WO2023157172A1 - Estimation device, estimation method, program, and storage medium - Google Patents

Estimation device, estimation method, program, and storage medium Download PDF

Info

Publication number
WO2023157172A1
WO2023157172A1 PCT/JP2022/006369 JP2022006369W WO2023157172A1 WO 2023157172 A1 WO2023157172 A1 WO 2023157172A1 JP 2022006369 W JP2022006369 W JP 2022006369W WO 2023157172 A1 WO2023157172 A1 WO 2023157172A1
Authority
WO
WIPO (PCT)
Prior art keywords
independence
degree
index
estimation
time
Prior art date
Application number
PCT/JP2022/006369
Other languages
French (fr)
Japanese (ja)
Inventor
隆行 小笠原
東一郎 後藤
健太郎 田中
信吾 塚田
真澄 山口
Original Assignee
日本電信電話株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電信電話株式会社 filed Critical 日本電信電話株式会社
Priority to PCT/JP2022/006369 priority Critical patent/WO2023157172A1/en
Publication of WO2023157172A1 publication Critical patent/WO2023157172A1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons

Definitions

  • the present invention relates to an estimation device, an estimation method, a program and a storage medium.
  • Non-Patent Document 1 In rehabilitation medicine, the degree of independence is used as a scale to evaluate the disability when a person suffers a disability due to paralysis associated with cerebrovascular disease. FIM (Functional Independence Measure) is frequently used in clinical settings not only in Japan but also overseas to measure independence. FIM is scored visually by medical personnel, and a method of automatically estimating FIM using machine learning from measured values measured with a wearable device has been proposed (Non-Patent Document 1).
  • Non-Patent Document 2 describes an example of the application of measurement results from wearable devices to rehabilitation.
  • Non-Patent Document 3 describes an imputation method in statistics.
  • Non-Patent Document 1 a wearable device measures physical activity, heart rate, and sleep quality.
  • sleep disturbances are often associated with hospitalized patients, and the association between sleep quality and independence may be poor, and independence may not be accurately estimated.
  • sleep which is one of the states related to the activity of the cranial nerve system, was attempted using a type of wearable device that can only measure information related to the circulatory system and physical movements. Therefore, the problem can be solved by appropriately based on the information of the circulatory system and physical motion.
  • An object of the present invention is to provide an estimation device, an estimation method, a program, and a storage medium capable of estimating the degree of independence with high accuracy.
  • One aspect of the present invention relates to the state and motion of a subject, and is an index that calculates a time-series index based on vital data regarding the state of the circulatory system of the subject and motion data regarding physical vibrations and angles.
  • a calculating unit, a statistical value calculating unit that calculates an index statistical value that is a statistical value of the time-series index based on the time-series index, and a degree of independence of the subject based on the index statistical value and an independence degree estimating unit that estimates the .
  • One aspect of the present invention relates to the state and motion of a subject, and is an index that calculates a time-series index based on vital data regarding the state of the circulatory system of the subject and motion data regarding physical vibrations and angles.
  • One aspect of the present invention is a program for causing a computer to function as the above estimation device.
  • One aspect of the present invention is a computer-readable recording medium recording a program for causing a computer to function as the above estimation device.
  • the degree of independence can be estimated with high accuracy.
  • FIG. 1 is a diagram showing an estimation system according to a first embodiment
  • FIG. It is a figure which shows the structure of an estimation apparatus. It is a flow chart which shows operation of an estimation system. It is a figure which shows an example of the hardware constitutions of an estimation apparatus. It is a figure which shows the structure of the estimation apparatus based on the modification of 1st Embodiment. It is a figure which shows the structure of the estimation apparatus based on the modification of 1st Embodiment.
  • FIG. 4 is a diagram showing explanatory variables and correlation coefficients;
  • FIG. 1 is a diagram showing an estimation system 1 according to the first embodiment.
  • Estimation system 1 includes estimation device 2 , sensor terminal 202 , and relay terminal 203 .
  • a sensor terminal 202 is attached to the trunk of a person 201 to be measured. The result of measurement by the sensor terminal 202 is relayed by the relay terminal 203 and transmitted to the estimation device 2 .
  • the sensor terminal 202 measures vital data and motion data of the subject 201 .
  • the vital data is a feature quantity relating to the condition of the circulatory system of the subject 201 .
  • the motion data is a feature quantity relating to physical vibrations and angles of the subject 201 .
  • Vital data are, for example, electrocardiographic potential, heart rate, pulse rate, RRI, and body temperature.
  • the motion data are, for example, acceleration and angular velocity.
  • the sensor terminal 202 may be a computer device such as a smart phone or tablet.
  • the relay terminal 203 transmits the data received from the sensor terminal 202 to the estimation device 2 .
  • the relay terminal 203 is connected to the sensor terminal 202 via Bluetooth (registered trademark) and connected to the estimation device 2 via Wi-Fi, but the connection is not limited to this.
  • the relay terminal 203 is a computer device such as a smart phone or a tablet, and may process data received from the sensor terminal 202 and transmit the processed data to the estimation device 2 .
  • FIG. 2 is a diagram showing the configuration of the estimation device 2. As shown in FIG. The estimation device 2 includes a reception unit 10 , a received data storage unit 11 , an index calculation unit 12 , a statistical value calculation unit 14 , an independence estimation model storage unit 16 , an independence estimation unit 18 and a presentation unit 20 .
  • the receiving unit 10 receives from the relay terminal 203 the vital data and motion data, which are the results of measurement by the sensor terminal 202 .
  • the received data storage unit 11 stores the data received by the receiving unit 10.
  • the received data storage unit 11 stores data in association with the time when the data was measured.
  • the received data storage unit 11 stores, for example, time-series data of the subject's 201 electrocardiographic potential and acceleration.
  • the index calculation unit 12 calculates a time-series index indicating changes over time in the subject's 201 state and movement.
  • the time-series index is, for example, the time change of %HRR (percent heart rate reserve), which indicates a value obtained by normalizing the heart rate calculated based on the electrocardiographic potential by the maximum and minimum heart rates of the subject 201 .
  • the time-series index is, for example, activity time for each posture (lying, sitting, standing, walking) calculated based on acceleration.
  • the time-series index is, for example, the change in body motion over time, which is the standard deviation per predetermined time (for example, one second) of a value obtained by synthesizing three-axis acceleration calculated based on the acceleration.
  • the time-series index is, for example, the temporal change in the number of steps per predetermined time (for example, one minute) calculated based on acceleration.
  • the time-series index is, for example, the change over time of the activity cost index, which is a value obtained by dividing %HRR calculated based on electrocardiographic potential and acceleration by body motion.
  • the index calculator 12 may calculate at least one of the time change of %HRR, the activity time by posture, the time change of body movement, the time change of the number of steps per hour, and the time change of the activity cost index.
  • the index calculation unit 12 does not need to calculate the time-series index indicating sleep as the time-series index indicating the time-dependent changes in the state and motion of the person 201 to be measured.
  • a time-series indicator of sleep is, for example, sleep duration or sleep quality.
  • the index calculation unit 12 may normalize the time-series data stored in the received data storage unit 11. For example, when the time-series data stored in the received data storage unit 11 is time-series data over 24 hours or more, the index calculation unit 12 may normalize the time-series data into time-series data over 24 hours. The index calculation unit 12 may generate time-series data over 24 hours by taking an ensemble average at the same time in the time-series data.
  • the statistical value calculation unit 14 calculates statistical values of the indices based on the time-series indices calculated by the index calculation unit 12 .
  • Statistics are, for example, mean values, quantiles, and deviations.
  • the statistic value of the index may be a statistic value of the time-series index value for each predetermined time (for example, 30 minutes).
  • the time-series index is the time change of %HRR for 24 hours, for example, 48 average values of %HRR are calculated every 30 minutes, and finally the average value of the 48 average values is calculated as the statistical value of the index. be done.
  • the time-series index is the activity time for each posture, the average value of the time in each posture for every 30 minutes, for example, is calculated as the statistic value of the index.
  • the time-series index is a change in body movement over a period of 24 hours, for example, 48 average values of body movement are calculated every 30 minutes, and finally the average value of the 48 average values is used as the statistical value of the index. Calculated.
  • the time-series index is the time change of the number of steps
  • the average value of the number of steps for every 30 minutes for example, is calculated as the statistic value of the index.
  • the time-series index is the time change of the activity cost index for 24 hours, for example, 48 average values of the activity cost index are calculated every 30 minutes
  • the average value of the 48 average values is the statistics of the index. calculated as a value.
  • the independence estimation model storage unit 16 stores an independence estimation model.
  • the degree-of-independence estimation model is a model that outputs the degree of independence with the statistical values of the indices calculated by the statistical value calculation unit 14 as input.
  • the independence degree estimation model is created by machine learning using a data set of index statistical values and independence degrees.
  • the explanatory variable in machine learning is the statistical value of the index, and the objective variable is the degree of independence.
  • Independence is a scale that assesses the disability of people with disabilities, especially older people with disabilities.
  • the degree of independence is an index that indicates how much a person with a disability can independently carry out daily life, and it is also an index that indicates whether a person with a disability is bedridden or not.
  • the degree of independence is, for example, FIM.
  • the degree of independence is, for example, the sum of scores of 13 items related to motor function included in FIM.
  • Machine learning techniques are not limited, for example neural networks, random forests, support vector machines, logistic regression or ensemble learning.
  • the statistic value of the index may be the value of the time-series index for each predetermined time.
  • the index statistic may be 48 average %HRR values for each 30-minute period.
  • the statistic of the metric is activity time by posture, the statistic of the metric may be the time in each posture every 30 minutes.
  • the statistic value of the index may be 48 average values of body movement every 30 minutes.
  • the time-series metric is the time change in the number of steps
  • the statistical value of the metric may be the number of steps per 30 minutes.
  • the indicator statistic may be 48 average values of the activity cost index for every 30 minutes.
  • the statistical value of the index may be a time-series index. At this time, since the number of inputs to the independence degree estimation model increases, the calculation load at the time of estimation increases, but the estimation accuracy improves.
  • the independence degree estimating unit 18 estimates the independence degree from the statistical values of the indicators using the independence degree estimation model.
  • the independence degree estimating unit 18 estimates the independence degree by inputting the statistic value of the index into the independence degree estimation model and outputting the independence degree.
  • the presentation unit 20 presents the degree of independence estimated by the degree-of-independence estimation unit 18 .
  • the presentation unit 20 presents the degree of independence by outputting data to a display device such as a display.
  • FIG. 3 is a flowchart showing the operation of the estimation system 1.
  • the sensor terminal 202 measures vital data and motion data (step S101).
  • the relay terminal 203 relays the vital data and motion data measured by the sensor terminal 202 and transmits them to the estimation device 2 (step S102).
  • the receiving unit 10 of the estimation device 2 receives vital data and motion data from the relay terminal 203 (step S201).
  • the index calculator 12 calculates a time-series index based on the vital data and the motion data (step S202).
  • the statistical value calculation unit 14 calculates statistical values of indices based on the time-series indices (step S203).
  • the degree-of-independence estimation unit 18 estimates the degree of independence by inputting the statistical values of the indices into the degree-of-independence estimation model (step S204).
  • the presentation unit 20 presents the estimated degree of independence (step S205).
  • FIG. 4 is a diagram showing an example of the hardware configuration of the estimation device 2.
  • the estimating device 2 includes, for example, a computing device 102 having a CPU 103 and a main memory device 104 connected via a bus 101, a computer having a communication interface 105, an external storage device 107, a clock 108, and a display device 109; It can be implemented by a program that controls hardware resources.
  • the CPU 103 and the main storage device 104 constitute an arithmetic device 102 .
  • Programs for the CPU 103 to perform various controls and calculations are stored in advance in the main memory device 104 .
  • Each function of the estimation device 2 shown in FIG. 2 is realized by the arithmetic device 102 .
  • the communication interface 105 is an interface and control device for connecting the estimation device 2 and various external electronic devices such as the relay terminal 203 via a communication network.
  • the estimating device 2 may receive heart rate, electrocardiographic waveform, and acceleration data from the relay terminal 203 via the communication interface 105 and the communication network.
  • the communication interface 105 for example, a calculation interface and an antenna compatible with wireless data communication standards such as LTE, 3G, wireless LAN, and Bluetooth are used.
  • the communication interface 105 implements the receiver 10 in FIG.
  • the external storage device 107 is composed of a readable and writable storage medium and a drive device for reading and writing various information such as programs and data on the storage medium.
  • a semiconductor memory such as a hard disk or a flash memory can be used as a storage medium for the external storage device 107 .
  • the external storage device 107 includes a storage area for storing vital data and motion data measured by the sensor terminal 202, a program storage unit for storing a program for the estimation device 2 to analyze the vital data and motion data, Other storage devices (not shown) may include, for example, a storage device for backing up programs and data stored in the external storage device 107 .
  • the received data storage unit 11 and the independence degree estimation model storage unit 16 in FIG. 2 are implemented by the external storage device 107 .
  • the clock 108 is configured with a built-in clock provided in the estimation device 2, and measures time.
  • the time information obtained by the clock 108 is used for sampling of vital data and motion data and data analysis processing.
  • the display device 109 functions as the presentation unit 20 of the estimation device 2.
  • the display device 109 is implemented by a liquid crystal display or the like.
  • the independence degree estimating model may be a model that receives the statistical values of the indices calculated by the statistical value calculation unit 14 and the profile of the subject 201 as inputs and outputs the independence degree. That is, the independence degree estimation model is generated using the statistical values of the indices and the profile of the subject 201 as explanatory variables and the independence degree as the objective variable.
  • the profile of the person to be measured 201 is the age, height, weight or sex of the person to be measured 201, for example.
  • the profile of the person to be measured 201 may be data obtained by combining some of the age, height, weight and gender of the person to be measured 201 .
  • FIG. 5 is a diagram showing the configuration of the estimation device 2 according to the modification of the first embodiment.
  • the estimation device 2 includes a subject profile acquisition unit 22 .
  • the subject profile acquisition unit 22 acquires the profile of the subject 201 .
  • the degree of independence estimation unit 18 inputs the statistical values of the indices and the profile of the person to be measured 201 acquired by the person to be measured profile acquisition unit 22 into the degree of independence estimation model and outputs the degree of independence, thereby estimating the degree of independence.
  • the independence degree estimation model may be a model that takes as inputs the statistical values of the indices calculated by the statistical value calculation unit 14 and the past independence degree of the subject 201 and outputs the independence degree. That is, the independence degree estimation model is generated by using the statistical value of the index and the past independence degree of the subject 201 as explanatory variables and the independence degree as the objective variable.
  • FIG. 6 is a diagram showing the configuration of the estimation device 2 according to the modification of the first embodiment.
  • the estimation device 2 includes a past independence degree acquisition unit 24 .
  • the past independence degree acquisition unit 24 acquires the past independence degree of the subject 201 .
  • the independence degree estimation unit 18 inputs the past independence degree of the subject 201 acquired by the past independence degree acquisition unit 24 and the statistical value of the index into the independence degree estimation model and outputs the independence degree, thereby estimating the independence degree. do.
  • the independence degree estimation model may be generated using the statistical value of the index, the profile of the subject 201, and the past independence degree of the subject 201 as explanatory variables, and the independence degree as the objective variable.
  • the independence degree estimating unit 18 inputs the statistical value of the index, the profile of the person to be measured 201, and the past independence degree of the person to be measured 201 into the independence degree estimation model, and outputs the degree of independence, thereby estimating the degree of independence. do.
  • FIG. 7 is a diagram showing explanatory variables and correlation coefficients.
  • the time-series indices of the subject 201 are statistics in 30-minute increments as explanatory variables. Correlation coefficients were calculated by performing 5-fold cross-validation on the values.
  • the explanatory variables of the second condition include the profile of the person to be measured 201 (age, height, weight and sex) in addition to the explanatory variables of the first condition.
  • the explanatory variables of the third condition are the profile of the subject 201 (age, height, weight and sex) and the degree of independence of the subject 201 in the past.
  • the past independence degree of the subject 201 is the independence degree in the first week of hospitalization
  • the independence degree of the objective variable is the independence degree in the fifth week of hospitalization. That is, the past independence degree is the independence degree four weeks before the estimated independence degree.
  • the explanatory variables of the fourth condition include the past degree of independence of the subject 201 in addition to the explanatory variables of the first condition.
  • the explanatory variables of the fifth condition include the profile of the subject 201 (age, height, weight and sex) in addition to the explanatory variables of the fourth condition.
  • the correlation coefficient was calculated by performing 5-fold cross-validation on the statistical values for the second to fifth conditions as well. In k-fold cross-validation, k is the number of data sets for evaluation, and k values of 2, 5, or 10 can be used.
  • the correlation coefficient exceeded 0.6 under all conditions.
  • the estimation device 2 can estimate the degree of independence with high accuracy.
  • the estimation device 2 may The accuracy of estimating the degree becomes worse. Therefore, when the time-series index indicating time-dependent changes in the state and motion of the subject 201 includes a time-series index indicating sleep, the estimation device 2 estimates the degree of independence without being affected by the degree of independence. can be used, leading to improved estimation accuracy.
  • the estimating device 2 according to the second embodiment includes a loss complementing unit 30 in addition to the estimating device 2 according to the first embodiment.
  • the loss complementing unit 30 complements missing data. Missing data is data that could not be received by the estimating device 2 due to a malfunction of the sensor terminal 202, a communication situation between the relay terminal 203 and the estimating device 2, or the like. Missing data may include data whose value exceeds the upper threshold or falls below the lower threshold, based on a predetermined upper threshold and a predetermined lower threshold.
  • the loss complementing unit 30 may invalidate the missing data.
  • the index calculation unit 12 generates time-series data over 24 hours by taking the ensemble average of the same time in the time-series data
  • the loss complementing unit 30 takes the average value of the same time excluding missing data as the ensemble average can be calculated.
  • the missing data complementing unit 30 may complement missing data by applying a multiple imputation method to the profile of the subject 201 and the past degree of independence of the subject 201 .
  • the estimation device 2 according to the second embodiment can compensate for data deficiencies, handle more data sets of explanatory variables without deficiencies, and improve the reliability of the estimation of the degree of independence. can.
  • explanatory variables of the degree of independence estimation model are not limited to the statistical values of indices, the profile of the subject 201, or the degree of independence of the subject 201 in the past.
  • explanatory variables of the degree-of-independence estimation model may include disease information of subject 201 .
  • the disease information of the person to be measured 201 is, for example, information as to whether or not the person to be measured 201 has a disease such as a cerebrovascular disease, spinal cord injury, or femoral fracture.
  • the degree-of-independence estimation unit 18 may include the disease information of the person to be measured 201 in the inputs to the degree-of-independence estimation model in accordance with the explanatory variables.
  • the estimation device 2 may include an estimated independence degree storage unit 40 and store the independence degree estimated by the independence degree estimation unit 18 .
  • the estimated independence degree storage unit 40 may store the statistical value of the index and the independence degree in association with the profile of the subject 201, the past independence degree of the subject 201, or the disease information of the subject 201. good.
  • the estimation device 2 may include an estimated statistic value calculation unit 42 that calculates the statistic value of the estimated independence degree based on the estimated independence degree stored in the estimated independence degree storage unit 40 .
  • the estimated statistic value calculation unit 42 may calculate the statistic value of the estimated degree of independence based on the profile of the subject 201 .
  • the estimated statistic value calculation unit 42 calculates an average value and a standard deviation, which are statistic values of the estimated degree of independence, for each age group such as 50's and 60's.
  • the presentation unit 20 may present the results calculated by the estimated statistical value calculation unit 42 .
  • the independence degree estimation model By using different explanatory variables and objective variables for the independence degree estimation model, it is possible to create a model that estimates characteristics other than the independence degree. For example, one item in the profile of the subject 201 as the objective variable, the statistical value of the index as the explanatory variable, the items in the profile of the subject 201 that are not the objective variable, the degree of independence of the subject 201, and the subject 201 , and an estimation model for estimating one item of the profile of the person to be measured 201 may be created. By inputting an explanatory variable into the estimation model, an estimated value of one item of the profile of the subject 201 is output. For example, when an explanatory variable is input to the estimation model, an estimated value of the age of the person to be measured 201 is output. The person to be measured 201 can grasp that the is good.
  • the disease of the person to be measured 201 is set as the objective variable, and at least one of the statistical value of the index, the profile item of the person to be measured 201, and the degree of independence of the person to be measured 201 is set as the explanatory variable, and the disease is estimated.
  • the presentation unit 20 may present data compressed by the independence estimation model.
  • the dimensions of the explanatory variables are compressed, leaving only essential information necessary for learning, so essential information can be grasped.
  • the estimation device 2 and the sensor terminal 202 may be realized by the same device. At this time, the estimation device 2 and the sensor terminal 202 do not need to communicate via the relay terminal 203 . Also, the relay terminal 203 may perform a part of the functions of the estimation device 2 . For example, the relay terminal 203 may perform part of the functions of the index calculation unit 12 , the statistical value calculation unit 14 , the independence estimation unit 18 , and the presentation unit 20 and transmit the processed data to the estimation device 2 .

Abstract

An estimation device provided with an indicator calculation unit for calculating a time series indicator relating to the condition and movements of a subject on the basis of vital data relating to the condition of the circulatory organs of the subject and movement data relating to physical oscillations and angles; a statistic calculation unit for calculating an indicator statistic, which is a statistic for the time series indicator, on the basis of the time series indicator; and an independence estimation unit for estimating the independence of the subject on the basis of the indicator statistic.

Description

推定装置、推定方法、プログラム及び記憶媒体Estimation device, estimation method, program and storage medium
　本発明は、推定装置、推定方法、プログラム及び記憶媒体に関する。 The present invention relates to an estimation device, an estimation method, a program and a storage medium.
　リハビリテーション医療において、脳血管疾患などに伴う麻痺により障害を負った際に、その障害を評価する尺度として自立度が用いられている。自立度にはFIM（Functional Independence Measure）が日本のみならず海外でも臨床現場にて頻繁に用いられている。FIMは医療者が目視で採点するが、ウェアラブルデバイスで計測した計測値から機械学習を用いて自動的に推定する方法が提案されている（非特許文献１）。 In rehabilitation medicine, the degree of independence is used as a scale to evaluate the disability when a person suffers a disability due to paralysis associated with cerebrovascular disease. FIM (Functional Independence Measure) is frequently used in clinical settings not only in Japan but also overseas to measure independence. FIM is scored visually by medical personnel, and a method of automatically estimating FIM using machine learning from measured values measured with a wearable device has been proposed (Non-Patent Document 1).
　非特許文献２には、ウェアラブルデバイスによる計測結果のリハビリテーションへの応用例が記載されている。非特許文献３には、統計学における代入法が記載されている。 　Non-Patent Document 2 describes an example of the application of measurement results from wearable devices to rehabilitation. Non-Patent Document 3 describes an imputation method in statistics.
特開２０２０－０３６７８１号公報Japanese Patent Application Laid-Open No. 2020-036781
　非特許文献１において、ウェアラブルデバイスは身体活動量、心拍数、睡眠の質を計測する。しかしながら、入院患者には睡眠障害がしばしば伴い、睡眠の質と自立度の関連性が低く、自立度を正確に推定できないことがある。これは、脳神経系の活動に関する状態のひとつである睡眠を、循環器系や物理動作に関する情報しか測定できない種類のウェアラブルデバイスを用いて得ようとしたことに起因するためである。よって循環器系や物理動作の情報に適切に基づくことで、課題を解決することができる。
　本発明は、高い精度で自立度を推定することができる推定装置、推定方法、プログラム及び記憶媒体を提供することを目的としている。 In Non-Patent Document 1, a wearable device measures physical activity, heart rate, and sleep quality. However, sleep disturbances are often associated with hospitalized patients, and the association between sleep quality and independence may be poor, and independence may not be accurately estimated. This is because sleep, which is one of the states related to the activity of the cranial nerve system, was attempted using a type of wearable device that can only measure information related to the circulatory system and physical movements. Therefore, the problem can be solved by appropriately based on the information of the circulatory system and physical motion.
An object of the present invention is to provide an estimation device, an estimation method, a program, and a storage medium capable of estimating the degree of independence with high accuracy.
　本発明の一態様は、被測定者の状態や動作に関し、前記被測定者の循環器の状態に関するバイタルデータ及び物理的な振動や角度に関する動作データに基づいて、時系列の指標を算出する指標算出部と、前記時系列の指標に基づいて、前記時系列の指標の統計値である指標統計値を算出する統計値算出部と、前記指標統計値に基づいて、前記被測定者の自立度を推定する自立度推定部と、を備える推定装置である。 One aspect of the present invention relates to the state and motion of a subject, and is an index that calculates a time-series index based on vital data regarding the state of the circulatory system of the subject and motion data regarding physical vibrations and angles. a calculating unit, a statistical value calculating unit that calculates an index statistical value that is a statistical value of the time-series index based on the time-series index, and a degree of independence of the subject based on the index statistical value and an independence degree estimating unit that estimates the .
　本発明の一態様は、被測定者の状態や動作に関し、前記被測定者の循環器の状態に関するバイタルデータ及び物理的な振動や角度に関する動作データに基づいて、時系列の指標を算出する指標算出ステップと、前記時系列の指標に基づいて、前記時系列の指標の統計値である指標統計値を算出する統計値算出ステップと、前記指標統計値に基づいて、前記被測定者の自立度を推定する自立度推定ステップと、を有する推定方法である。 One aspect of the present invention relates to the state and motion of a subject, and is an index that calculates a time-series index based on vital data regarding the state of the circulatory system of the subject and motion data regarding physical vibrations and angles. a calculating step, a statistical value calculating step of calculating an index statistical value, which is a statistical value of the time-series index, based on the time-series index; and an independence degree estimation step of estimating .
　本発明の一態様は、上記の推定装置としてコンピュータを機能させるためのプログラムである。 One aspect of the present invention is a program for causing a computer to function as the above estimation device.
　本発明の一態様は、上記の推定装置としてコンピュータを機能させるためのプログラムを記録したコンピュータで読み取り可能な記録媒体である。 One aspect of the present invention is a computer-readable recording medium recording a program for causing a computer to function as the above estimation device.
　本発明により、高い精度で自立度を推定することができる。 With the present invention, the degree of independence can be estimated with high accuracy.
第１の実施形態に係る推定システムを示す図である。1 is a diagram showing an estimation system according to a first embodiment; FIG. 推定装置の構成を示す図である。It is a figure which shows the structure of an estimation apparatus. 推定システムの動作を示すフローチャートである。It is a flow chart which shows operation of an estimation system. 推定装置のハードウェア構成の一例を示す図である。It is a figure which shows an example of the hardware constitutions of an estimation apparatus. 第１の実施形態の変形例に係る推定装置の構成を示す図である。It is a figure which shows the structure of the estimation apparatus based on the modification of 1st Embodiment. 第１の実施形態の変形例に係る推定装置の構成を示す図である。It is a figure which shows the structure of the estimation apparatus based on the modification of 1st Embodiment. 説明変数と相関係数を示した図である。FIG. 4 is a diagram showing explanatory variables and correlation coefficients;
〈第１の実施形態〉
　図１は、第１の実施形態に係る推定システム１を示す図である。推定システム１は、推定装置２、センサ端末２０２、中継端末２０３を含む。推定システム１において、例えば被測定者２０１の体幹にセンサ端末２０２が装着される。センサ端末２０２が計測した結果は中継端末２０３により中継され推定装置２に送信される。センサ端末２０２は、被測定者２０１のバイタルデータ及び動作データを測定する。バイタルデータは、被測定者２０１の循環器の状態に関する特徴量である。動作データは、被測定者２０１の物理的な振動や角度に関する特徴量である。バイタルデータは例えば心電位、心拍数、脈拍数、ＲＲＩ、体温である。動作データは例えば加速度、角速度である。センサ端末２０２は、スマートフォンやタブレットなどのコンピュータ機器としてもよい。 <First Embodiment>
FIG. 1 is a diagram showing an estimation system 1 according to the first embodiment. Estimation system 1 includes estimation device 2 , sensor terminal 202 , and relay terminal 203 . In the estimation system 1, for example, a sensor terminal 202 is attached to the trunk of a person 201 to be measured. The result of measurement by the sensor terminal 202 is relayed by the relay terminal 203 and transmitted to the estimation device 2 . The sensor terminal 202 measures vital data and motion data of the subject 201 . The vital data is a feature quantity relating to the condition of the circulatory system of the subject 201 . The motion data is a feature quantity relating to physical vibrations and angles of the subject 201 . Vital data are, for example, electrocardiographic potential, heart rate, pulse rate, RRI, and body temperature. The motion data are, for example, acceleration and angular velocity. The sensor terminal 202 may be a computer device such as a smart phone or tablet.
　中継端末２０３は、センサ端末２０２から受信したデータを推定装置２に送信する。中継端末２０３は、例えばセンサ端末２０２とＢｌｕｅｔｏｏｔｈ（登録商標）により接続され、推定装置２とＷｉ－Ｆｉにより接続されるがこれに限られない。中継端末２０３は、スマートフォンやタブレットなどのコンピュータ機器であって、センサ端末２０２から受信したデータを処理し推定装置２に送信してもよい。 The relay terminal 203 transmits the data received from the sensor terminal 202 to the estimation device 2 . For example, the relay terminal 203 is connected to the sensor terminal 202 via Bluetooth (registered trademark) and connected to the estimation device 2 via Wi-Fi, but the connection is not limited to this. The relay terminal 203 is a computer device such as a smart phone or a tablet, and may process data received from the sensor terminal 202 and transmit the processed data to the estimation device 2 .
　図２は、推定装置２の構成を示す図である。推定装置２は、受信部１０、受信データ記憶部１１、指標算出部１２、統計値算出部１４、自立度推定モデル記憶部１６、自立度推定部１８、提示部２０を備える。 FIG. 2 is a diagram showing the configuration of the estimation device 2. As shown in FIG. The estimation device 2 includes a reception unit 10 , a received data storage unit 11 , an index calculation unit 12 , a statistical value calculation unit 14 , an independence estimation model storage unit 16 , an independence estimation unit 18 and a presentation unit 20 .
　受信部１０は、センサ端末２０２による計測結果であるバイタルデータ及び動作データを中継端末２０３から受信する。 The receiving unit 10 receives from the relay terminal 203 the vital data and motion data, which are the results of measurement by the sensor terminal 202 .
　受信データ記憶部１１は、受信部１０が受信したデータを記憶する。受信データ記憶部１１には、データと当該データが測定された時間が紐づけて記憶される。受信データ記憶部１１は、例えば被測定者２０１の心電位及び加速度の時系列データを記憶する。 The received data storage unit 11 stores the data received by the receiving unit 10. The received data storage unit 11 stores data in association with the time when the data was measured. The received data storage unit 11 stores, for example, time-series data of the subject's 201 electrocardiographic potential and acceleration.
　指標算出部１２は、受信データ記憶部１１に記憶されたデータに基づいて、被測定者２０１の状態や動作の時間変化を示す時系列の指標を算出する。時系列の指標は、例えば心電位に基づき算出される心拍数を被測定者２０１の最大・最小心拍数で規格化した値を示す％ＨＲＲ(percent heart rate reserve)の時間変化である。時系列の指標は、例えば加速度に基づき算出される体勢別（臥位、座位、立位、歩行）の活動時間である。時系列の指標は、例えば加速度に基づき算出される３軸加速度を合成した値の所定の時間（例えば１秒間）当たりの標準偏差である体動の時間変化である。時系列の指標は、例えば加速度に基づき算出される所定の時間（例えば１分間）当たりの歩数の時間変化である。時系列の指標は、例えば心電位及び加速度に基づき算出される％ＨＲＲを体動で割った値である活動コスト指数の時間変化である。指標算出部１２は、％ＨＲＲの時間変化、体勢別の活動時間、体動の時間変化、時間当たりの歩数の時間変化、活動コスト指数の時間変化のうち少なくとも１つを算出すればよい。 Based on the data stored in the received data storage unit 11, the index calculation unit 12 calculates a time-series index indicating changes over time in the subject's 201 state and movement. The time-series index is, for example, the time change of %HRR (percent heart rate reserve), which indicates a value obtained by normalizing the heart rate calculated based on the electrocardiographic potential by the maximum and minimum heart rates of the subject 201 . The time-series index is, for example, activity time for each posture (lying, sitting, standing, walking) calculated based on acceleration. The time-series index is, for example, the change in body motion over time, which is the standard deviation per predetermined time (for example, one second) of a value obtained by synthesizing three-axis acceleration calculated based on the acceleration. The time-series index is, for example, the temporal change in the number of steps per predetermined time (for example, one minute) calculated based on acceleration. The time-series index is, for example, the change over time of the activity cost index, which is a value obtained by dividing %HRR calculated based on electrocardiographic potential and acceleration by body motion. The index calculator 12 may calculate at least one of the time change of %HRR, the activity time by posture, the time change of body movement, the time change of the number of steps per hour, and the time change of the activity cost index.
　指標算出部１２は、被測定者２０１の状態や動作の時間変化を示す時系列の指標として、睡眠を示す時系列の指標を算出しなくてよい。睡眠を示す時系列の指標は例えば睡眠時間又は睡眠品質である。 The index calculation unit 12 does not need to calculate the time-series index indicating sleep as the time-series index indicating the time-dependent changes in the state and motion of the person 201 to be measured. A time-series indicator of sleep is, for example, sleep duration or sleep quality.
　指標算出部１２は、受信データ記憶部１１に記憶された時系列データを規格化してもよい。指標算出部１２は、例えば受信データ記憶部１１に記憶された時系列データが２４時間以上にわたる時系列データであるとき、規格化を行い２４時間にわたる時系列データとしてもよい。指標算出部１２は、時系列データにおいて同時刻のアンサンブル平均をとることで２４時間にわたる時系列データを生成してもよい。 The index calculation unit 12 may normalize the time-series data stored in the received data storage unit 11. For example, when the time-series data stored in the received data storage unit 11 is time-series data over 24 hours or more, the index calculation unit 12 may normalize the time-series data into time-series data over 24 hours. The index calculation unit 12 may generate time-series data over 24 hours by taking an ensemble average at the same time in the time-series data.
　統計値算出部１４は、指標算出部１２により算出された時系列の指標に基づいて指標の統計値を算出する。統計値は例えば平均値、分位数、偏差である。指標の統計値は、時系列の指標の所定の時間（例えば３０分）ごとの値の統計値であってよい。時系列の指標が２４時間の％ＨＲＲの時間変化のとき、例えば３０分間ごとの％ＨＲＲの平均値４８個が算出され、最終的に４８個の平均値の平均値が指標の統計値として算出される。時系列の指標が体勢別の活動時間であるとき、例えば３０分間ごとの各態勢の時間の平均値が指標の統計値として算出される。時系列の指標が２４時間の体動の時間変化であるとき、例えば３０分間ごとの体動の平均値４８個が算出され、最終的に４８個の平均値の平均値が指標の統計値として算出される。時系列の指標が歩数の時間変化であるとき、指標の統計値として例えば３０分間ごとの歩数の平均値が算出される。時系列の指標が２４時間の活動コスト指数の時間変化であるとき、例えば３０分間ごとの活動コスト指数の平均値４８個が算出され、最終的に４８個の平均値の平均値が指標の統計値として算出される。 The statistical value calculation unit 14 calculates statistical values of the indices based on the time-series indices calculated by the index calculation unit 12 . Statistics are, for example, mean values, quantiles, and deviations. The statistic value of the index may be a statistic value of the time-series index value for each predetermined time (for example, 30 minutes). When the time-series index is the time change of %HRR for 24 hours, for example, 48 average values of %HRR are calculated every 30 minutes, and finally the average value of the 48 average values is calculated as the statistical value of the index. be done. When the time-series index is the activity time for each posture, the average value of the time in each posture for every 30 minutes, for example, is calculated as the statistic value of the index. When the time-series index is a change in body movement over a period of 24 hours, for example, 48 average values of body movement are calculated every 30 minutes, and finally the average value of the 48 average values is used as the statistical value of the index. Calculated. When the time-series index is the time change of the number of steps, the average value of the number of steps for every 30 minutes, for example, is calculated as the statistic value of the index. When the time-series index is the time change of the activity cost index for 24 hours, for example, 48 average values of the activity cost index are calculated every 30 minutes, and finally the average value of the 48 average values is the statistics of the index. calculated as a value.
　自立度推定モデル記憶部１６は、自立度推定モデルを記憶する。自立度推定モデルは、統計値算出部１４により算出される指標の統計値を入力として、自立度を出力するモデルである。自立度推定モデルは、指標の統計値と自立度とのデータセットを用いて機械学習により作成される。機械学習における説明変数は指標の統計値であり、目的変数は自立度である。自立度は、障害を有する者、特に障害を有する高齢者の障害を評価する尺度である。自立度は、障害を有する者が日常生活をどのくらい自立してできるかを示す指標であり、障害を有する者が寝たきりであるか否かを示す指標でもある。自立度は例えばＦＩＭである。自立度は例えばＦＩＭに含まれる運動機能に関する１３項目の点数の和である。機械学習の手法は限定されず、例えばニューラルネットワーク、ランダムフォレスト、サポートベクトルマシン、ロジスティック回帰又はアンサンブル学習である。 The independence estimation model storage unit 16 stores an independence estimation model. The degree-of-independence estimation model is a model that outputs the degree of independence with the statistical values of the indices calculated by the statistical value calculation unit 14 as input. The independence degree estimation model is created by machine learning using a data set of index statistical values and independence degrees. The explanatory variable in machine learning is the statistical value of the index, and the objective variable is the degree of independence. Independence is a scale that assesses the disability of people with disabilities, especially older people with disabilities. The degree of independence is an index that indicates how much a person with a disability can independently carry out daily life, and it is also an index that indicates whether a person with a disability is bedridden or not. The degree of independence is, for example, FIM. The degree of independence is, for example, the sum of scores of 13 items related to motor function included in FIM. Machine learning techniques are not limited, for example neural networks, random forests, support vector machines, logistic regression or ensemble learning.
　指標の統計値は、時系列の指標の所定の時間ごとの値であってもよい。時系列の指標が２４時間の％ＨＲＲの時間変化のとき、指標の統計値は３０分間ごとの％ＨＲＲの平均値４８個であってもよい。時系列の指標が体勢別の活動時間であるとき、指標の統計値は、３０分間ごとの各態勢の時間であってもよい。時系列の指標が２４時間の体動の時間変化であるとき、指標の統計値は、３０分間ごとの体動の平均値４８個であってもよい。時系列の指標が歩数の時間変化であるとき、指標の統計値は、３０分間ごとの歩数であってよい。時系列の指標が２４時間の活動コスト指数の時間変化であるとき、指標の統計値は、３０分間ごとの活動コスト指数の平均値４８個であってよい。このとき、自立度推定モデルの入力の数が増大するため、推定時の計算負荷は増大するが、推定精度は向上する。 The statistic value of the index may be the value of the time-series index for each predetermined time. When the time-series index is a 24-hour %HRR change over time, the index statistic may be 48 average %HRR values for each 30-minute period. When the time-series metric is activity time by posture, the statistic of the metric may be the time in each posture every 30 minutes. When the time-series index is the time change of body movement for 24 hours, the statistic value of the index may be 48 average values of body movement every 30 minutes. When the time-series metric is the time change in the number of steps, the statistical value of the metric may be the number of steps per 30 minutes. When the time-series indicator is the time change of the activity cost index for 24 hours, the indicator statistic may be 48 average values of the activity cost index for every 30 minutes. At this time, since the number of inputs to the independence degree estimation model increases, the calculation load at the time of estimation increases, but the estimation accuracy improves.
　指標の統計値は、時系列の指標であってもよい。このとき、自立度推定モデルの入力の数が増大するため、推定時の計算負荷は増大するが、推定精度は向上する。 The statistical value of the index may be a time-series index. At this time, since the number of inputs to the independence degree estimation model increases, the calculation load at the time of estimation increases, but the estimation accuracy improves.
　自立度推定部１８は、自立度推定モデルを用いて指標の統計値から自立度を推定する。自立度推定部１８は、指標の統計値を自立度推定モデルに入力し自立度を出力させることで自立度を推定する。 The independence degree estimating unit 18 estimates the independence degree from the statistical values of the indicators using the independence degree estimation model. The independence degree estimating unit 18 estimates the independence degree by inputting the statistic value of the index into the independence degree estimation model and outputting the independence degree.
　提示部２０は、自立度推定部１８により推定された自立度を提示する。提示部２０は例えばディスプレイなど表示装置にデータを出力することで自立度を提示する。 The presentation unit 20 presents the degree of independence estimated by the degree-of-independence estimation unit 18 . The presentation unit 20 presents the degree of independence by outputting data to a display device such as a display.
　図３は、推定システム１の動作を示すフローチャートである。センサ端末２０２がバイタルデータ及び動作データを測定する（ステップＳ１０１）。中継端末２０３はセンサ端末２０２が測定したバイタルデータ及び動作データを中継し、推定装置２に送信する（ステップＳ１０２）。推定装置２の受信部１０は中継端末２０３からバイタルデータ及び動作データを受信する（ステップＳ２０１）。指標算出部１２はバイタルデータ及び動作データに基づいて時系列の指標を算出する（ステップＳ２０２）。統計値算出部１４は、時系列の指標に基づいて指標の統計値を算出する（ステップＳ２０３）。自立度推定部１８は、指標の統計値を自立度推定モデルに入力することで自立度を推定する（ステップＳ２０４）。提示部２０は、推定された自立度を提示する（ステップＳ２０５）。 FIG. 3 is a flowchart showing the operation of the estimation system 1. FIG. The sensor terminal 202 measures vital data and motion data (step S101). The relay terminal 203 relays the vital data and motion data measured by the sensor terminal 202 and transmits them to the estimation device 2 (step S102). The receiving unit 10 of the estimation device 2 receives vital data and motion data from the relay terminal 203 (step S201). The index calculator 12 calculates a time-series index based on the vital data and the motion data (step S202). The statistical value calculation unit 14 calculates statistical values of indices based on the time-series indices (step S203). The degree-of-independence estimation unit 18 estimates the degree of independence by inputting the statistical values of the indices into the degree-of-independence estimation model (step S204). The presentation unit 20 presents the estimated degree of independence (step S205).
　図４は、推定装置２のハードウェア構成の一例を示す図である。推定装置２は、例えば、バス１０１を介して接続されるＣＰＵ１０３と主記憶装置１０４とを有する演算装置１０２、通信インターフェース１０５、外部記憶装置１０７、時計１０８、表示装置１０９を備えるコンピュータと、これらのハードウェア資源を制御するプログラムによって実現することができる。 FIG. 4 is a diagram showing an example of the hardware configuration of the estimation device 2. As shown in FIG. The estimating device 2 includes, for example, a computing device 102 having a CPU 103 and a main memory device 104 connected via a bus 101, a computer having a communication interface 105, an external storage device 107, a clock 108, and a display device 109; It can be implemented by a program that controls hardware resources.
　ＣＰＵ１０３と主記憶装置１０４とは、演算装置１０２を構成する。主記憶装置１０４には、ＣＰＵ１０３が各種制御や演算を行うためのプログラムが予め格納されている。演算装置１０２によって、図２に示した推定装置２の各機能が実現される。 The CPU 103 and the main storage device 104 constitute an arithmetic device 102 . Programs for the CPU 103 to perform various controls and calculations are stored in advance in the main memory device 104 . Each function of the estimation device 2 shown in FIG. 2 is realized by the arithmetic device 102 .
　通信インターフェース１０５は、推定装置２と中継端末２０３など各種外部電子機器との間を通信ネットワークにて接続するためのインターフェースおよび制御装置である。推定装置２は、通信インターフェース１０５を介して、中継端末２０３から通信ネットワークを介して心拍数や心電波形、加速度のデータを受信してもよい。 The communication interface 105 is an interface and control device for connecting the estimation device 2 and various external electronic devices such as the relay terminal 203 via a communication network. The estimating device 2 may receive heart rate, electrocardiographic waveform, and acceleration data from the relay terminal 203 via the communication interface 105 and the communication network.
　通信インターフェース１０５としては、例えば、ＬＴＥ、３Ｇ、無線ＬＡＮ、Ｂｌｕｅｔｏｏｔｈなどの無線データ通信規格に対応した演算インターフェースおよびアンテナが用いられる。通信インターフェース１０５によって、図２における受信部１０が実現される。 As the communication interface 105, for example, a calculation interface and an antenna compatible with wireless data communication standards such as LTE, 3G, wireless LAN, and Bluetooth are used. The communication interface 105 implements the receiver 10 in FIG.
　外部記憶装置１０７は、読み書き可能な記憶媒体と、その記憶媒体に対してプログラムやデータなどの各種情報を読み書きするための駆動装置とで構成されている。外部記憶装置１０７には、記憶媒体としてハードディスクやフラッシュメモリなどの半導体メモリを使用することができる。 The external storage device 107 is composed of a readable and writable storage medium and a drive device for reading and writing various information such as programs and data on the storage medium. A semiconductor memory such as a hard disk or a flash memory can be used as a storage medium for the external storage device 107 .
　外部記憶装置１０７は、センサ端末２０２により計測されたバイタルデータ及び動作データを記憶する記憶領域や、推定装置２がバイタルデータ及び動作データの解析処理を行うためのプログラムを格納するプログラム格納部や、図示しないその他の格納装置で、例えば、この外部記憶装置１０７内に格納されているプログラムやデータなどをバックアップするための格納装置などを有することができる。外部記憶装置１０７によって、図２における受信データ記憶部１１及び自立度推定モデル記憶部１６が実現される。 The external storage device 107 includes a storage area for storing vital data and motion data measured by the sensor terminal 202, a program storage unit for storing a program for the estimation device 2 to analyze the vital data and motion data, Other storage devices (not shown) may include, for example, a storage device for backing up programs and data stored in the external storage device 107 . The received data storage unit 11 and the independence degree estimation model storage unit 16 in FIG. 2 are implemented by the external storage device 107 .
　時計１０８は、推定装置２に設けられた内蔵時計などで構成され、時間を計時する。時計１０８によって得られた時刻情報は、バイタルデータ及び動作データのサンプリングやデータ解析処理に用いられる。 The clock 108 is configured with a built-in clock provided in the estimation device 2, and measures time. The time information obtained by the clock 108 is used for sampling of vital data and motion data and data analysis processing.
　表示装置１０９は、推定装置２の提示部２０として機能する。表示装置１０９は液晶ディスプレイなどによって実現される。 The display device 109 functions as the presentation unit 20 of the estimation device 2. The display device 109 is implemented by a liquid crystal display or the like.
〈変形例〉
　自立度推定モデルは、統計値算出部１４により算出される指標の統計値及び被測定者２０１のプロフィールを入力として、自立度を出力するモデルであってもよい。つまり、自立度推定モデルは、指標の統計値及び被測定者２０１のプロフィールを説明変数とし、自立度を目的変数として生成される。被測定者２０１のプロフィールは例えば被測定者２０１の年齢、身長、体重又は性別である。被測定者２０１のプロフィールは被測定者２０１の年齢、身長、体重及び性別のうちいくつかが組み合わさったデータであってもよい。 <Modification>
The independence degree estimating model may be a model that receives the statistical values of the indices calculated by the statistical value calculation unit 14 and the profile of the subject 201 as inputs and outputs the independence degree. That is, the independence degree estimation model is generated using the statistical values of the indices and the profile of the subject 201 as explanatory variables and the independence degree as the objective variable. The profile of the person to be measured 201 is the age, height, weight or sex of the person to be measured 201, for example. The profile of the person to be measured 201 may be data obtained by combining some of the age, height, weight and gender of the person to be measured 201 .
　図５は、第１の実施形態の変形例に係る推定装置２の構成を示す図である。推定装置２は、被測定者プロフィール取得部２２を備える。被測定者プロフィール取得部２２は被測定者２０１のプロフィールを取得する。自立度推定部１８は、指標の統計値及び被測定者プロフィール取得部２２により取得された被測定者２０１のプロフィールを自立度推定モデルに入力し自立度を出力させることで自立度を推定する。 FIG. 5 is a diagram showing the configuration of the estimation device 2 according to the modification of the first embodiment. The estimation device 2 includes a subject profile acquisition unit 22 . The subject profile acquisition unit 22 acquires the profile of the subject 201 . The degree of independence estimation unit 18 inputs the statistical values of the indices and the profile of the person to be measured 201 acquired by the person to be measured profile acquisition unit 22 into the degree of independence estimation model and outputs the degree of independence, thereby estimating the degree of independence.
　自立度推定モデルは、統計値算出部１４により算出される指標の統計値及び被測定者２０１の過去の自立度を入力として、自立度を出力するモデルであってもよい。つまり、自立度推定モデルは、指標の統計値及び被測定者２０１の過去の自立度を説明変数とし、自立度を目的変数として生成される。 The independence degree estimation model may be a model that takes as inputs the statistical values of the indices calculated by the statistical value calculation unit 14 and the past independence degree of the subject 201 and outputs the independence degree. That is, the independence degree estimation model is generated by using the statistical value of the index and the past independence degree of the subject 201 as explanatory variables and the independence degree as the objective variable.
　図６は、第１の実施形態の変形例に係る推定装置２の構成を示す図である。推定装置２は、過去自立度取得部２４を備える。過去自立度取得部２４は被測定者２０１の過去の自立度を取得する。自立度推定部１８は、指標の統計値及び過去自立度取得部２４により取得された被測定者２０１の過去の自立度を自立度推定モデルに入力し自立度を出力させることで自立度を推定する。 FIG. 6 is a diagram showing the configuration of the estimation device 2 according to the modification of the first embodiment. The estimation device 2 includes a past independence degree acquisition unit 24 . The past independence degree acquisition unit 24 acquires the past independence degree of the subject 201 . The independence degree estimation unit 18 inputs the past independence degree of the subject 201 acquired by the past independence degree acquisition unit 24 and the statistical value of the index into the independence degree estimation model and outputs the independence degree, thereby estimating the independence degree. do.
　自立度推定モデルは、指標の統計値、被測定者２０１のプロフィール及び被測定者２０１の過去の自立度を説明変数とし、自立度を目的変数として生成されてもよい。このとき、自立度推定部１８は、指標の統計値、被測定者２０１のプロフィール及び被測定者２０１の過去の自立度を自立度推定モデルに入力し自立度を出力させることで自立度を推定する。 The independence degree estimation model may be generated using the statistical value of the index, the profile of the subject 201, and the past independence degree of the subject 201 as explanatory variables, and the independence degree as the objective variable. At this time, the independence degree estimating unit 18 inputs the statistical value of the index, the profile of the person to be measured 201, and the past independence degree of the person to be measured 201 into the independence degree estimation model, and outputs the degree of independence, thereby estimating the degree of independence. do.
〈実験例〉
　自立度推定モデルの生成に用いる説明変数を変え、推定された自立度と実際の自立度との相関係数を算出した。図７は、説明変数と相関係数を示した図である。第１条件においては説明変数として被測定者２０１の時系列の指標である指標（％ＨＲＲ、臥位時間、立位座位時間、歩行時間、歩数、活動コスト指数）の各々の３０分単位の統計値に対して5分割交差検定を行うことで相関係数を算出した。第２条件の説明変数は第１条件の説明変数に加え、被測定者２０１のプロフィール（年齢、身長、体重及び性別）が含まれる。第３条件の説明変数は被測定者２０１のプロフィール（年齢、身長、体重及び性別）及び被測定者２０１の過去の自立度である。被測定者２０１の過去の自立度は入院第１週目の自立度であり、目的変数の自立度は入院第５週目の自立度である。つまり、過去の自立度は推定される自立度の４週間前の自立度である。第４条件の説明変数は第１条件の説明変数に加え、被測定者２０１の過去の自立度が含まれる。第５条件の説明変数は第４条件の説明変数に加え、被測定者２０１のプロフィール（年齢、身長、体重及び性別）が含まれる。第２条件から第５条件においても統計値に対して5分割交差検定を行うことで相関係数を算出した。ｋ分割交差検定において、kは評価用のデータセットの数であり、kの値として２、５、又は１０を用いることができる。 <Experimental example>
We changed the explanatory variables used to generate the independence degree estimation model, and calculated the correlation coefficient between the estimated independence degree and the actual independence degree. FIG. 7 is a diagram showing explanatory variables and correlation coefficients. In the first condition, the time-series indices of the subject 201 (%HRR, lying time, standing and sitting time, walking time, number of steps, activity cost index) are statistics in 30-minute increments as explanatory variables. Correlation coefficients were calculated by performing 5-fold cross-validation on the values. The explanatory variables of the second condition include the profile of the person to be measured 201 (age, height, weight and sex) in addition to the explanatory variables of the first condition. The explanatory variables of the third condition are the profile of the subject 201 (age, height, weight and sex) and the degree of independence of the subject 201 in the past. The past independence degree of the subject 201 is the independence degree in the first week of hospitalization, and the independence degree of the objective variable is the independence degree in the fifth week of hospitalization. That is, the past independence degree is the independence degree four weeks before the estimated independence degree. The explanatory variables of the fourth condition include the past degree of independence of the subject 201 in addition to the explanatory variables of the first condition. The explanatory variables of the fifth condition include the profile of the subject 201 (age, height, weight and sex) in addition to the explanatory variables of the fourth condition. The correlation coefficient was calculated by performing 5-fold cross-validation on the statistical values for the second to fifth conditions as well. In k-fold cross-validation, k is the number of data sets for evaluation, and k values of 2, 5, or 10 can be used.
　相関係数は、いずれの条件においても0.6を超える値となった。以上により推定装置２は、自立度を高い精度で推定することができる。 The correlation coefficient exceeded 0.6 under all conditions. As described above, the estimation device 2 can estimate the degree of independence with high accuracy.
　また、被測定者２０１が入院患者である場合、被測定者２０１には睡眠障害がしばしば伴い、自立度に関わらず睡眠の質が悪いことが多い。そのため、睡眠に関するデータと自立度の関連性が低く、被測定者２０１の状態や動作の時間変化を示す時系列の指標が睡眠を示す時系列の指標を含む場合には、推定装置２が自立度を推定する精度が悪くなる。そのため、被測定者２０１の状態や動作の時間変化を示す時系列の指標が睡眠を示す時系列の指標を含む場合には、推定装置２は自立度の影響を受けずに自立度を推定することができ、推定の精度の改善につながる。 In addition, when the person 201 to be measured is a hospitalized patient, the person 201 to be measured often suffers from sleep disorders, and the quality of sleep is often poor regardless of the degree of independence. Therefore, when the relationship between the data on sleep and the degree of independence is low, and the time-series index indicating the time-dependent change in the state and motion of the person to be measured 201 includes the time-series index indicating sleep, the estimation device 2 may The accuracy of estimating the degree becomes worse. Therefore, when the time-series index indicating time-dependent changes in the state and motion of the subject 201 includes a time-series index indicating sleep, the estimation device 2 estimates the degree of independence without being affected by the degree of independence. can be used, leading to improved estimation accuracy.
〈第２の実施形態〉
　第２の実施形態に係る推定装置２は、第１の実施形態に係る推定装置２に加え欠損補完部３０を備える。 <Second embodiment>
The estimating device 2 according to the second embodiment includes a loss complementing unit 30 in addition to the estimating device 2 according to the first embodiment.
　欠損補完部３０は、欠損したデータを補完する。欠損したデータとは、センサ端末２０２の不具合や、中継端末２０３と推定装置２の通信状況などにより推定装置２が受信できなかったデータである。欠損したデータは、所定の上限閾値及び所定の下限閾値に基づいて、その値が上限閾値を超える値又は下限閾値を下回るデータを含んでもよい。
　欠損補完部３０は、欠損したデータを無効にしてもよい。指標算出部１２が、時系列データにおいて同時刻のアンサンブル平均をとることで２４時間にわたる時系列データを生成するとき、欠損補完部３０は、欠損したデータを除く同時刻の平均値をアンサンブル平均として算出してもよい。 The loss complementing unit 30 complements missing data. Missing data is data that could not be received by the estimating device 2 due to a malfunction of the sensor terminal 202, a communication situation between the relay terminal 203 and the estimating device 2, or the like. Missing data may include data whose value exceeds the upper threshold or falls below the lower threshold, based on a predetermined upper threshold and a predetermined lower threshold.
The loss complementing unit 30 may invalidate the missing data. When the index calculation unit 12 generates time-series data over 24 hours by taking the ensemble average of the same time in the time-series data, the loss complementing unit 30 takes the average value of the same time excluding missing data as the ensemble average can be calculated.
　欠損補完部３０は、被測定者２０１のプロフィールや被測定者２０１の過去の自立度に対して、多重代入法を適用することで欠損したデータの補完をしてもよい。 The missing data complementing unit 30 may complement missing data by applying a multiple imputation method to the profile of the subject 201 and the past degree of independence of the subject 201 .
　第２の実施形態に係る推定装置２により、データの欠損の補完をすることができ、欠損のない説明変数のデータセットをより多く扱うことができ、自立度の推定の信頼性を高めることができる。 The estimation device 2 according to the second embodiment can compensate for data deficiencies, handle more data sets of explanatory variables without deficiencies, and improve the reliability of the estimation of the degree of independence. can.
〈他の実施形態〉
　自立度推定モデルの説明変数は、指標の統計値、被測定者２０１のプロフィール又は被測定者２０１の過去の自立度に限られない。例えば、自立度推定モデルの説明変数は被測定者２０１の疾患情報を含んでもよい。被測定者２０１の疾患情報は例えば被測定者２０１が脳血管疾患、脊椎損傷、大腿骨骨折などの疾患を持っているか否かという情報である。自立度推定部１８は、当該説明変数に合わせて、被測定者２０１の疾患情報を自立度推定モデルへの入力に含めてもよい。 <Other embodiments>
The explanatory variables of the degree of independence estimation model are not limited to the statistical values of indices, the profile of the subject 201, or the degree of independence of the subject 201 in the past. For example, explanatory variables of the degree-of-independence estimation model may include disease information of subject 201 . The disease information of the person to be measured 201 is, for example, information as to whether or not the person to be measured 201 has a disease such as a cerebrovascular disease, spinal cord injury, or femoral fracture. The degree-of-independence estimation unit 18 may include the disease information of the person to be measured 201 in the inputs to the degree-of-independence estimation model in accordance with the explanatory variables.
　推定装置２は、推定自立度記憶部４０を備え、自立度推定部１８により推定された自立度を記憶してもよい。また、推定自立度記憶部４０は、指標の統計値、自立度を被測定者２０１のプロフィール、被測定者２０１の過去の自立度又は被測定者２０１の疾患情報と紐づけて記憶してもよい。 The estimation device 2 may include an estimated independence degree storage unit 40 and store the independence degree estimated by the independence degree estimation unit 18 . In addition, the estimated independence degree storage unit 40 may store the statistical value of the index and the independence degree in association with the profile of the subject 201, the past independence degree of the subject 201, or the disease information of the subject 201. good.
　推定装置２は、推定自立度記憶部４０に記憶された推定自立度に基づいて、推定自立度の統計値を算出する推定統計値算出部４２を備えてもよい。推定統計値算出部４２は、被測定者２０１のプロフィールに基づいて推定自立度の統計値を算出してもよい。例えば、推定統計値算出部４２は、５０代、６０代など年齢ごとに推定自立度の統計値である平均値や標準偏差を算出する。提示部２０は、推定統計値算出部４２により算出された結果を提示してもよい。 The estimation device 2 may include an estimated statistic value calculation unit 42 that calculates the statistic value of the estimated independence degree based on the estimated independence degree stored in the estimated independence degree storage unit 40 . The estimated statistic value calculation unit 42 may calculate the statistic value of the estimated degree of independence based on the profile of the subject 201 . For example, the estimated statistic value calculation unit 42 calculates an average value and a standard deviation, which are statistic values of the estimated degree of independence, for each age group such as 50's and 60's. The presentation unit 20 may present the results calculated by the estimated statistical value calculation unit 42 .
　自立度推定モデルの説明変数と目的変数を異なるものにすることで、自立度以外の特性を推定するモデルを作成することができる。例えば、目的変数として被測定者２０１のプロフィールの１項目を、説明変数として指標の統計値と、被測定者２０１のプロフィールのうち目的変数でない項目、被測定者２０１の自立度及び被測定者２０１の疾患情報のうち少なくとも１つとし、被測定者２０１のプロフィールの１項目を推定する推定モデルを作成してもよい。
　当該推定モデルに説明変数を入力することで、被測定者２０１のプロフィールの１項目の推定値が出力される。例えば、当該推定モデルに説明変数を入力すると、被測定者２０１の年齢の推定値が出力される場合、年齢の推定値が実年齢より低ければ、説明変数に含まれる指標の統計値や自立度が良好であることを被測定者２０１は把握することができる。 By using different explanatory variables and objective variables for the independence degree estimation model, it is possible to create a model that estimates characteristics other than the independence degree. For example, one item in the profile of the subject 201 as the objective variable, the statistical value of the index as the explanatory variable, the items in the profile of the subject 201 that are not the objective variable, the degree of independence of the subject 201, and the subject 201 , and an estimation model for estimating one item of the profile of the person to be measured 201 may be created.
By inputting an explanatory variable into the estimation model, an estimated value of one item of the profile of the subject 201 is output. For example, when an explanatory variable is input to the estimation model, an estimated value of the age of the person to be measured 201 is output. The person to be measured 201 can grasp that the is good.
　例えば、目的変数として被測定者２０１が持っている疾患を、説明変数として指標の統計値、被測定者２０１のプロフィールの項目、被測定者２０１の自立度のうち少なくとも１つとし、疾患を推定する推定モデルを作成してもよい。 For example, the disease of the person to be measured 201 is set as the objective variable, and at least one of the statistical value of the index, the profile item of the person to be measured 201, and the degree of independence of the person to be measured 201 is set as the explanatory variable, and the disease is estimated. You may create an estimation model that
　自立度推定モデルがオートエンコーダである場合、提示部２０は、自立度推定モデルが圧縮したデータを提示してもよい。オートエンコーダの圧縮により、説明変数の次元が圧縮され、学習に必要な本質的な情報のみが残ることから、本質的な情報を把握することができる。 When the independence estimation model is an autoencoder, the presentation unit 20 may present data compressed by the independence estimation model. By compressing the autoencoder, the dimensions of the explanatory variables are compressed, leaving only essential information necessary for learning, so essential information can be grasped.
　推定装置２とセンサ端末２０２とは同一の装置により実現されてもよい。このとき、推定装置２とセンサ端末２０２とは中継端末２０３を介して通信しなくてもよい。また、中継端末２０３が推定装置２の一部機能を行ってもよい。例えば、中継端末２０３が指標算出部１２、統計値算出部１４、自立度推定部１８及び提示部２０の機能の一部を行い、処理されたデータを推定装置２に送信してもよい。 The estimation device 2 and the sensor terminal 202 may be realized by the same device. At this time, the estimation device 2 and the sensor terminal 202 do not need to communicate via the relay terminal 203 . Also, the relay terminal 203 may perform a part of the functions of the estimation device 2 . For example, the relay terminal 203 may perform part of the functions of the index calculation unit 12 , the statistical value calculation unit 14 , the independence estimation unit 18 , and the presentation unit 20 and transmit the processed data to the estimation device 2 .
　１…推定システム、２…推定装置、１０…受信部、１１…受信データ記憶部、１２…指標算出部、１４…統計値算出部、１６…自立度推定モデル記憶部、１８…自立度推定部、２０…提示部、２２…被測定者プロフィール取得部、２４…過去自立度取得部、１０１…バス、１０２…演算装置、１０３…ＣＰＵ、１０４…主記憶装置、１０５…通信インターフェース、１０７…外部記憶装置、１０８…時計、１０９…表示装置、２０１…被測定者、２０２…センサ端末、２０３…中継端末 DESCRIPTION OF SYMBOLS 1... Estimation system, 2... Estimation apparatus, 10... Reception part, 11... Received data storage part, 12... Index calculation part, 14... Statistics calculation part, 16... Independence degree estimation model storage part, 18... Independence degree estimation part , 20... presentation unit, 22... subject profile acquisition unit, 24... past independence degree acquisition unit, 101... bus, 102... arithmetic device, 103... CPU, 104... main storage device, 105... communication interface, 107... external Storage device 108 Clock 109 Display device 201 Subject 202 Sensor terminal 203 Relay terminal

Claims (9)

  1. 　被測定者の状態や動作に関し、前記被測定者の循環器の状態に関するバイタルデータ及び物理的な振動や角度に関する動作データに基づいて、時系列の指標を算出する指標算出部と、
    　前記時系列の指標に基づいて、前記時系列の指標の統計値である指標統計値を算出する統計値算出部と、
    　前記指標統計値に基づいて、前記被測定者の自立度を推定する自立度推定部と、
    　を備える推定装置。     an index calculation unit that calculates a time-series index regarding the condition and movement of a person to be measured based on vital data on the condition of the circulatory system of the person to be measured and movement data on physical vibrations and angles;
    a statistical value calculation unit that calculates an index statistical value, which is a statistical value of the time-series index, based on the time-series index;
    an independence degree estimating unit that estimates the degree of independence of the person to be measured based on the index statistics;
    An estimating device comprising:
  2. 　前記自立度推定部は、前記被測定者のプロフィールに基づいて自立度を推定する、
    　請求項１に記載の推定装置。     The degree of independence estimation unit estimates the degree of independence based on the profile of the person to be measured.
    The estimating device according to claim 1.
  3. 　前記自立度推定部は、前記被測定者の過去の自立度に基づいて自立度を推定する、
    　請求項１又は２に記載の推定装置。     The independence degree estimating unit estimates the degree of independence based on the past independence degree of the person to be measured.
    The estimation device according to claim 1 or 2.
  4. 　前記バイタルデータ及び前記動作データの欠損したデータを補完する欠損補完部と、
    　をさらに備える請求項１から３のいずれか一項に記載の推定装置。     a loss complementing unit that complements missing data of the vital data and the motion data;
    The estimation device according to any one of claims 1 to 3, further comprising:
  5. 　前記自立度推定部は、前記被測定者の疾患情報に基づいて自立度を推定する、
    　請求項１から４のいずれか一項に記載の推定装置。     The degree-of-independence estimating unit estimates the degree of independence based on the subject's disease information.
    The estimation device according to any one of claims 1 to 4.
  6. 　前記自立度推定部により推定された自立度に基づいて、自立度の統計値を算出する推定統計値算出部と、
    　前記推定統計値算出部により算出された統計値を提示する提示部と、
    　をさらに備える請求項１から５のいずれか一項に記載の推定装置。     an estimated statistical value calculation unit that calculates a statistical value of the degree of independence based on the degree of independence estimated by the degree of independence estimation unit;
    a presentation unit that presents the statistical values calculated by the estimated statistical value calculation unit;
    The estimation apparatus according to any one of claims 1 to 5, further comprising:
  7. 　被測定者の状態や動作に関し、前記被測定者の循環器の状態に関するバイタルデータ及び物理的な振動や角度に関する動作データに基づいて、時系列の指標を算出する指標算出ステップと、
    　前記時系列の指標に基づいて、前記時系列の指標の統計値である指標統計値を算出する統計値算出ステップと、
    　前記指標統計値に基づいて、前記被測定者の自立度を推定する自立度推定ステップと、
    　を有する推定方法。     an index calculation step of calculating a time-series index regarding the condition and movement of the subject based on vital data regarding the condition of the circulatory system of the subject and movement data regarding the physical vibration and angle;
    a statistical value calculation step of calculating an index statistical value, which is a statistical value of the time-series index, based on the time-series index;
    an independence degree estimation step of estimating the degree of independence of the subject based on the index statistics;
    An estimation method with
  8. 　請求項１から６のいずれか一項に記載の推定装置としてコンピュータを機能させるためのプログラム。 A program for causing a computer to function as the estimation device according to any one of claims 1 to 6.
  9. 　請求項１から６のいずれか一項に記載の推定装置としてコンピュータを機能させるためのプログラムを記録したコンピュータで読み取り可能な記録媒体。 A computer-readable recording medium recording a program for causing a computer to function as the estimation device according to any one of claims 1 to 6.
PCT/JP2022/006369 2022-02-17 2022-02-17 Estimation device, estimation method, program, and storage medium WO2023157172A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/006369 WO2023157172A1 (en) 2022-02-17 2022-02-17 Estimation device, estimation method, program, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/006369 WO2023157172A1 (en) 2022-02-17 2022-02-17 Estimation device, estimation method, program, and storage medium

Publications (1)

Publication Number Publication Date
WO2023157172A1 true WO2023157172A1 (en) 2023-08-24

Family

ID=87577829

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/006369 WO2023157172A1 (en) 2022-02-17 2022-02-17 Estimation device, estimation method, program, and storage medium

Country Status (1)

Country Link
WO (1) WO2023157172A1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014023825A (en) * 2012-07-30 2014-02-06 Mitsubishi Chemical Holdings Corp Ageing degree evaluation method, ageing degree evaluation device, program for executing computer, and computer readable record medium recording the same
JP2020161136A (en) * 2019-03-20 2020-10-01 国立大学法人電気通信大学 Information processing device
JP2021033441A (en) * 2019-08-20 2021-03-01 日本電信電話株式会社 System, method, and program for supporting rehabilitation
WO2021065182A1 (en) * 2019-10-04 2021-04-08 日本電気株式会社 Rehabilitation planning device, rehabilitation planning system, rehabilitation planning method, and computer-readable medium
JP2021131577A (en) * 2020-02-18 2021-09-09 Ｒｅｈａｂｉｌｉｔａｔｉｏｎ３．０株式会社 Occupational therapy supporting device, artificial intelligence learning device for occupational therapy supporting device, and method for using occupational therapy supporting device
JP2021157567A (en) * 2020-03-27 2021-10-07 株式会社博報堂Ｄｙメディアパートナーズ Information processing device and computer program

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014023825A (en) * 2012-07-30 2014-02-06 Mitsubishi Chemical Holdings Corp Ageing degree evaluation method, ageing degree evaluation device, program for executing computer, and computer readable record medium recording the same
JP2020161136A (en) * 2019-03-20 2020-10-01 国立大学法人電気通信大学 Information processing device
JP2021033441A (en) * 2019-08-20 2021-03-01 日本電信電話株式会社 System, method, and program for supporting rehabilitation
WO2021065182A1 (en) * 2019-10-04 2021-04-08 日本電気株式会社 Rehabilitation planning device, rehabilitation planning system, rehabilitation planning method, and computer-readable medium
JP2021131577A (en) * 2020-02-18 2021-09-09 Ｒｅｈａｂｉｌｉｔａｔｉｏｎ３．０株式会社 Occupational therapy supporting device, artificial intelligence learning device for occupational therapy supporting device, and method for using occupational therapy supporting device
JP2021157567A (en) * 2020-03-27 2021-10-07 株式会社博報堂Ｄｙメディアパートナーズ Information processing device and computer program

Similar Documents

Publication Publication Date Title
US10456078B2 (en) Wearable device and system for preventative health care for repetitive strain injuries
US9220444B2 (en) System method and device for determining the risk of dehydration
CN108348745B (en) System and method for predicting healthcare events
GB2436721A (en) Automated method for adapting the settings of a patient monitor
US20200085366A1 (en) Wearable sensor device and analysis platform for objective outcome assessment in spinal diseases
US20230148878A1 (en) System And Method for Determining an Adaptive Physiological Strain Index
Geman et al. Ubiquitous healthcare system based on the sensors network and android internet of things gateway
WO2014190254A1 (en) A system and method for assessing the clinical stability of critically ill patients under intensive care
WO2021070472A1 (en) Information processing device, information processing system, and information processing method
JP6534192B2 (en) Patient health composite score distribution and / or representative composite score based thereon
JP6373641B2 (en) Medical information processing system, medical information processing method, and medical information processing program
US20200126649A1 (en) Data processing apparatus, data processing method, and data processing program
CN113598721B (en) Wearable terminal, core body temperature monitoring method thereof and computer readable storage medium
US20200281536A1 (en) Personal health monitoring
JP6702559B2 (en) Electronic device, method and program
WO2023157172A1 (en) Estimation device, estimation method, program, and storage medium
US10456037B2 (en) Terminal device and information processing system
CN114947773A (en) Body temperature monitoring method and device, terminal equipment and storage medium
EP3900627A1 (en) Motion limiting apparatus for assessing status of spinal implants
JP7298685B2 (en) Rehabilitation support system and rehabilitation support method
JP7180216B2 (en) Biological information analysis device, biological information analysis method, and biological information analysis system
US11096580B2 (en) Adaptive health score
KR101148424B1 (en) Health care system for using wire/wireless network in ubiquitous environment
JP7294449B2 (en) MONITORING SYSTEM, MONITORING METHOD AND MONITORING PROGRAM
WO2022208581A1 (en) Learning device, determination device, method for generating trained model, and recording medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22927074

Country of ref document: EP

Kind code of ref document: A1