated as CPb(i), The Pa is CPa(i), Input from an I / O terminal For new patient N, Let NVb be an x-dimensional vector consisting of a sequence of b1 to bx. The NRS value before rehabilitation was defined as NPb. The cosine similarity between the CVb(i) and the NVb is Sim(i). After carrying out the above process, moreover, The ratio for narrowing down the control patients C with the highest Sim(i) is Input from an I / O terminal Let x be any value, If the value corresponding to x in the standard normal distribution table is the standard Z value z, then A first step of calculating the average value Sim(ave) and standard deviation of the similarity Sim (i = 1 to n) between the CVb(i) and the NVb of all control patients C; A second step of calculating a standard similarity SimZ corresponding to z from the Sim(ave) and standard deviation, and extracting control patients c (the number of which is m) having Sim(i) equal to or greater than SimZ; From the Sim(i) and NPb of the control patient c(i) (i = 1 to m), the NRS value after rehabilitation of the new patient N, NPa(lim), is calculated. NPa(lim)=NPb×{(Σ(i=1 to m)[(Sim(i) / Σ(i=1 to m)Sim(i))×(CPa(i) / CPb(i))]} The third step is to calculate the and outputting the calculated NPa(lim) from an input / output terminal. It is.

[0021] The method for calculating the predicted value of the NRS value after rehabilitation according to the invention of claim 2. Software-based information processing method From all the control patient groups C, the top control patients c whose pre-rehabilitation NPI value vector similarity is close to that of the new patient N by an arbitrary value x are extracted, and the same as the invention described in claim 1 is performed. process The predicted value is It is possible to calculate and output .

[0022] Although the NPI value is an index based on the subjective responses of the patient, as mentioned above, given that a correlation has been found between the damaged area of ​​the brain and the type of pain, a certain correlation can also be expected with the effect of rehabilitation. Therefore, if the number of control patient groups C is sufficiently large, it is possible to calculate a more accurate prediction value by narrowing down the data used to calculate the prediction value to only the data of control patients c whose pre-rehabilitation NPI value vectors are highly similar to those of new patient N.

[0023] The invention described in claim 3 relates to the method for calculating the predicted value of the NRS value after rehabilitation described in claim 2. Software-based information processing method Among the control patients c having Sim(i) equal to or greater than SimZ extracted in the second step, the Sim(i) of the control patient having the smallest Sim(i) is θ(i)=10*[1-cos -1 (Sim(i)) / (π / 2)] The value of θ(i) calculated by the formula is used as the reliability of NPa(lim). It is possible to output from an input / output terminal. It is characterized by:

[0024] In the present invention, the method according to claim 2 Software-based information processing By converting the similarity of the control patient with the smallest similarity among the control patients c narrowed down by 1 into a reliability using the above formula, it is possible to show that the calculated predicted value can be expected to have a reliability of at least the value of θ, and this can be used as a guideline for determining whether rehabilitation is appropriate for new patient N.

[0025] Next, the invention according to claim 4 is as follows: A system for predicting the effect of rehabilitation aimed at improving physical pain caused by stroke, the system calculates a predicted NRS value after rehabilitation for a new patient N from the similarity between the Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) indicating an index for each type of pain before rehabilitation for a control patient C (control patient) group who has already undergone rehabilitation, the Numeric Rating Scale (NRS) value indicating an overall index of pain regardless of type of pain, and the NPI value of the new patient N before rehabilitation, The system comprises at least a database, an information processing device, software, and an input / output terminal; The database comprises at least For each of the control patients C, a total of n in number, Cb(before)1~bx, which is the NPI value before rehabilitation, CPb (Pain before) is the NRS value before rehabilitation, and CPa (Pain after) is the NRS value after rehabilitation. I remember, The software, in the information processing device, from the input / output terminal, For new patient N, The NPI value before rehabilitation, Nb(before)1~Nbx, NPb (Pain before) is the NRS value before rehabilitation Upon receiving the input, For the i-th control subject C(i), The x-dimensional vector consisting of the sequence of Cb1 to Cbx is defined as CVb(i), The CPb is designated as CPb(i), The CPa is designated as CPa(i), For new patient N, The x-dimensional vector consisting of the sequence of numbers Nb1 to bx is defined as NVb, Calculate the cosine similarity Sim(i) between the CVb(i) and the NVb; From Sim(i) of all control patients C and the above NPb, NPa(all), which is the NRS value after rehabilitation of new patient N, is calculated. NPa(all)=NPb×{(Σ(i=1 to n)[(Sim(i) / Σ(i=1 to n)Sim(i))×(CPa(i) / CPb(i))]} and outputting the calculated value from the input / output terminal. It is characterized by:

[0026] The invention according to claim 5 is as follows: A system for predicting the effect of rehabilitation aimed at improving physical pain caused by stroke, the system calculates a predicted NRS value after rehabilitation for a new patient N from the similarity between the Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) indicating an index for each type of pain before rehabilitation for a control patient C (control patient) group who has already undergone rehabilitation, the Numeric Rating Scale (NRS) value indicating an overall index of pain regardless of type of pain, and the NPI value of the new patient N before rehabilitation, The system comprises at least a database, an information processing device, software, and an input / output terminal; The database comprises at least For each of the control patients C, a total of n in number, Cb(before)1~bx, which is the NPI value before rehabilitation, CPb (Pain before) is the NRS value before rehabilitation, and CPa (Pain after) is the NRS value after rehabilitation. I remember, The software, in the information processing device, from the input / output terminal, For new patient N, The NPI value before rehabilitation, Nb(before)1~Nbx, NPb (Pain before) is the NRS value before rehabilitation Upon receiving the input, For the i-th control subject C(i), The x-dimensional vector consisting of the sequence of Cb1 to Cbx is defined as CVb(i), The CPb is designated as CPb(i), The CPa is designated as CPa(i), For new patient N, The x-dimensional vector consisting of the sequence of numbers Nb1 to bx is defined as NVb, Calculate the cosine similarity Sim(i) between the CVb(i) and the NVb; moreover, An arbitrary value x is input as a ratio for narrowing down the control patients C having the top Sim(i) from the input / output terminal, Let the value corresponding to x in the standard normal distribution table be the standard Z value z. Calculate the average value Sim(ave) and standard deviation of the similarity Sim (i = 1 to n) between the CVb(i) and the NVb of all control patients C, A standard similarity SimZ corresponding to z is calculated from the Sim(ave) and the standard deviation, and control patients c (the number of patients is m) having Sim(i) equal to or greater than SimZ are extracted. From the Sim(i) and the NPb of the control patient c(i) (i = 1 to m), NPa(lim)=NPb×{(Σ(i=1 to m)[(Sim(i) / Σ(i=1 to m)Sim(i))×(CPa(i) / CPb(i))]} The NRS value after rehabilitation of new patient N, NPa(lim), was calculated using the formula: The input / output terminal can output the data. It is characterized by:

[0027] Finally, the invention according to claim 6 is as follows: 6. A system for calculating a predicted value of an NRS value after rehabilitation of a new patient N according to claim 5, wherein the software is configured to: Among the control patients c having Sim(i) equal to or greater than SimZ, the Sim(i) of the target patient having the smallest Sim(i) is θ(i)=10*[1-cos -1 (Sim(i)) / (π / 2)] The value of θ(i) calculated by the formula is set as the reliability of the NVa(lim), The information can be output from the input / output terminal. Effect of the Invention

[0028] The present invention Software-based information processing method , and software A system that implements this can provide physical therapists and other medical professionals with predicted values ​​for the effectiveness of rehabilitation prior to the start of rehabilitation for new patients complaining of post-stroke pain. This makes it possible to provide numerical evidence for the appropriateness of rehabilitation decisions and planning, which previously relied solely on the experience of medical professionals and the complaints of patients, and is expected to improve treatment outcomes by selecting appropriate therapies and reduce the burden on patients and medical professionals. BEST MODE FOR CARRYING OUT THEINVENTION

[0029] Hereinafter, the predicted value of the NRS value after rehabilitation of a new stroke patient according to the present invention will be calculated. Implemented the software An embodiment of the system will be described with reference to the drawings.

[0030] FIG. 2 is a system configuration diagram of a system (hereinafter, referred to as "this system 1") according to one embodiment of the present invention. This system 1 is broadly composed of an information processing device P that implements software S capable of executing the method of calculating a predicted value of an NRS value according to the present invention, a database D, and an input / output terminal T, and allows data communication between them. However, these system configurations are merely examples, and each component is not limited to being physically independent. For example, the information processing device P and the input / output terminal T may be integrated as an information processing terminal, and the installed software S may be configured to be able to communicate with the database D through a network line, or the information processing device P and the database D may be provided in a server, and the ASP system may be configured to enable the software S to be executed from the input / output terminal T through a network line.

[0031] 3 illustrates an example of the data configuration in the database of the present system 1 according to the embodiment. The database D stores a total of 12 pieces of data for each of the control patients C (the total number of which is n), including a 10-dimensional vector (CVb) consisting of a sequence of 10 NPI values ​​before rehabilitation obtained from a questionnaire survey (in this embodiment, 10 types of pain are surveyed), and NRS values ​​(CPa, CPb) before and after rehabilitation.

[0032] Figures 4 and 5 show schematic flows of data input / output and processing in the system according to the present invention. Also, Figure 6 is a flowchart explaining the data processing process in the first embodiment of the present system 1 according to claim 4 of the present invention by dividing it into more detailed steps.

[0033] First embodiment 4, in the present system 1 according to the first embodiment, a 10-dimensional vector NVb consisting of a sequence of NPI values ​​for a specific new patient N before rehabilitation and an NRS value NPb before rehabilitation are input from the input / output terminal T to an information processing device P, whereby the software S refers to the data (CVb, CPa, CPb) of the target patient group C stored in the database D, calculates NPa(all), which is the NRS value after rehabilitation of the new patient N calculated based on the data of all the control patients C, and can output it to the input / output terminal T. In the following, the total number of control patients C is assumed to be n.

[0034] In FIG. 6, first, in the flow of "10. System startup and initialization," the system is started in step 101, and in step 102, all variables that were set in the previous data processing are initialized, and a new data processing flow is started.

[0035] First, in the flow of "11. Read control patient data", in step 111, CVb, CPb, and CPa, which are 12×n pieces of data for n control patients C in database D, are read in order and set as reference values ​​(1) to (12n), respectively. For example, CVb(i) of the i-th control patient C(i) is a total of 12 values, consisting of 10 NPI values ​​(b1 to b10) before rehabilitation and CPb(i) and CPa(i), which are NRS values ​​before and after rehabilitation. At this time, for example, reference value (1) represents b1 of the i-th control patient C, and reference value (12n) represents the CPa of the n-th patient.

[0036] Next, in step 112, for i = 1, 2, 3, n, the sequence from the reference value (1+12(i-1)) to the reference value (10+12(i-1)) of all control patients C is defined as CVb(i). This CVb(i) represents a vector whose component element is the NPI value of the i-th control patient. The above numerical operations are intended to identify the data of the i-th control patient (i) from n people × 12 values.

[0037] Next, in step 113, it is determined whether the reference value (11+12(i-1)), which is the NRS value of the i-th control patient C before the start of rehabilitation, is "0". If it is not "0", in step 115, the reference value (12+12(i-1)), which is the NRS value after the rehabilitation, is divided by the reference value (11+(12i-1) to obtain the NRS ratio (i). On the other hand, if the reference value (12i-1) is "0", it is set to "1" in step 114, and the NRS ratio (i) is calculated in step 115. This NRS ratio (i) indicates the change in the NRS value of the i-th control patient (i) before and after rehabilitation, that is, the degree of improvement in pain. Note that the processes in steps 113 and 114 are intended to correct such seemingly irrational indexes due to subjective responses, since there may be cases in which some control patients C answer the NRS value as "0" even though any of the NPI values ​​is "1" or more, as described above.

[0038] Next, in step 116, the formula Calculation is performed using JPEG2025009221000018.jpg2773 to calculate the vector size (i). The vector size (i) indicates the size of the 10-dimensional vector CVb(i). Here, "a" indicates the order of the NPI value obtained from the questionnaire survey of the i-th control patient (i). This ends the flow of "11. Read reference patient data" and proceeds to the flow of "12. Input new patient data".

[0039] Next, in the flow of “12. Input new patient data”, in step 121, a total of 11 values, namely, NVb (NVb is the 10 NPI values ​​(b1 to b10) obtained from a questionnaire survey before rehabilitation is implemented) and NPb (NRS value) of new patient N are input from input / output terminal T.

[0040] Next, the flow proceeds to "Calculate 13 NPa (all)". In step 131, of the data input in step 121, NVb (10 NPI values) are set as input values ​​(1) to (10), and NPb, which is the NRS value, is set as input value (11).

[0041] Next, in step 132, the sequence of input values ​​(1) to (11) is set as DS(n+1), and in step 133, the formula Calculation is performed using JPEG2025009221000019.jpg2773, and the calculation result is set as the vector size (n+1).

[0042] Next, in step 134, it is determined whether the values ​​of vector size (i) and vector size (n+1) are "0". As described above, vector size (i) is the size of CVb(i) in which the NPI value of the i-th control patient C before rehabilitation is taken as a 10-dimensional vector, and vector size (n+1) is the size of NVb in which the NPI value of the new patient N before rehabilitation is taken as a 10-dimensional vector. Here, if vector size (i) and vector size (n+1) are not both "0", the similarity between the vectors can be calculated, so the cosine similarity between the two can be Sim(i). On the other hand, if both are "0", there is no difference between the two, so the cosine similarity Sim(i)=1. And if only one of them is "0", the similarity between the vectors cannot be calculated, so the cosine similarity Sim(i)=0.

[0043] Next, in step 135, the formula JPEG2025009221000020.jpg2641 calculates the Sim sum by adding up the similarity between the vectors of CVb of all control patients C (n people) in the database and NVb of new patient N, and then in the next step 136, the formula The NRS ratio (all) is calculated using JPEG2025009221000021.jpg27165. The NRS ratio (all) obtained by this calculation is a coefficient predicting the degree of improvement in NRS based on the similarity between the CVb of all control patients C (n people) and the NVb of the new patient N. Then, in the final step 137, the NRS ratio (all) is multiplied by NPb, which is the NRS value before rehabilitation of the new patient N, to obtain NPa(all), which is the predicted NRS value after rehabilitation of the new patient N.

[0044] Since the NRS value is an integer exponent between "1" and "10", in this embodiment, the predicted value obtained here is rounded off to an integer, and if a predicted value of 11 or more is calculated, this is set to "10", and then in "output 14 NPa(all)", NPa(all) is output from the input / output terminal T to complete the process. The formula that integrates the above data processing is NPa(all)=NPb×{(Σ(i=1 to n)[(Sim(i) / Σ(i=1 to n)Sim(i))×(CPa(i) / CPb(i))]} It is.

[0045] Second Embodiment 7 and 8 are flowcharts for explaining the data processing process in the second embodiment of the present system according to claims 5 and 6, broken down into more detailed steps. Note that the process from "20 System startup / initialization" to "23 NPa(all) calculation" shown in Fig. 7 is basically the same as the flowchart of the first embodiment shown in Fig. 5, and therefore detailed explanations will be omitted.

[0046] In the system according to the second embodiment, a process is performed to narrow down the control patients C (n people) to the top patients (m people) with the greatest cosine similarity between the 10-dimensional vector obtained from the new patient N and the NPI value, and then a predicted value NPa(lim) of the post-rehabilitation NRS value of the new patient N is calculated. This is multiplied by NPb, the NRS value of the new patient N, to calculate a more accurate post-rehabilitation NRS value NPa(lim), and further the reliability of this NPa(lim) is also calculated.

[0047] In the second embodiment, after calculating NPa(all) of a new patient N in the final step 237 of the flowchart in Fig. 7, the process proceeds to the flowchart in Fig. 8, where an arbitrary variable x is input from the input / output terminal T in step 241 of process "24 Input x value". The variable x is a coefficient for narrowing down the control patients C (n people) to the top x% of patients (m people) in descending order of Sim(i), which is the cosine similarity between CVb and NVb, which are 10-dimensional vectors consisting of NPI values ​​before rehabilitation.

[0048] Next, proceed to the "25 NPa(lim) calculation" step. In step 251, the Z value corresponding to the variable x is searched for in the standard normal distribution table, and this is set as the "reference Z value z." The Z value is a unit that expresses the distance from the average value as a multiple of the standard deviation, and in statistics, it is a measure of how far each piece of data is from the overall average.

[0049] Next, in step 252, Sim total / n is set as Sim(ave), and in step 253, the formula Calculate the Sim standard deviation using JPEG2025009221000022.jpg2582.

[0050] Next, in step 254, z×Sim standard deviation+Sim(ave) is set as the reference similarity SimZ. After that, in step 255, it is determined whether Sim(i)-SimZ is equal to or greater than "0", and if it is "yes (equal to or greater than 0)", control patients C (the number of which is m) who have Sim(i) equal to or greater than SimZ are extracted. This "m" is an integer indicating the number of control patients C who have a similarity equal to or greater than SimZ.

[0051] Here, "m = m + 1", Sim top (m) = Sim (i), NRS ratio top (m) = NRS ratio (i), and then determine whether Sim top (m) - Sim top minimum is "0" or less, and if it is "yes (0 or less)", Sim top (m) = Sim top minimum. Incidentally, in the initial setting of the variables, Sim top (i) = 0 (i = 1, 2 ... n), and Sim top minimum (the smallest Sim (i) value among Sim (i) equal to or greater than SimZ) = 1. Also, NRS ratio top (i) = 0 (i = 1, 2 ... m). In other words, of the n Sim top (i) and NRS ratio top (i) prepared, only the number of control patients C with a similarity equal to or greater than SimZ is used in the calculations after this step 255. Note that if the two branches in step 255 are "no", nothing is done.

[0052] Next, in step 256, the formula The Sim top elements (i) for all the control patients C who have a similarity equal to or higher than the extracted reference similarity are summed up to obtain the Sim top sum, and in step 257, the formula The NRS ratio (lim) is calculated using JPEG2025009221000024.jpg18125. The "top NRS ratio" is a calculation that is the same as that in step 236, in which the ratio of NRS values ​​before and after rehabilitation (i.e., the degree of improvement in pain) NRS(all) is calculated for all control patients C, but this calculation is performed for control patients C having a similarity equal to or higher than the extracted standard similarity.

[0053] Then, in step 258, similar to the first embodiment, the input value (11) of new patient N, i.e., the NRS value before rehabilitation, is multiplied by the upper NRS ratio to calculate NPa(lim), which is the predicted NRS value after rehabilitation of new patient N based on the extracted data of highly similar control patient C. Since the NRS value is an integer exponent between "1" and "10", the decimal part of the obtained predicted value is rounded off to an integer, similar to the first embodiment, and if a predicted value of 11 or more is calculated, this is set to "10" and the process is completed. The formula integrating the data processing in the second embodiment described above is NPa(lim)=NPb×{(Σ(i=1 to m)[(Sim(i) / Σ(i=1 to m)Sim(i))×(CPa(i) / CPb(i))]} It is.

[0054] Next, the flow proceeds to "26 NPa (lim) reliability calculation". In step 261, the formula Using JPEG2025009221000025.jpg24165, the lower limit of the reliability of NPa(lim), which is the predicted value of the NRS value after rehabilitation of new patient N, is calculated. Here, "Sim upper minimum" refers to the "Sim upper minimum" calculated in step 255, i.e., the smallest Sim(i) value among Sim(i) equal to or greater than SimZ, converted into the reliability of NPa(lim). This value can take a value between "0" and "10," with the higher the value, the higher the reliability.

[0055] Finally, in "27 Output of reliability of NPa(all), NPa(lim), and NPa(lim)", the numerical values ​​of NPa(all), NPa(lim), and the lower limit of the reliability of NPa(lim) calculated in the above flow are output and displayed together on the input / output terminal T, completing all processes of the system related to the second embodiment.

[0056] If the NPa(lim) value calculated by this system for new patient N is greater than the patient's NRS value before rehabilitation, physical therapists or other medical professionals can determine that rehabilitation will not be effective and may even worsen the condition, and can choose another treatment method without rehabilitation. Conversely, if an NPa(lim) is calculated that is sufficiently smaller than the NRS value before rehabilitation and the lower limit of reliability is high, it can be determined that rehabilitation is likely to be effective and can be selected for implementation.

[0057] The above describes the specific configuration of the system for calculating the predicted NRS value after rehabilitation of a new stroke patient according to the present invention. However, the present invention is not limited to the above embodiment, and improvements or modifications are possible within the scope of the technical idea of ​​the present invention, and these belong to the technical scope of the present invention. [Industrial Applicability]

[0058] The present invention Implemented the software The system allows medical professionals, such as doctors and physical therapists, to calculate the predicted effect of rehabilitation treatment prior to treatment of a new patient complaining of stroke pain, and obtain a certain guideline for judging whether or not rehabilitation should be implemented. This reduces the psychological anxiety and burden on patients and medical professionals when implementing rehabilitation, and also avoids unnecessary burden on new patients who are not expected to benefit from rehabilitation. [Brief description of the drawings]

[0059] [Figure 1] Questionnaire items for the neuropathic pain severity assessment tool [Diagram 2] Schematic diagram showing the configuration of this system [Diagram 3] Data structure in the database [Figure 4] Overview of data processing steps of the system according to the first embodiment [Diagram 5]Overview of data processing steps of the system according to the second embodiment [Figure 6] A detailed flowchart of the data processing steps of the system according to the first embodiment. [Figure 7] 13 is a detailed flowchart of the data processing process of the system according to the second embodiment (first half) [Figure 8] 13 is a detailed flowchart of the data processing process of the system according to the second embodiment (second half) [Explanation of symbols]

[0060] D Database P Information processing device S Software T input / output terminal

Claims

1. To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (which ranges from 0 to 10) indicates the type of pain. The NRS (Numeric Rating Scale) value indicates an overall pain index regardless of the type of pain, Based on the similarity with the NPI value of new patient N (before rehabilitation), A method for calculating a predicted post-rehabilitation NRS value of a new patient N, comprising: For each x type of pain, The NPI value before rehabilitation is expressed as b (before) 1 to bx. The NRS value before rehabilitation was defined as Pb (Pain before), The NRS value after rehabilitation is Pa (Pain after), For the i-th control patient C(i), where the total number of control patients C is n, The x-dimensional vector consisting of the sequence of b1 to bx is defined as CVb(i), The Pb is CPb(i), The Pa is CPa(i), New patient N, The x-dimensional vector consisting of the sequence of b1 to bx is defined as NVb. The NRS value before rehabilitation was defined as NPb. When the cosine similarity between the CVb(i) and the NVb is Sim(i), From the Sim(i) and NPb of all control patients C, NPa(all), which is the NRS value after rehabilitation of new patient N, is calculated. NPa(all)=NPb×{(Σ(i=1 to n)[(Sim(i) / Σ(i=1 to n)Sim(i))×(CPa(i) / CPb(i))]} A method of calculation using the formula:

2. To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (which ranges from 0 to 10) indicates the type of pain. The NRS (Numeric Rating Scale) value indicates an overall pain index regardless of the type of pain, Based on the similarity with the NPI value of new patient N (before rehabilitation), A method for calculating a predicted post-rehabilitation NRS value of a new patient N, comprising: For each x type of pain, The NPI value before rehabilitation is expressed as b (before) 1 to bx. The NRS value before rehabilitation was defined as Pb (Pain before), The NRS value after rehabilitation is Pa (Pain after), For the i-th control patient C(i), where the total number of control patients C is n, The x-dimensional vector consisting of the sequence of b1 to bx is defined as CVb(i), The Pb is CPb(i), The Pa is CPa(i), New patient N, The x-dimensional vector consisting of the sequence of b1 to bx is defined as NVb. The NRS value before rehabilitation was defined as NPb. The cosine similarity between the CVb(i) and the NVb is Sim(i), moreover, Let x be an arbitrary value as a ratio for narrowing down the control patients C whose Sim(i) is high, If the value corresponding to x in the standard normal distribution table is the standard Z value z, A first step of calculating the average value Sim(ave) and standard deviation of the similarity Sim (i=1 to n) between the CVb(i) and the NVb of all control patients C; A second step of calculating a standard similarity SimZ corresponding to z from the Sim(ave) and standard deviation, and extracting control patients c (the number of patients is m) having Sim(i) equal to or greater than SimZ; From the Sim(i) and NPb of the control patient c(i) (i = 1 to m), the NRS value after rehabilitation of the new patient N, NPa(lim), is calculated. NPa(lim)=NPb×{(Σ(i=1 to m)[(Sim(i) / Σ(i=1 to m)Sim(i))×(CPa(i) / CPb(i))]} A third step of calculating the value of the saturation potential by the formula: The method comprising:

3. Among the control patients c having Sim(i) equal to or greater than SimZ extracted in the second step, Sim(i) of the control patient with the smallest Sim(i) is θ(i)=10*[1-cos -1 (Sim(i)) / (π / 2)] The value of θ(i) calculated by the formula was used as the reliability of the NPa(lim). Characterized by: A method for calculating a predicted value of the NRS value after rehabilitation of a new patient N, as claimed in claim 2.

4. To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) indicates an index for each type of pain, and the Numeric Rating Scale (NRS) value indicates an index for overall pain regardless of the type of pain. Based on the similarity with the NPI value of new patient N (before rehabilitation), A system for calculating a predicted NRS value after rehabilitation of a new patient N, The system comprises at least a database, an information processing device, software, and an input / output terminal; The database comprises at least For each of the control patients C, a total of n in number, The NPI value before rehabilitation, Cb(before)1~bx, CPb (Pain before), which is the NRS value before rehabilitation, CPa (Pain after) is the NRS value after rehabilitation. I remember, The software, in the information processing device, from the input / output terminal, For new patient N, The NPI value before rehabilitation is Nb(before)1~Nbx, NPb (Pain before) is the NRS value before rehabilitation Upon receiving the input, For the i-th control subject C(i), The x-dimensional vector consisting of the sequence of Cb1 to Cbx is defined as CVb(i), The CPb is designated as CPb(i), The CPa is designated as CPa(i), New patient N, Let NVb be an x-dimensional vector consisting of the sequence of Nb1 to Nbx, Calculate the cosine similarity Sim(i) between the CVb(i) and the NVb; From Sim(i) of all control patients C and the above NPb, NPa(all), which is the NRS value after rehabilitation of new patient N, is calculated. NPa(all)=NPb×{(Σ(i=1 to n)[(Sim(i) / Σ(i=1 to n)Sim(i))×(CPa(i) / CPb(i))]} and outputting the calculated value from the input / output terminal. A system characterized by.

5. To predict the effectiveness of rehabilitation aimed at improving physical pain caused by stroke Before rehabilitation in the control patient group C (control patient group) who had undergone rehabilitation, The Neuropathic Pain Inventory (NPI) value (NPI is a value between 0 and 10) indicates an index for each type of pain, and the Numeric Rating Scale (NRS) value indicates an index for overall pain regardless of the type of pain. Based on the similarity with the NPI value of new patient N (before rehabilitation), A system for calculating a predicted NRS value after rehabilitation of a new patient N, The system comprises at least a database, an information processing device, software, and an input / output terminal; The database comprises at least For each of the control patients C, a total of n in number, The NPI value before rehabilitation, Cb(before)1~bx, CPb (Pain before), which is the NRS value before rehabilitation, CPa (Pain after) is the NRS value after rehabilitation. I remember, The software, in the information processing device, from the input / output terminal, For new patient N, The NPI value before rehabilitation is Nb(before)1~Nbx, NPb (Pain before) is the NRS value before rehabilitation Upon receiving the input, For the i-th control subject C(i), The x-dimensional vector consisting of the sequence of Cb1 to Cbx is defined as CVb(i), The CPb is designated as CPb(i), The CPa is designated as CPa(i), New patient N, Let NVb be an x-dimensional vector consisting of the sequence of Nb1 to Nbx, Calculate the cosine similarity Sim(i) between the CVb(i) and the NVb; moreover, An arbitrary value x is input as a ratio for narrowing down the control patients C having the top Sim(i) from the input / output terminal, The value corresponding to x in the standard normal distribution table is the standard Z value z. Calculate the average value Sim(ave) and standard deviation of the similarity Sim (i = 1 to n) between the CVb(i) and the NVb of all control patients C; A standard similarity SimZ corresponding to z is calculated from the Sim(ave) and the standard deviation, and control patients c (the number of patients is m) having Sim(i) equal to or greater than SimZ are extracted. From the Sim(i) and the NPb of the control patient c(i) (i = 1 to m), NPa(lim)=NPb×{(Σ(i=1 to m)[(Sim(i) / Σ(i=1 to m)Sim(i))×(CPa(i) / CPb(i))]} The NRS value after rehabilitation for new patient N, NPa(lim), was calculated using the formula: Output is possible from the input / output terminal; A system characterized by.

6. The software, in the information processing device, Among the control patients c having Sim(i) equal to or greater than SimZ, the Sim(i) of the target patient having the smallest Sim(i) is θ(i)=10*[1-cos -1 (Sim(i)) / (π / 2)] The value of θ(i) calculated by the formula is set as the reliability of the NVa(lim), Output is possible from the input / output terminal; Characterized by: A system for calculating a predicted NRS value after rehabilitation of a new patient N according to claim 5.

Images