JP2023138175A - Processing device, program, method, and processing system - Google Patents

Abstract

To provide a processing device, a program, a method, and a processing system that can more easily output conditions of knees and risks related to the knees.SOLUTION: A processing system 1 includes: a detection device 200 that is attached to a user and detects an output value during exercise of the user; and a processing device 100 that is communicably connected to the detection device 200 and processes the detected output value. The processing system 1 is connected to a server device 300 via a network for wireless communication. The server device 300 includes a processor, a memory, a communication interface, and the like, and appropriately transmits and receives instruction commands, information, and the like necessary for processing by the processing device 100.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a processing device, a program, a method, and a processing system capable of estimating or assisting in estimating a state during human exercise.

It has long been known that the knee joint is the largest joint in the human body and plays a major role in ensuring smooth and stable movement of the leg, especially during movements such as walking and changing direction. However, the knee joint causes various symptoms due to wear and tear on the bones, cartilage, ligaments, etc. of the knee joint. For example, knee osteoarthritis caused by wear of the cartilage of the knee joint is known to cause pain during exercise such as walking. Thus, it is extremely important to appropriately evaluate the condition of the knee and know the degree of risk to the knee.

For example, in Patent Document 1, in order to evaluate the state of movement of an evaluation subject, analysis is performed using image data taken of either the front or back of the subject and image data taken of either the left or right side of the subject. However, a motion evaluation device for calculating parameters related to body motion is described.

Japanese Patent Application Publication No. 2016-140591

However, using such a large-scale evaluation device to estimate or assist with estimation of the condition during exercise has placed a heavy burden on users who are at some risk due to knee conditions. . Therefore, various embodiments of the present disclosure provide a processing device, a program, a method, and a processing system that can more easily output the condition of the knee and the risks related to the knee.

According to one aspect of the present disclosure, there is provided a processing device comprising at least one processor, wherein the at least one processor is attached to or around the knee of a leg of a user, Receive the acceleration detected by wire or wirelessly from a sensor for detecting acceleration via a communication interface, calculate an index indicating the state of the knee based on the received acceleration, and calculate the calculated index. A processing device configured to perform processing for outputting information regarding the degree of risk of the knee based on an index and a predetermined threshold value is provided.

According to one aspect of the present disclosure, "in a processing device comprising at least one processor, the at least one processor is attached to or around the knee of a user's leg, and detects acceleration during exercise of the user. The acceleration detected by wire or wirelessly is received from the sensor for the purpose of the test, via a communication interface, an index indicating the state of the knee is calculated based on the received acceleration, and the calculated index and the previously detected acceleration are calculated. A computer program is provided that functions to output information regarding the degree of risk of the knee based on a determined threshold value.

According to one aspect of the present disclosure, there is provided a method, in a processing device comprising at least one processor, performed by the at least one processor, the method comprising a step of receiving the acceleration detected by wire or wirelessly via a communication interface from a sensor for detecting acceleration during exercise; and calculating an index indicating the condition of the knee based on the received acceleration. and outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold.

According to one aspect of the present disclosure, "the processing device described above is connected to the processing device by wire or wirelessly, is attached to the knee of a user's leg or its surroundings, and is attached to the knee of a user's leg or its surroundings, and and a sensor configured to detect.

According to the present disclosure, it is possible to provide a processing device, a program, a method, and a processing system that can more easily output the condition of the knee and the risks related to the knee.

Note that the above effects are merely illustrative for convenience of explanation, and are not limiting. In addition to or in place of the above effects, any effects described in the present disclosure or effects obvious to those skilled in the art may be achieved.

FIG. 1 is a diagram showing a usage state of a processing system 1 according to the present disclosure. FIG. 2 is a schematic diagram of the processing system 1 according to the embodiment of the present disclosure. FIG. 3 is a block diagram showing the configuration of the processing system 1 according to the embodiment of the present disclosure. FIG. 4 is a diagram showing the appearance of the detection device 200 according to the embodiment of the present disclosure. FIG. 5A is a diagram illustrating an example of an acceleration table stored in the processing device 100 according to the embodiment of the present disclosure. FIG. 5B is a diagram illustrating an example of a user table stored in the processing device 100 according to the embodiment of the present disclosure. FIG. 5C is a diagram illustrating an example of a KAM threshold table stored in the processing device 100 according to the embodiment of the present disclosure. FIG. 6 is a diagram illustrating a processing flow related to generation of a trained estimation model according to an embodiment of the present disclosure. FIG. 7 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. FIG. 8 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. FIG. 9 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. FIG. 10 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. FIG. 11 is a diagram showing the correlation between actually measured acceleration and KAM value. FIG. 12 is a diagram showing the correlation between actually measured output values and KAM values. FIG. 13 is a diagram showing the correlation between actually measured output values and KAM values. FIG. 14 is a diagram showing an example of a screen output from the output interface 114 of the processing device 100.

Various embodiments of the present disclosure will be described with reference to the accompanying drawings. Note that common components in the drawings are given the same reference numerals.

1. Overview of Processing System 1 The processing system 1 according to the present disclosure includes a processing device 100 and a detection device 200, and processes an output value detected by the detection device 200 attached to the user in the processing device 100, so that the processing system 1 according to the present disclosure can be used. It is used to calculate an index indicating the condition of a person's knee and output information regarding the degree of risk. In particular, the processing system 1 is attached to the user's knee or its surroundings, and uses the output value detected by the detection device 200 when the user exercises to determine or determine the degree of risk for knee osteoarthritis. Used to assist. Therefore, in the following, a case will be mainly described in which the processing system 1 according to the present disclosure is used to determine or assist in determining the degree of risk for knee osteoarthritis. However, the degree of risk for knee osteoarthritis is only one example of the degree of risk for the knee, and the mounting position of the detection device is also only one example.

FIG. 1 is a diagram showing a usage state of a processing system 1 according to the present disclosure. Specifically, it is a diagram showing a state in which a detection device 200 of the processing system 1 is attached to a user 10 and is being used. According to FIG. 1, the detection device 200a of the processing system 1 is attached to the lower part of the right knee 11a of the right leg 12a of the user 10 by wrapping the auxiliary tool 400a around the right leg 12a. Similarly, the detection device 200b of the processing system 1 is attached to the lower part of the left knee 11b of the left leg 12b of the user 10 by wrapping the auxiliary tool 400b around the left leg 12b. After that, the user wearing the detection device 200a and the detection device 200b is caused to walk in a predetermined direction, and the detection device 200a and the detection device 200b detect output values (for example, acceleration) that occur during this walking movement. do. Then, the detection device 200a and the detection device 200b transmit the detected output values to the processing device 100 (not shown in FIG. 1) of the processing system 1.

Note that in the present disclosure, the user to whom the detection device 200 is attached may include any human being, such as a patient, a subject, and a person to be diagnosed. The detection device 200 according to the present disclosure is not limited to being used in a medical institution, for example, but may be used in any place such as a sports gym, an orthopedic clinic, an osteopathic clinic, or even a user's workplace or home. The attributes of the person wearing the detection device 200 do not matter. Furthermore, in the present disclosure, an operator simply means a person who operates the processing device 100. Therefore, the user may be the same as the above-mentioned user, or may be a different person, such as a medical worker or a gym trainer.

Further, although the example in FIG. 1 shows an example in which the detection device 200a or 200b is attached below the knee (on the front side), it may be attached above the knee, on the knee, or on any other part of the leg. Further, the detection device 200a or 200b is not limited to the front side of the leg 12a or 12b, but may be attached to either the rear side or the side side. Further, although FIG. 1 shows an example in which two detection devices 200a and 200b are used, the number of detection devices 200a and 200b may naturally be one on one leg side, or multiple detection devices on each leg 12a and/or 12b. Two or more pieces may be attached to a position.

Further, an acceleration sensor is typically used as the detection device 200, and acceleration during exercise is detected. However, the sensor is not limited to only an acceleration sensor, but any sensor that can detect the movement of the user 10, especially movement such as bending and straightening the knee, such as a gyro sensor, a geomagnetic sensor, and a telescopic sensor, can be used. It is possible to use the corresponding output value to estimate the motion state or to assist therein. It is also possible to use a combination of multiple sensors, such as an acceleration sensor and a gyro sensor.

Further, the state of the knee that is judged or assisted in judgment by the processing system 1 is the state during exercise. This movement is typically the user's walking, but may include various other movements such as running, bending and stretching, and jumping. It is not necessary to perform these movements only for estimating the state or assisting in the estimation; for example, the detection device 200 may be worn on the user 10 on a daily basis to detect output values during exercise during daily life. .

Further, the auxiliary tools 400 (auxiliary tools 400a and 400b) may be any device as long as it can assist in attaching the detection device 200 to the user. Typically, a flexible band-like body is used, but adhesive plasters, taping tapes, bandages, bandages, wound dressings, adhesive tapes, supporters, etc. may also be used. Furthermore, the auxiliary tool 400 does not need to be configured as a separate body that can be separated from the detection device 200, and double-sided tape directly attached to the detection device 200, a wristwatch-shaped band, etc. can also be used as the auxiliary tool 400. is possible.

2. Configuration of Processing System 1 FIG. 2 is a schematic diagram of the processing system 1 according to the embodiment of the present disclosure. The processing system 1 includes a detection device 200 that is attached to a user and detects an output value during exercise of the user, and a processing device that is communicably connected to the detection device 200 and processes the detected output value. 100. Such processing system 1 is connected to server device 300 via a network for wireless communication. The server device 300 includes a processor, a memory, a communication interface, etc., and appropriately transmits and receives instructions, information, etc. necessary for processing by the processing device 100.

FIG. 3 is a block diagram showing the configuration of the processing system 1 according to the embodiment of the present disclosure. According to FIG. 3, the processing system 1 includes a processing device 100 and a detection device 200 that is communicably connected to the processing device 100 wirelessly or by wire. The processing device 100 receives an operation input from the user and controls the detection of an output value by the detection device 200 during exercise. Furthermore, the processing device 100 processes the output value detected by the detection device 200 to determine or assist in determining the state of the user's knee during exercise. Furthermore, the processing device 100 allows the user etc. to check the output value detected by the detection device 200, an index indicating the knee condition calculated based on the output value, and information regarding the degree of risk of the knee. .

The processing system 1 includes a processing device 100 including a processor 111, a memory 112, an operation input interface 113, a display 114, and a communication interface 115, and a detection device 200 including a processor 211, a sensor 212, a memory 213, and a communication interface 214. Each of these components is electrically connected to each other via control lines and data lines. Note that the processing system 1 does not need to include all of the components shown in FIG. 3; it is possible to omit some of them, or to add other components. For example, the processing system 1 can include a battery for driving each component, a communication interface for transmitting results processed by the processing system 1 to the outside, receiving instructions from the outside, etc. be.

Note that the processing system 1 includes the processing device 100 and the detection device 200 as separate bodies. However, the present invention is not limited to this, and it is also possible to configure the processing device 100 and the detection device 200 as one unit, for example, in a wearable terminal device, a smartphone, or the like. Further, the processing device 100 is not limited to being configured as a single component, and causes other components connected by wire or wirelessly (for example, a cloud server device, etc.) to execute at least a part of the processing. In such a case, the other components may also be referred to as the processing device 100. Specifically, the learned KAM value estimation model is stored in the memory of the cloud server device, the processor of the cloud server device calculates the KAM value based on the KAM value estimation model, and uses the calculated KAM value and the threshold value. There are cases where the degree of risk of knee osteoarthritis is determined by comparing the results and output on a display such as a smartphone via a communication interface. In such a case, naturally, the cloud server device itself or a combination of the cloud server device and the smartphone functions as the processing device 100.

First, the processing device 100 will be explained based on FIG. The processor 111 functions as a control unit that controls other components of the processing system 1 based on a program stored in the memory 112. The processor 111 controls the driving of each component of the detection device 200 based on the program stored in the memory 112, stores the output value received from the detection device 200 in the memory 112, and stores the stored output value. Process. Specifically, the processor 111 receives acceleration detected via the communication interface 115 from a detection device attached to or around the knee of the user's leg for detecting acceleration during exercise of the user. , a process of calculating an index indicating the state of the knee based on the received acceleration, a process of outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold value, etc., are stored in the memory 112. Execute based on the stored program. The processor 111 is mainly composed of one or more CPUs, but may be appropriately combined with a GPU or the like.

The memory 112 includes RAM, ROM, nonvolatile memory, HDD, etc., and functions as a storage unit. The memory 112 stores instructions for various controls of the processing system 1 according to the present embodiment as programs. Specifically, the memory 112 receives acceleration detected via the communication interface 115 from a detection device attached to or around the knee of the user's leg for detecting acceleration during movement of the user. The processor 111 executes processes such as calculating an index indicating the state of the knee based on the received acceleration, and outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold. Memorize the program to do this. In addition to the program, the memory 112 also stores an acceleration table, a user table, a KAM threshold table, and the like. Furthermore, when machine learning is used to calculate the KAM value, the memory 112 stores a learned KAM value estimation model. Note that the memory 112 can be a storage medium communicably connected to the outside, or a combination of such storage media.

The operation input interface 113 functions as an operation input unit that receives instructions input from the user to the processing device 100 and the detection device 200. Examples of the operation input interface 113 include a "start button" for instructing the detection device 200 to start and end detection, a "confirm button" for making various selections, and returning to the previous screen or canceling the entered confirmation operation. Examples include a "back/cancel button" for moving the icons, etc., a cross key button for moving icons displayed on the display 114, an on/off key for turning on and off the power of the processing device 100, and the like. Note that it is also possible to use a touch panel for the operation input interface 113, which is provided to be superimposed on the display 114 and has an input coordinate system corresponding to the display coordinate system of the display 114. The method for detecting the user's instruction input using the touch panel may be any method such as a capacitance method or a resistive film method.

The display 114 displays the output value detected by the detection device 200 or the value calculated based on the output value, or displays an index indicating the knee condition or knee risk calculated based on the output value. It functions as a display section for displaying the degree, etc. Although it is composed of a liquid crystal panel, it is not limited to a liquid crystal panel, and may be composed of an organic EL display, a plasma display, or the like.

The communication interface 115 sends and receives various commands related to the start of detection, output values detected by the detection device 200, etc. to the detection device 200 connected by wire or wirelessly, and sends and receives information to and from the server device 300. It functions as a communications department for Examples of the communication interface 115 include connectors for wired communication such as USB and SCSI, transmitting/receiving devices for wireless communication such as LTE, Bluetooth (registered trademark), WiFi, and infrared rays, and various connections for printed mounting boards and flexible mounting boards. There are various types of terminals and their combinations.

An example of such a processing device 100 is a portable terminal device capable of wireless communication, typified by a smartphone. However, in addition to this, any device that can execute the process according to the present disclosure can be suitably applied, such as a tablet terminal, laptop computer, desktop computer, feature phone, personal digital assistant, or PDA. It is.

Next, the detection device 200 will be explained. The processor 211 functions as a control unit that controls other components of the detection device 200 based on a program stored in the memory 213. Based on the program stored in the memory 213, the processor 211 specifically performs processing for controlling the detection of the output value by the sensor 212, processing for storing the detected output value in the memory 213, and processing for storing the detected output value in the memory 213. It executes processing such as transmitting the obtained output value to the processing device 100 via the communication interface 214. The processor 111 is mainly composed of one or more CPUs, but may be appropriately combined with a GPU or the like.

The memory 213 includes RAM, ROM, nonvolatile memory, HDD, etc., and functions as a storage unit. The memory 213 stores instructions for various controls of the detection device 200 according to this embodiment as a program. Specifically, the memory 213 performs processing for controlling detection of an output value by the sensor 212, processing for storing the detected output value in the memory 213, and processing for the output value stored in the memory 213 via the communication interface 214. A program for the processor 211 to execute processing such as sending to the device 100 is stored. Furthermore, the memory 213 stores the output value detected by the sensor 212 in addition to the program. Note that the memory 112 can be a storage medium communicably connected to the outside, or a combination of such storage media.

The sensor 212 is driven by instructions from the processor 211 and functions as a detection unit for detecting an output value when the user 10 exercises. As the sensor 212, an acceleration sensor is used, for example. The acceleration sensor detects the rate of change in the amount of movement (velocity) per unit time. The types include a capacitance method, a piezo method, a heat detection method, etc., and any method can be suitably used. Further, it is preferable that the acceleration sensor is capable of detecting at least acceleration in the horizontal direction, and further detecting acceleration in the vertical direction and/or acceleration in the depth direction. Further, as the sensor 212, a gyro sensor can also be used in combination with an acceleration sensor. In this case, it is possible to obtain three output values from the gyro sensor: an angular velocity with respect to the horizontal axis, an angular velocity with respect to the vertical axis, and an angular velocity with respect to the depth axis. That is, in addition to a total of three accelerations in the horizontal direction, vertical direction, and depth direction, the above three angular velocities (that is, output values of a total of six axes) can be used. In addition to this example, it is possible to use an appropriate combination of sensors capable of detecting the movement of the user 10, particularly movements such as bending/extending the knee or shaking, such as a geomagnetic sensor or a telescopic sensor.

The communication interface 214 functions as a communication unit for transmitting and receiving various commands related to the start of detection, output values detected by the detection device 200, etc. to and from the processing device 100 connected by wire or wirelessly. Examples of the communication interface 214 include wired communication connectors such as USB and SCSI, wireless communication transmitting and receiving devices such as LTE, Bluetooth (registered trademark), WiFi, and infrared rays, and various connections for printed mounting boards and flexible mounting boards. There are various types of terminals and their combinations.

FIG. 4 is a diagram showing the appearance of the detection device 200 according to the embodiment of the present disclosure. Specifically, an example of the appearance when an acceleration sensor is used as the sensor 212 of the detection device 200 is shown. According to FIG. 4, the detection device 200 has a power switch 216 on its top surface that turns on/off the power of the detection device 200. Furthermore, the detection device 200 has a USB terminal 215 as an example of the communication interface 214. Furthermore, the detection device 200 includes an indicator 217 for notifying a driving state such as an abnormality.

Here, KAM (external knee varus moment) value is used as an index for evaluating the degree and prognosis of knee osteoarthritis. It is said that when this KAM value exceeds a certain value, the risk of knee osteoarthritis progressing after a certain period of time is high, and when it falls below another certain value, the risk of progression is low (Reference 1: T Miyazakira, Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis, Ann Rheum Dis., 20 02, No. 61, pp. 617-622, Reference 2: Kim L. Bennell et all, Higher dynamic medial knee load predictors greater cartilage loss over 12 months in medial knee osteoarthritis, Ann Rheum Dis., 2011, No. 70, pp. 1770-1774, Reference 3: Nicholas M. Brisson, Baseline Knee Adduc tion Moment Interacts With Body Mass Index to Predict Loss of Medial Tibial Cartilage Volume Over 2.5 Years in Knee Osteoarthritis, JOURNAL OF ORTHOPAEDIC RESEARCH, 2017, NOVEMBER, pp. 2476-2483). Here, there is a certain correlation between the KAM value and the acceleration detected by the acceleration sensor attached to or around the knee. Therefore, by calculating the KAM value from the detected acceleration, it is possible to determine or assist in determining the degree or prognosis of knee osteoarthritis.

3. Information Stored in Processing Device 100 FIG. 5A is a diagram illustrating an example of an acceleration table stored in processing device 100 according to an embodiment of the present disclosure. The acceleration table is prepared for each user and stored in association with user ID information that identifies the user. As an example, FIG. 5A shows an acceleration table for a user whose user ID information is "U1". According to FIG. 5A, acceleration information is stored in the acceleration table in association with time information. “Time information” is information that specifies the time when each acceleration was measured in the detection device 200. The information may be information on a specific date and time using a timer included in the detection device 200, or may be information on elapsed time from the start of measurement. "Acceleration information" is information indicating a specific acceleration value detected in the corresponding time information. When both "time information" and "acceleration information" are detected by the detection device 200, they are transmitted from the detection device 200 and stored in the memory 112 of the processing device 100 that received them. Note that in FIG. 5A, horizontal acceleration is typically stored as acceleration information in association with each piece of time information. However, the present invention is not limited to this, and in addition to the horizontal acceleration, vertical acceleration may also be stored in association with each piece of time information.

FIG. 5B is a diagram illustrating an example of a user table stored in the processing device 100 according to the embodiment of the present disclosure. According to FIG. 5B, the user table includes user name information, gender information, KAM value information, estimated KL information, progression risk information, walking stability information, and fall risk information in association with user ID information. are respectively memorized. The estimated KL information, progression risk information, walking stability information, and fall risk information are used as an example of information indicating the degree of knee risk. "User ID information" is information that is generated each time the detection device 200 is attached and a user whose acceleration is to be measured is newly registered, and is information unique to each user. This is information to identify the person. "Gender information" is information indicating the user's gender. Typically, the biological sex of the user or the sex selected based on the operational input received via the input interface 116 is stored. "User name information" is information indicating the user's name displayed on, for example, the display 114 of the processing device. "KAM value information" is information calculated based on the acceleration detected by the detection device 200, and is used as an index indicating the state of the knee. As mentioned above, the KAM value is information used to determine the degree of knee osteoarthritis, which is one of the risks of the knee.

Here, the walking movement is performed by periodically repeating a stance phase from landing on the heel to the release of the fingertips, and a free period from the release of the fingertips to landing on the heel. Therefore, a two-dimensional curve (KAM value curve) is plotted with time on the horizontal axis and KAM values calculated at each time on the vertical axis from the output value (for example, acceleration) from the detection device 200 during such a walking movement. is required. In the present disclosure, the following two KAM values are appropriately used as the KAM value. As the first KAM value, the highest peak value (KAM peak value) detected during the stance phase is used. This KAM peak value can reflect the value at the moment when the largest force is applied to the hip joint during the stance phase. As the second KAM value, the area value (KAM area value) between the KAM value curve and the horizontal axis (straight line) in the stance phase can be used. This KAM area value can reflect the value of the entire load applied to the hip joint during the stance phase.

It is also known that knee osteoarthritis causes pain when walking and affects the ability to bend and straighten the knee joint, making walking difficult. Furthermore, it is known that these factors cause a loss of left and right balance, increasing the risk of falling. The degree of progression of such knee osteoarthritis is generally classified according to the Kellgren-Lawrence classification (KL). KL = 2 (early stage of knee osteoarthritis), KL = 3 (middle stage of knee osteoarthritis), and KL = 4 (late stage of knee osteoarthritis). Normally, the above classification of KL is determined based on medical images near the knee joint obtained from X-rays, etc., but as shown in Figure 5C, it is based on the KAM value (that is, the external knee joint varus moment value). It is also possible to estimate the Therefore, "estimated KL information" is information indicating KL estimated based on the calculated KAM value. This information stores three classifications: healthy, equivalent to KL=0 to 2, and equivalent to KL=3 to 4.

"Progression risk information" is information indicating the degree of progression risk of knee osteoarthritis. If the estimated KL based on the KAM value is classified as equivalent to KL=3 to 4, it can be said that the risk of knee osteoarthritis progressing is high. Therefore, if the estimated KL information is classified as "equivalent to healthy", the progression risk of knee osteoarthritis is "low", and if it is classified as "equivalent to KL = 0 to 2", the progression risk information is classified as "equivalent to healthy". If the risk of progression of knee osteoarthritis is classified as "moderate", and if the risk of progression of knee osteoarthritis is classified as "equivalent to KL = 3 to 4", the risk of progression of knee osteoarthritis is classified as "high". "Walking stability information" is information that is an index of stability during walking and is calculated according to the degree of knee osteoarthritis determined based on the KAM value information. Examples include "good" (corresponding to a healthy person) indicating high gait stability, "progress observation" (corresponding to KL = 0 to 2) indicating a tendency for deterioration of gait stability, and gait stability. Three classifications of "unstable" (corresponding to KL=3 to 4) in which the walking stability is worsening are stored as walking stability information. "Fall risk information" is information that is calculated according to the degree of knee osteoarthritis determined based on KAM value information and serves as an index of stability when walking, similar to gait stability information. . As an example, "low" is considered to be equivalent to a normal person with respect to knee osteoarthritis, "medium" is considered to be equivalent to KL = 0 to 2 with regard to knee osteoarthritis, and If KL is equivalent to 3 to 4, "high" is stored as fall risk information. Furthermore, since gait stability information and fall risk information are also evaluations obtained based on the degree of risk of knee osteoarthritis, it is natural to use these information as part of the information that indicates the degree of knee risk. Is possible.

Note that the estimated KL information, progression risk information, gait stability information, and fall risk information in Figure 5B are each classified into three categories, but they can be classified more finely or more roughly. is also possible.

FIG. 5C is a diagram illustrating an example of a KAM threshold table stored in the processing device 100 according to the embodiment of the present disclosure. According to FIG. 5C, gender information, KAM value numerical range information, and estimated KL information are stored in association with each other in the KAM value threshold table. "Gender information" is information classified into male and female. Although these are typically classified based on biological sex, they may also be classified based on gender information input via the operation input interface 113 by a user, operator, or the like. “KAM value numerical range information” is information that determines the relationship between the calculated KAM value and each threshold value. If the gender is male, M1 is set as the first threshold and M2 is set as the second threshold. If the gender is female, F1 is set as the first threshold and F2 is set as the second threshold. "Estimated KL information" is information for estimating the corresponding KL according to the relationship between the calculated KAM value and each threshold value. That is, according to the KAM value threshold table in FIG. 5C, if the KAM value calculated when a male user walks, for example, is smaller than the first threshold (M1), the estimated KL is "equivalent to a healthy person". ”. Further, for example, if the KAM value calculated when a female user performs walking exercise is equal to or higher than the second threshold value (F2), the estimated KL is determined to be equivalent to KL=3 to 4.

In addition, the inventors found that the KAM value (i.e., external knee varus moment value) and KL (i.e., Kellgren-Lawrence classification), which indicates the degree of risk of knee osteoarthritis, depend on male and female gender. I found it. In other words, men have lower KAM values than women even if they have the same low risk. Therefore, in order to accurately determine the degree of risk of knee osteoarthritis, it is important to make decisions according to gender. Therefore, in FIG. 5C, different threshold values are set depending on gender.

Furthermore, in the example of FIG. 5C, the first threshold value and the second threshold value are used to classify into three categories: healthy equivalent, KL=0 to 2 equivalent, and KL=3 to 4 equivalent. However, the invention is not limited to this, and may be divided into categories such as healthy equivalent, KL=0 equivalent, KL=1 equivalent, KL=2 equivalent, KL=3 equivalent, and KL=4 equivalent, or more or less classification. It may also be classified.

4. Calculation of KAM value using learned estimation model FIG. 5A explains how the acceleration information detected by the detection device 200 is stored, and FIG. 5B explains how the KAM value calculated based on the obtained acceleration information is stored. did. FIG. 6 is a diagram illustrating a processing flow related to generation of a trained estimation model according to an embodiment of the present disclosure. Specifically, FIG. 6 shows a process for generating a trained estimation model used to estimate a KAM value from acceleration information detected by the detection device 200. The processing flow may be executed by the processor 111 of the processing device 100, or may be executed by a processor of another processing device.

According to FIG. 6, a step of acquiring an output value from the detection device 200 is executed (S111). The output value detected by the detection device 200 attached below the knee of a healthy person or a user with knee osteoarthritis is used as the output value. Note that, as the output value, it is also possible to use horizontal acceleration detected at a predetermined period in a predetermined period, or it is also possible to use other output values. Examples of other output values include, in addition to horizontal acceleration, vertical acceleration, depth acceleration, angular velocity with respect to the horizontal axis, angular velocity with respect to the vertical axis, and angular velocity with respect to the depth axis. In total, there are output values for six axes. Then, a predetermined number of such output values are acquired for deep learning.

Next, the output value acquired in S111 and the KAM value measured in advance by another method as the correct answer label are input as learning data to a convolutional neural network (CNN) for generating an estimation model. , KAM values are output in a learning device including the convolutional neural network (S112). Then, by repeating the learning in S112, a learned estimation model for finally calculating the KAM value is generated (S113). Note that the KAM value as the correct label is measured, for example, by a method using motion capture.

Here, when detecting an output value in the detection device 200 for the obtained trained estimation model, a KAM value is calculated separately using another method, and this output value and KAM value are used. Verification may also be performed (S114). Then, it is possible to adjust the parameter values used in the convolutional neural network based on the feedback of the results.

Note that in this disclosure, a learning method using a convolutional neural network has been exemplified, but the learning method is not limited to this, and other deep learning methods may be used, and other machine learning methods may also be used. For example, by preparing a plurality of combinations of output values and KAM values for which the correspondence between the output values from the detection device 200 and KAM values has been confirmed in advance, and using these as training data, a learned estimation model is generated. is also possible.

The learned estimation model obtained in this way is stored in the memory 112, processed by the processor 111, and used to calculate the KAM value. Note that when the cloud server device functions as the processing device 100, the learned estimation model is stored in the memory of the cloud server device.

5. Processing flow executed in the processing device 100 [Processing related to mode selection]
FIG. 7 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. Specifically, FIG. 7 shows a processing flow executed by the processor 111 at a predetermined cycle after the program according to the embodiment of the present disclosure is started in the processing device 100.

First, when the processor 111 receives an interrupt signal indicating that the operation input interface 113 has accepted an input instruction to start the program, the processor 111 displays the top screen on the display 114 (S311). Although the top screen is not particularly illustrated, it includes a measurement mode for measuring the output value of the user in the detection device 200, and an icon for transitioning to a result display mode for displaying the measurement results. Thereafter, the processor 111 determines whether a mode has been selected based on an interrupt signal indicating that an operation input for the icon by the operator has been received from the operation input interface 113 (S312). If it is determined that the mode is not selected, the top screen remains displayed and the processing flow ends.

On the other hand, if it is determined that a mode has been selected, the processor 111 determines whether a measurement mode is to be selected based on the coordinates input by the operator (S313). If it is determined that the measurement mode is selected, the processor 111 controls the display 114 to display a measurement screen, and ends the processing flow. On the other hand, if it is determined that it is not the measurement mode, the processor 111 controls the display 114 to display a result screen, and ends the processing flow.

Although not particularly shown, the measurement screen includes areas for inputting or selecting user gender information, user name information, user ID information, etc., and areas for measurement. A start button icon etc. for starting is displayed.

[Processing related to starting measurement]
FIG. 8 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. Specifically, FIG. 8 shows a processing flow executed by the processor 111 at a predetermined period after the measurement mode is selected and the measurement screen is displayed in FIG. Although not particularly shown in FIG. 8, before the process shown in FIG. 8, the measurer or user inputs or selects user name information, user ID information, etc. on the measurement screen, and stores these input or selected information in the memory 112. Further, the detection device 200 is attached in advance to the lower part of the knee of one of the legs of the user 10 by means of the auxiliary tool 400, and all preparations for the start of exercise (for example, walking) by the user 10 are completed. Furthermore, in the description of the processing flow from FIG. 8 onwards, for convenience of explanation, a case will be described in which the detection device 200 is attached only to the knee of one of the legs.

According to FIG. 8, the processor 111 determines whether or not the operation input interface 113 receives an operation input from the operator on the measurement button icon (S411). If it is determined that the operation input to the start button icon has been accepted, the processor 111 controls the detection device 200 to transmit a measurement start instruction signal to instruct the detection device 200 to start measurement via the communication interface 115. (S412). After that, the processor 111 controls the display 114 to display a measurement standby screen (S413). Although not particularly shown, an end button icon for ending the measurement is displayed on the measurement standby screen.

Here, processing on the detection device 200 side will be explained. When the processor 211 of the detection device 200 receives a measurement start signal via the communication interface 214 after being worn by the user, it drives the sensor 212 to start detecting acceleration at a predetermined period. Then, the processor 211 stores the detected acceleration as an output value in the memory 213 in association with the detected time. Then, the processor 211 executes this process until it receives a measurement end instruction signal from the processing device 100.

The case has been described in which the measurement start instruction signal is transmitted by detecting the press of the start button displayed on the display 114. However, the present invention is not limited to this, and the transmission may be performed by the processor 111 detecting a press of a start button provided as a physical key on the operation input interface 113. Further, for example, when a press operation on the power switch 216 of the detection device 200 is accepted, the detection device 200 may transmit a measurement start signal to the processing device 100, and then the processor 111 may control the display of a measurement standby screen. .

With the above steps, the processing flow related to the start of measurement is completed.

[Processing performed during measurement standby]
FIG. 9 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. Specifically, FIG. 9 shows a processing flow executed by the processor 111 at a predetermined period after the measurement standby screen is displayed in FIG.

According to FIG. 9, the processor 111 receives an operation input from the operator on the end button icon through the operation input interface 113, and determines whether the measurement has ended (S511). If it is determined that the measurement has ended, the processor 111 controls the detection device 200 to transmit a measurement end instruction signal to instruct the detection device 200 to end the measurement via the communication interface 115 (S512).

Here, in the detection device 200 that has received the measurement end instruction signal from the processing device 100 via the communication interface 214, the processor 211 controls the sensor 212 to end acceleration detection. Then, the processor 211 controls the output value and time information stored in the memory 213 until the end to be transmitted to the processing device 100 via the communication interface 214.

In the processing device 100, it is determined whether the processor 111 has received the output value and time information from the detection device 200 via the communication interface 115 (S513). If it is determined that the received output value (acceleration information) has been received, the processor 111 associates the received output value (acceleration information) with the time information in the acceleration table of the memory 112 in association with the user ID information input or selected on the measurement screen. and stored (S514). Next, the processor 111 performs various estimation processes based on the output values stored in the memory 112 (S515). Details of this processing will be described later. Then, the processor 111 stores various information obtained during the estimation process in the user table of the memory 112 in association with the user ID information (S516).

Note that the case has been described in which the measurement end instruction signal is transmitted by detecting the press of the end button displayed on the display 114. However, the present invention is not limited thereto, and the transmission may be performed by the processor 111 detecting a press of an end button provided as a physical key on the operation input interface 113. Further, for example, when a pressing operation on the power switch 216 of the detection device 200 is accepted, the detection device 200 may end the measurement and transmit the output value etc. to the processing device 100.

With the above steps, the processing flow performed during measurement standby ends.

[Calculation and judgment processing]
FIG. 10 is a diagram showing a processing flow executed in the processing device 100 according to the embodiment of the present disclosure. Specifically, FIG. 10 shows a specific method for calculating the KAM value in FIG. 5B from the acceleration information detected by the detection device 200 in FIG. 5A, and determining the degree of knee risk based on the calculated KAM value. This shows the typical processing flow.

According to FIG. 10, the processor 111 first reads out the output value associated with the user ID information input or selected on the measurement screen from the acceleration table stored in the memory 112 (S211). Note that although only acceleration information is stored in FIG. 5A, six-axis output values including angular velocity information may be stored as described above. Therefore, as the output values read in S211, it is possible to read not only acceleration information but also six-axis output values including angular velocity information. Next, the processor 111 applies the read output value to the learned estimation model for estimating the KAM value generated in FIG. 6 (S212). The processor 111 then calculates the KAM value using the estimated model (S213). The processor 111 stores the calculated KAM value as KAM value information in the user table in association with the user ID information.

Although not described in detail in this disclosure, the KAM value and output values such as acceleration have a correlation with each other. Therefore, a conversion table between the output value and the KAM value is prepared in advance based on the correlation, and the conversion table is stored in the memory 112. After that, when an output value is obtained in actual measurement, it is also possible to calculate the KAM value by referring to the conversion table. Further, in the example of FIG. 10, the KAM value was calculated based on the learned estimation model and the output value from the detection device 200. However, the KAM value is not limited to this, and it is also possible to calculate the KAM value by other methods such as a method using motion capture.

Next, when the KAM value is calculated, the processor 111 refers to the KAM threshold table based on the calculated KAM value. Then, the processor 111 determines which of the estimated KL information classified by each threshold value the calculated KAM value corresponds to. Furthermore, based on the estimated KL information, the risk of progression of knee osteoarthritis is evaluated as shown in FIG. 5B (S214). For example, the processor 111 refers to the user table in FIG. 5B and reads gender information associated with user ID information. The processor 111 refers to the KAM threshold table based on the read gender information and determines whether to use the threshold associated with men or the threshold associated with women. Next, the processor 111 compares the calculated KAM value with each threshold value depending on the gender, and determines which of the estimated KL information corresponds to the corresponding one. For example, if the KAM value calculated by the walking exercise of a male user is smaller than the first threshold value (M1), the processor 111 determines that the estimated KL is "equivalent to a healthy person" and The risk of progression of arthropathy is evaluated as "low." Furthermore, if the KAM value calculated by a female user performing walking exercise is equal to or higher than the second threshold (F2), the processor 111 evaluates the risk of progression of arthropathy as "low". The patient's KL is determined to be equivalent to KL=3 to 4, and the risk of progression of knee osteoarthritis is evaluated as "high." The obtained results are stored in the user table in the memory 112 as knee risk level information.

Here, assuming that the KAM area value is used as the KAM value, the first threshold for determining whether the classification is "equivalent to healthy" or "equivalent to KL = 0 to 2" is set when the user is male. The first threshold value (KAM value)=15.0, more preferably the first threshold value (KAM value)=16.5 is used. Further, as the first threshold value, when the user is a woman, the first threshold value (KAM value)=8.5, more preferably the first threshold value (KAM value)=10.5 is used. Furthermore, the second threshold value that determines whether the classification is "equivalent to KL = 0 to 2" or "equivalent to KL = 3 to 4" is that if the user is male, the second threshold value (KAM value) = 17. .0, more preferably the second threshold value (KAM value)=18.0 is used. Further, as the second threshold, when the user is a woman, second threshold (KAM value)=11.0, more preferably second threshold (KAM value)=13.5 is used. In this way, by determining the degree of risk of knee osteoarthritis using a specific threshold value, a more accurate determination can be made.

Next, the processor 111 evaluates walking stability and fall risk based on the calculated KAM value (S215). Specifically, based on the estimated KL information determined in S214, the processor 111 determines that the walking stability is "good" and the risk of falling is "low" if it is determined that the person is "equivalent to being healthy." evaluate. Furthermore, if it is determined that "KL = 0 to 2" based on the estimated KL information determined in S214, the processor 111 determines the walking stability as "progress observation" and the fall risk as "medium". ” Furthermore, if it is determined that "KL = 3 to 4" based on the estimated KL information determined in S214, the processor 111 determines that the walking stability is "unstable" and the fall risk is "high." ” The processor 111 then stores the evaluated walking stability and fall risk in the user table of the memory 112 based on the user ID information.

[Processing related to displaying the result screen]
The results obtained by the process in FIG. 10 (for example, "degree of knee risk" including evaluation results of walking stability and fall risk) are output to the display 114. Specifically, the processor 111 first receives an operation input from an operator through the operation input interface 113, and selects user ID information associated with the user whose information is to be displayed. The processor 111 then refers to the user table and provides estimated KL information, knee osteoarthritis progression risk information, gait stability information, and fall risk information associated with the selected user ID information. are read out, respectively, and control is performed to output the result screen on the display 114 (S515). Note that here, as information indicating knee risk, information on the degree of knee osteoarthritis risk, gait stability information, and fall risk information is displayed, but it is also possible to display only one of them. .

FIG. 14 is a diagram showing an example of a screen output from the output interface 114 of the processing device 100. Specifically, FIG. 14 is a diagram illustrating an example of a result screen showing knee osteoarthritis progression risk information as knee risk degree information. According to FIG. 14, the calculated KAM values are shown in a bar graph for each measured knee (left and right) along with the date of this measurement, and the KAM values are for the knee osteoarthritis determined in FIG. 10. The progression risk information is shown in three stages: "high," "medium," and "low." In other words, by referring to this screen, the user can visually grasp what KAM value each of the left and right knees currently has and what degree of progression risk it has. .

Although not particularly shown in the user table, past history information regarding estimated KL information, knee osteoarthritis progression risk information, gait stability information, and fall risk information may also be stored. It is possible. In such a case, as shown in FIG. 14, in addition to the currently calculated degree of progression risk, it is possible to show the degree of past risk in chronological order. By doing so, it becomes possible to visually grasp the changes in the progress risk information from the past.

In the example of FIG. 14, the bar graph is displayed based on the KAM value, but this is based on the estimated KL information, knee osteoarthritis progression risk information, gait stability information, and fall risk information, or a combination thereof. Any information indicating the degree of knee risk may be displayed. Further, the display method may be to describe the obtained concrete values, classifications, etc. as text information, or may be expressed by other methods such as other graphs or abstract diagrams.

In this way, in this embodiment, it is possible to more easily calculate and output the condition of the knee and the risks associated with the knee. Specifically, by using the output value detected by the detection device 200, it is possible to more easily and accurately judge the degree of risk of the knee, such as information on the risk of progression of knee osteoarthritis. be.

6. Other Embodiments In the embodiments described above, a case has been described in which an acceleration sensor is used as the detection device 200 to detect acceleration during exercise. However, in place of the acceleration sensor or in combination with the acceleration sensor, any sensor capable of detecting the movement of the user 10, especially movement such as bending and straightening the knee, may be used, such as a gyro sensor, geomagnetic sensor, or telescopic sensor. Is possible.

7. Example Table 1 showing the correspondence between KAM values and KL classifications of knee osteoarthritis shows the relationship between KL classifications and KAM values for each user, with healthy subjects and patients with knee osteoarthritis as users. This is a table showing. Specifically, 26 healthy subjects (including 4 males and 22 females), 56 knee osteoarthritis patients with KL = 0 to 2 (including 9 males and 47 females), and KL = 3 to 2. KAM values (KAM area values in this example) were calculated for 34 patients with knee osteoarthritis equivalent to 4 (including 7 men and 27 women).

Regarding the KL classification, doctors take X-ray images of each patient diagnosed with knee osteoarthritis, observe the degree of wear of the articular cartilage and the degree of ossification around the knee, and Each classification of KL=0 to 4 was assigned based on the degree. In addition, regarding the KAM value, we used the motion capture system "Oqus" manufactured by Qualisys (using 8 cameras at 200 frames/second) and the floor reaction force meter "AM6110" (2000 Hz) manufactured by Bertec. It was measured by detecting the movement of reflective markers placed on the user's standard 46 bony markers. Then, from the movement of the obtained reflective marker, the kinematic behavior of the knee is calculated using "Visual 3D" manufactured by C-motion Company, a KAM value curve is obtained, and the KAM value (KAM area value) is calculated. was calculated.

Table 1 shows that all 116 users were divided into groups of healthy people, patients with knee osteoarthritis with KL = 0 to 2, and patients with knee osteoarthritis with KL = 3 to 4, and the methods described above were used. The results of calculating the average value of the calculated KAM values are shown. First, focusing on the overall population (men and women), the average value for healthy subjects was 9.1, whereas the average value for patients with knee osteoarthritis with KL = 0 to 2 and those with knee osteoarthritis with KL = 3 to 4. It was confirmed that the average KAM value of arthropathy patients increased to 10.8 and 13.6 as the risk of knee osteoarthritis increased. Similarly, focusing on women, the average value for healthy subjects was 8.1, whereas for patients with knee osteoarthritis with KL = 0 to 2 and patients with knee osteoarthritis with KL = 3 to 4. It was confirmed that the average KAM value increased to 9.6 and 12.4 as the degree of risk of knee osteoarthritis increased. In addition, focusing on men, the average value for healthy subjects was 14.7, whereas the average value for knee osteoarthritis patients with KL = 0 to 2 and knee osteoarthritis patients with KL = 3 to 4. It was confirmed that as the degree of risk of knee osteoarthritis increases, the KAM value also increases to 16.6 and 18.2. Based on these results, there is a correlation between KL and KAM value (KAM area value), which indicates the degree of knee osteoarthritis, and the higher the degree of knee osteoarthritis in a patient, the higher the KAM value (KAM area value). ), and conversely, it was confirmed that when the KAM value exceeds a certain standard, it is appropriate to assign a high KL classification.

In addition, in all KL classifications, there was a clear difference in KAM values for each KL classification between men and women. This confirmed that determining the degree of risk of knee osteoarthritis based on the KAM value requires judgment according to male and female gender.

From the above, it was confirmed that the KL classification of knee osteoarthritis can be estimated based on the calculated KAM value.

8. Example showing the correlation between the output value and the KAM value FIG. 11 is a diagram showing the correlation between the actually measured acceleration peak width at the beginning of the stance phase and the KAM value. Specifically, FIG. 11 shows a total of 44 output values actually measured by the detection device 200 attached to the lower part of the knee of both legs of 22 subjects with knee osteoarthritis. A KAM value curve was obtained by a method similar to that described in "Example showing correspondence between KAM values and KL classification of knee osteoarthritis", and KAM peak values were calculated. On the other hand, acceleration was measured by attaching a small wireless multifunctional sensor "TSND151" manufactured by ATR-Promotions to the lower part of the knee of both legs of the subject, obtaining the output value (acceleration) during walking, and using the obtained output value to determine the stance leg. The initial peak width was calculated. The Pearson correlation coefficient (p) was calculated based on each obtained value, and when p<0.05, it was evaluated that there was a correlation between the two. According to FIG. 11, in the measurement, a correlation coefficient p<0.001 between both values is shown, confirming that there is a high correlation between the KAM value and the peak width of acceleration. In other words, it was confirmed that the KAM value can be estimated by detecting the peak width of the acceleration at the beginning of the stance phase using the detection device 200.

In addition, among the 44 output values actually measured by the detection device 200 attached to the lower part of the knee of both legs of the 22 subjects with knee osteoarthritis used in the example of FIG. 11 above, A case will be described in which output values for a total of six axes are used: horizontal acceleration, vertical acceleration, depth acceleration, angular velocity with respect to the horizontal axis, angular velocity with respect to the vertical axis, and angular velocity with respect to the depth axis. First, a trained estimation model was generated using the output values of 18 randomly selected subjects out of the output values of 22 subjects and the KAM value (KAM area value) prepared as the correct answer label. The KAM values (KAM area values) of the correct labels were calculated from the KAM value curves obtained by the same 18 people using the same method as described above using motion capture. Next, in order to verify the generated estimation model, the output values of the remaining four people were applied to the generated estimation model to calculate the KAM value (KAM area value). Then, the KAM value (KAM area value) estimated by the estimation model was verified by comparing it with the KAM value (KAM area value) obtained by the same method described above using motion capture.

Figure 12 shows the correlation between the KAM value (KAM area value) estimated by applying the output values of four people to the generated estimation model and the KAM value (KAM area value) calculated by motion capture. FIG. Specifically, in FIG. 12, the horizontal axis represents the number of repeated learnings (0 to 239 times) using the estimation model, and the vertical axis represents the KAM value (KAM area value) and motion obtained by applying the estimation model. A graph in which a correlation coefficient with a KAM value (KAM area value) obtained using capture is assigned is shown. According to FIG. 12, a correlation coefficient of 0.4 or more, which is generally evaluated as "correlated", was stably obtained in the 47th and subsequent steps. In particular, an extremely high correlation coefficient of 0.934 at maximum was obtained in 239 learning steps. This means that the KAM value (KAM area value) calculated using the estimation model can be used to determine the degree of risk of knee osteoarthritis, etc., similar to the KAM value (KAM area value) obtained by motion capture. It has been shown that it can be fully used for this purpose.

Figure 13 shows the correlation between the KAM value (KAM peak value) estimated by applying the output values of four people to the generated estimation model and the KAM value (KAM peak value) calculated by motion capture. FIG. Specifically, in FIG. 13, the horizontal axis represents the number of repetitions of learning in the estimation model (0 to 239 times), and the vertical axis represents the KAM value (KAM peak value) obtained by applying the estimation model and the motion. A graph in which a correlation coefficient with a KAM value (KAM peak value) obtained using capture is assigned is shown. According to FIG. 13, a correlation coefficient of 0.4 or more, which is generally evaluated as "correlated", was stably obtained in the 29th and subsequent steps. In particular, an extremely high correlation coefficient of 0.633 at maximum was obtained in 218 learning steps. This means that the KAM value (KAM peak value) estimated using the estimation model can be used sufficiently for the evaluation of knee osteoarthritis, etc., in the same way as the KAM value (KAM peak value) obtained by motion capture. showed that it is possible. Although not described in detail, the estimation model was generated in the same manner as the KAM value (KAM area value) except that the KAM peak value was used as the KAM value.

It is also possible to configure a system by appropriately combining the elements described in each embodiment or replacing them.

The processes and procedures described herein can be implemented not only by what is explicitly described in the embodiments, but also by software, hardware, or a combination thereof. Specifically, the processes and procedures described in this specification can be realized by implementing logic corresponding to the processes in a medium such as an integrated circuit, volatile memory, nonvolatile memory, magnetic disk, or optical storage. be done. Further, the processes and procedures described in this specification can be implemented as a computer program and executed by various computers including processing devices and server devices.

Even if processes and procedures described herein are described as being performed by a single device, software, component, or module, such processes or procedures may be performed by multiple devices, software, components, or modules. It may be implemented by multiple components and/or multiple modules. Further, even if it is explained that the various information described in this specification is stored in a single memory or storage unit, such information may be stored in multiple memories or storage units provided in a single device. The information may be stored in a distributed manner in a plurality of memories distributed and arranged in a plurality of devices. Additionally, the software and hardware elements described herein may be implemented by integrating them into fewer components or decomposing them into more components.

1 processing system 100 processing device 200 detection device 300 server device 400 auxiliary tool

Claims (10)

A processing device comprising at least one processor,
the at least one processor,
Receiving the acceleration detected by wire or wirelessly via a communication interface from a sensor attached to the knee of the user's leg or around the knee and detecting acceleration during exercise of the user,
calculating an index indicating the state of the knee based on the received acceleration;
A processing device configured to perform processing for outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold. The processing device according to claim 1, wherein the threshold value is a different value depending on the gender of the user. The processing device according to claim 1 or 2, wherein when the threshold value is equal to or greater than a first threshold value, it is determined that the degree of risk to the knee is high. The index is a value indicating a knee varus moment,
The first threshold is 15.0 when the user is male, and 8.5 when the user is female.
The processing device according to claim 3. The threshold includes, in addition to the first threshold, a second threshold different from the first threshold,
When the index is greater than or equal to the first threshold and less than the second threshold, the knee Judging that the level of risk is high,
The processing device according to claim 3 or 4. The processing device according to any one of claims 1 to 5, wherein the knee risk is a progression risk for knee osteoarthritis. The index is calculated by a learned estimation model obtained by learning using the acceleration detected by the sensor and diagnostic information regarding knee osteoarthritis prepared in advance as a correct label. 6. The processing device according to any one of items 6 to 6. In a processing device comprising at least one processor, the at least one processor
Receiving the acceleration detected by wire or wirelessly via a communication interface from a sensor attached to the knee of the user's leg or around the knee and detecting acceleration during exercise of the user,
calculating an index indicating the state of the knee based on the received acceleration;
outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold;
A program that makes it work. A method carried out by the at least one processor in a processing device comprising at least one processor, the method comprising:
receiving the detected acceleration via a communication interface, by wire or wirelessly, from a sensor attached to or around the knee of the user's leg for detecting acceleration during exercise of the user;
calculating an index indicating the condition of the knee based on the received acceleration;
outputting information regarding the degree of risk of the knee based on the calculated index and a predetermined threshold;
method including. A processing device according to any one of claims 1 to 7,
A sensor connected to the processing device by wire or wirelessly, attached to the knee of the user's leg or around the knee, and configured to detect acceleration during exercise of the user;
processing systems including;

Images