Patient monitoring system, patient monitoring device, patient monitoring method, and program
The system addresses the invasive nature of sensor failures in artificial hip joint monitoring by using a body-surface motion sensor to calculate and notify dislocation risk, reducing patient burden and dislocation risk without surgery.
Patent Information
- Application Number
- PCT/JP2026/017073
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-09
- Filing Date
- 2026-04-30
- Publication Date
- 2026-11-12
AI Technical Summary
Existing patient monitoring systems for artificial hip joints require invasive surgery to access the sensor when it fails, increasing patient burden.
A patient monitoring system with a motion sensor attached to the body surface that calculates a joint limb position index using detected physical quantities, estimates dislocation risk, and provides notifications without requiring surgery.
Reduces patient burden by allowing non-invasive monitoring of joint dislocation risk and providing timely feedback, thereby decreasing the risk of joint dislocation and the need for revision surgeries.
Smart Images

Figure JP2026017073_12112026_PF_FP_ABST
Abstract
Description
Patient monitoring system, patient monitoring device, patient monitoring method, and program
[0001] The present invention relates to a patient monitoring system, a patient monitoring device, a patient monitoring method, and a program.
[0002] A technique has been proposed for monitoring the state of a replacement artificial hip joint based on data transmitted from a sensor, using the replacement artificial hip joint that includes an artificial femoral stem, an artificial femoral head coupled to the femoral stem, an artificial acetabular assembly coupled to the femoral head, and a sensor provided in at least one of the femoral stem, the femoral head, and the acetabular assembly (see, for example, Patent Document 1).
[0003] Japanese Unexamined Patent Application Publication (Translation of PCT Application) No. 2016-516489
[0004] However, in the monitoring technique described in Patent Document 1, when a sensor failure occurs, surgery is required to access the replacement artificial hip joint implanted in the patient's body, which may increase the burden on the patient.
[0005] The present invention has been made in view of the above circumstances, and an object thereof is to provide a patient monitoring system, a patient monitoring device, a patient monitoring method, and a program capable of appropriately monitoring the state of an artificial joint while reducing the burden on a patient in whom the artificial joint is implanted.
[0006] To achieve the above object, a patient monitoring system according to the present invention includes: a motion sensor attached to the body surface of at least one of two body parts arranged via a joint portion of a patient; a joint limb position index calculation unit that calculates a joint limb position index reflecting the relative positional relationship of one bone with respect to any one of a plurality of bones connected to each other via the joint present in the joint portion, based on a physical quantity detected by the motion sensor; a dislocation risk estimation unit that estimates a dislocation risk of the joint portion based on the joint limb position index; and a dislocation risk notification unit that notifies a user of the estimated dislocation risk.
[0007] According to the present invention, since the motion sensor is attached to the surface of the patient's body, surgery is not required even if the motion sensor malfunctions, thereby reducing the burden on the patient. Furthermore, the joint position index calculation unit calculates a joint position index that reflects the relative positional relationship of one of the multiple bones connected to each other via joints in the patient's joint area, based on the physical quantities detected by the motion sensor. Then, the dislocation risk estimation unit determines the dislocation risk of the joint area based on the calculated joint position index, and the dislocation risk notification unit notifies the user of the estimated dislocation risk. As a result, the condition of the patient's joints can be appropriately monitored, thereby reducing the patient's risk of dislocation.
[0008] This is a schematic diagram of a patient monitoring system according to an embodiment of the present invention. This is a block diagram showing the hardware configuration of the patient monitoring system according to the embodiment. This is a block diagram showing the functional configuration of the sensor unit and terminal device according to the embodiment. This is a diagram illustrating the dislocation risk estimation model according to the embodiment. This is a block diagram showing the functional configuration of the management server according to the embodiment. This is a flowchart showing an example of the patient monitoring process flow executed by the terminal device according to the embodiment. This is a flowchart showing an example of the patient monitoring information management process flow executed by the management server according to the embodiment.
[0009] Hereinafter, a patient monitoring system according to one embodiment of the present invention will be described in detail with reference to the drawings. The patient monitoring system according to this embodiment includes: a motion sensor attached to the surface of at least one of two body parts aligned via a joint of the patient; a joint position index calculation unit that calculates a joint position index that reflects the joint position, which is the relative positional relationship of one of a plurality of bones connected to each other via a joint present in the joint, based on a physical quantity detected by the motion sensor; a dislocation risk estimation unit that determines the risk of dislocation of the joint based on the joint position index; and a risk notification unit that notifies the user of the determination result.
[0010] The patient monitoring system according to this embodiment aims to prevent dislocation of the artificial joint in patients who have undergone artificial joint replacement surgery, in which a portion of a joint is replaced with an artificial joint, by monitoring the risk of dislocation of the artificial joint and notifying the patient of such risk. As shown in Figure 1, the patient monitoring system according to this embodiment comprises a sensor unit 1 attached to the body surface of the patient HU1, a terminal device 2 that communicates wirelessly with the sensor unit 1, and a management server 3 that can communicate with the terminal device 2. Although Figure 1 shows an example where only one sensor unit 1 is attached, the sensor unit 1 may be attached to multiple locations on the body surface of the patient HU1 (for example, two locations near the greater trochanter and near the acromion). The terminal device 2 has the function of communicating wirelessly with the sensor unit 1 and the function of communicating wirelessly with an access point 8 connected to a network NW. The network NW is, for example, a wired LAN (Local Area Network), and a switch 9 that controls communication in the network NW is connected to it. The management server 3 is connected to the network NW. The terminal device 2 and the management server 3 communicate via the access point 8 and the network NW. In this embodiment, an example is described in which the terminal device 2 communicates with the management server 3 via the access point 8, but it is not limited to this, and for example, the terminal device 2 and the management server 3 may communicate via a public telephone network.
[0011] The sensor unit 1 is attached to the body surface of at least one of two adjacent body parts (in the example shown in Figure 1, near the greater trochanter) via a joint part Po1 corresponding to, for example, the hip joint of the patient HU1. As shown in Figure 2, the sensor unit 1 comprises a motion sensor 107, a processing circuit 101, and a wireless module 106. The motion sensor 107 has, for example, a three-axis gyro sensor and a three-axis accelerometer, and detects the angular velocity of the sensor unit 1 and the acceleration of the sensor unit 1 in the vertical direction and in directions perpendicular to the vertical direction. Specifically, the motion sensor 107 outputs a total of six types of signals: three types of signals reflecting the angular velocity in three types of Euler angle directions centered on each of the three coordinate axes in the sensor coordinate system of the sensor unit 1 itself, namely the roll angle direction, pitch angle direction, and yaw angle direction, and three types of signals reflecting the acceleration in the vertical direction and in two directions perpendicular to the vertical direction and mutually perpendicular. The wireless module 106 has the function of wirelessly communicating with the terminal device 2 using a communication protocol that conforms to short-range wireless communication standards such as BR / EDR (Bluetooth Basic Rate / Enhanced Data Rate) and BLE (Bluetooth Low Energy).
[0012] The processing circuit 101 is composed of, for example, a dedicated integrated circuit. When power is supplied to the sensor unit 1, it periodically converts the six types of signals input from the motion sensor 107 into three types of angular velocity information indicating the angular velocity in the rotational direction around each of the three axes in the sensor coordinate system, and three types of acceleration information indicating the acceleration in the three axial directions, including the vertical direction. The processing circuit 101 then generates a physical quantity notification frame containing the three types of angular velocity information and the two types of acceleration information, and wirelessly transmits the generated physical quantity notification frame to the terminal device 2 via the wireless module 106.
[0013] Terminal device 2 is, for example, a smartphone and comprises a CPU (Central Processing Unit) 201, a main memory unit 202, an auxiliary memory unit 203, a display unit 241, a speaker 242, a vibration generator 243, an audio signal generation unit 244, an input unit 205, a first wireless module 261, a second wireless module 262, a timing unit 208, and a bus 209 connecting the units, and functions as a patient monitoring device that monitors the patient's risk of dislocation. The main memory unit 202 is composed of volatile memory such as RAM (Random Access Memory) and is used as a working area for the CPU 201. The auxiliary memory unit 203 is composed of non-volatile memory such as ROM (Read Only Memory), or non-volatile memory such as SSD (Solid State Drive) or HDD (Hard Disk Drive), and stores programs for realizing various functions of terminal device 2. The display unit 241 is a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display.
[0014] When the audio signal generation unit 244 receives a dislocation risk notification command signal from the CPU 101 (described later), it generates an audio signal indicating a warning sound and outputs it to the speaker 242, thereby causing the speaker 242 to output a warning sound. When the vibration generator 243 receives a dislocation risk notification command signal from the CPU 101 (described later), it vibrates the entire terminal device 2 in response. The input unit 205 is, for example, a transparent touch pad and is placed on top of the display unit 241. The first wireless module 261 has the function of wirelessly communicating with the terminal device 2 using a communication protocol compliant with short-range wireless communication standards such as BR / EDR and BLE. The second wireless module 262 has the function of wirelessly communicating with the access point 8 using a communication protocol compliant with a wireless communication standard such as IEEE 802.11. The timing unit 208 has, for example, a software timer and timing the time when the joint position index (described later) is calculated.
[0015] In terminal device 2, the CPU 201 reads the program stored in auxiliary storage unit 203 into main storage unit 202 and executes it, so that, as shown in Figure 3, it functions as a physical quantity acquisition unit 211, a reference setting unit 217, a joint position index calculation unit 212, a dislocation risk estimation unit 213, a dislocation risk notification unit 214, a display control unit 216, a joint position index history information notification unit 218, a model acquisition unit 219, and a high dislocation risk period notification acquisition unit 220. Furthermore, the auxiliary storage unit 203 shown in Figure 2 has a reference information storage unit 235, a joint position index storage unit 232, a model storage unit 233, and an estimation result storage unit 234, as shown in Figure 3. In addition, the main storage unit 202 shown in Figure 2 has a physical quantity buffer 221 that temporarily stores three types of angular velocity information transmitted from sensor unit 1, as shown in Figure 3. The reference information storage unit 235 stores reference attitude information, which indicates the roll angle, pitch angle, and yaw angle of the sensor unit 1, representing the reference attitude of the sensor unit 1 in an absolute coordinate system with respect to the gravitational acceleration axis.
[0016] The joint position index memory unit 232 stores joint position index information that reflects the posture of the sensor unit 1, associating it with time information that represents the time when the joint position index was calculated. This joint position index reflects the joint position, which is the relative positional relationship between one of several bones connected to each other via a joint in the joint area and other bones. If there are multiple sensor units 1, the joint position index memory unit 232 stores joint position index information that reflects the posture of each of the multiple sensor units 1, associating it with time information that represents the time when the joint position index was calculated. Here, the joint position index is information that represents a three-dimensional vector showing the relative roll angle, pitch angle, and yaw angle in the absolute coordinate system at each time with respect to the aforementioned reference posture of the sensor unit 1. If the sensor unit 1 is attached near the patient's hip joint, this joint position index reflects at least one movement selected from the patient's hip flexion, hip adduction, and hip internal rotation movements. Furthermore, the joint position index storage unit 232 stores joint position index information, which represents the joint position index calculated from the angular velocity and acceleration detected within each second unit period, in association with time information indicating a representative point in time within the second unit period, when the predetermined first unit period is divided into multiple second unit periods shorter than the first unit period, and these multiple second unit periods are used as the period for calculating the joint position index. Here, the length of the first unit period is set to, for example, 24 hours, and the second unit period is set to, for example, 1 second to 10 seconds. The representative point in time is set to, for example, the start time or period of the second unit period. Moreover, the joint position index storage unit 232 has a storage capacity that can store joint position indexes for the first unit period, and for example, after the program for performing patient monitoring processing is started in the terminal device 2, once the aforementioned first unit period has elapsed, the previously stored joint position index information and time information are sequentially deleted each time new joint position index information and time information are stored thereafter.
[0017] The model storage unit 233 stores dislocation risk estimation model information, which represents a dislocation risk estimation model that calculates the patient's dislocation risk from the joint position index information. This dislocation risk estimation model is, for example, a trained feedforward neural network, and the model storage unit 233 stores information indicating the structure of the neural network and information indicating the weight coefficients in that neural network. Here, the information indicating the structure of the neural network includes information indicating the number of nodes, the number of layers, the weight coefficients corresponding to each node, and the activation function. Nonlinear functions such as sigmoid functions, ramp functions, and step functions are used as activation functions. Furthermore, as shown in Figure 4, when a three-dimensional joint position index vector indicating the roll angle, pitch angle, and yaw angle relative to the reference posture of the sensor unit 1, as indicated by the joint position index information, is input to the dislocation risk estimation model, it outputs an expectation value vector whose elements are the expected probabilities of each of a set number of pre-set dislocation risk levels. In Figure 4, the dislocation risk level is such that, for example, "Level 0" has the lowest risk of dislocation, and the risk of dislocation increases as the level number increases, such as "Level 1," "Level 2," and so on.
[0018] Returning to Figure 3, the estimated result storage unit 234 stores the dislocation risk level information, which indicates the dislocation risk level determined by the dislocation risk estimation unit 213, in association with time information indicating the time when the corresponding joint position index was calculated.
[0019] The physical quantity acquisition unit 211 acquires multiple physical quantity notification frames, which are periodically transmitted from the sensor unit 1, via the first wireless module 261, when the user performs a reference setting operation on the input unit 205 as described below, or when a predetermined physical quantity acquisition period arrives after the user performs a dislocation risk determination start operation on the input unit 205 as described below. The physical quantity acquisition unit 211 then extracts three types of angular velocity information and three types of acceleration information contained in each of the acquired physical quantity notification frames, and stores these extracted pieces of information in the physical quantity buffer 221 in chronological order, associating them with each other. Here, the physical quantity acquisition unit 211 repeatedly acquires the angular velocity and acceleration detected by the motion sensor 107 of the sensor unit 1 at a predetermined sampling period. The sampling period is set to, for example, 10 msec to 100 msec.
[0020] When the reference setting unit 217 receives a reference setting operation performed by the user on the input unit 205, it then calculates the roll angle, pitch angle, and yaw angle in the absolute coordinate system when the user wearing the sensor unit 1 is in a pre-specified reference posture, based on the angular velocity indicated by each of the multiple angular velocity information stored in the physical quantity buffer 221 and the acceleration indicated by the corresponding acceleration information. Here, the reference posture can be, for example, an upright posture. The reference setting unit 217 then stores the calculated reference information indicating the roll angle, pitch angle, and yaw angle in the reference information storage unit 235.
[0021] After the joint position index calculation unit 212 receives the dislocation risk determination start operation performed by the user on the input unit 205 as described below, and when the predetermined physical quantity acquisition time arrives, it then calculates the roll angle, pitch angle, and yaw angle of the sensor unit 1 in the absolute coordinate system based on the angular velocity indicated by each of the multiple angular velocity information stored in the physical quantity buffer 221 and the acceleration indicated by the corresponding acceleration information. Next, the joint position index calculation unit 212 calculates the roll angle, pitch angle, and yaw angle relative to the reference posture obtained by subtracting the calculated roll angle, pitch angle, and yaw angle of the sensor unit 1 from the calculated roll angle, pitch angle, and yaw angle and the roll angle, pitch angle, and yaw angle indicated by the reference information stored in the reference information storage unit 235. The joint position index calculation unit 212 generates joint position index information representing a joint position index vector whose elements are the roll angle, pitch angle, and yaw angle relative to the calculated reference posture, and stores the generated joint position index information in the joint position index storage unit 232, associating it with time information representing the time measured by the timing unit 208.
[0022] The dislocation risk estimation unit 213 selects dislocation risk estimation model information corresponding to the patient identification information included in the physical quantity notification frame from the dislocation risk estimation model information stored in the model storage unit 233. Then, using the dislocation risk estimation model indicated by the selected dislocation risk estimation model information, the dislocation risk estimation unit 213 estimates the expected value of the aforementioned multiple dislocation risk levels from the joint position index vector indicated by the joint position index information stored in the joint position index storage unit 232. Then, the dislocation risk estimation unit 213 selects the dislocation risk level with the highest expected value among the estimated multiple dislocation risk levels and stores the dislocation risk level information indicating the selected dislocation risk level in the estimation result storage unit 234 in association with the aforementioned time information.
[0023] When the high-risk period notification acquisition unit 220 acquires the high-risk period start notification information, described later, transmitted from the management server 3, it extracts the high-risk period start information and high-risk period cycle information contained in the acquired high-risk period start notification information. The high-risk period notification acquisition unit 220 then notifies the dislocation risk notification unit 214 of this extracted information.
[0024] The dislocation risk notification unit 214 refers to the dislocation risk level information stored in the estimated result storage unit 234 and, if the dislocation risk level indicated by the most recent dislocation risk information is equal to or greater than a preset dislocation risk threshold, executes a dislocation risk notification process to notify the user that the dislocation risk is high. Specifically, the dislocation risk notification unit 214 notifies the display control unit 216 of dislocation risk notification command information that instructs the display control unit 216 to notify the user that the dislocation risk is high, and generates a dislocation risk notification command signal which is output to the audio signal generation unit 244 and the vibration generator 243. Furthermore, when the dislocation risk notification unit 214 receives high dislocation risk period start information and high dislocation risk period cycle information from the high dislocation risk period notification acquisition unit 220, it notifies the display control unit 216 of the notified high dislocation risk period start information and high dislocation risk period cycle information.
[0025] When the display control unit 216 receives the aforementioned dislocation risk notification command information from the dislocation risk notification unit 214, it forms a dislocation risk notification image to inform the user that there is a high risk of dislocation and displays it on the display unit 241. Furthermore, when the display control unit 216 receives information on the start date of the high dislocation risk period and information on the cycle of the high dislocation risk period from the dislocation risk notification unit 214, it forms a high dislocation risk period notification image based on this information to inform the user that the high dislocation risk period has started, and the length and cycle of the high dislocation risk period, and displays it on the display unit 241.
[0026] The joint position index history information notification unit 218 generates joint position index history notification information, which includes joint position index information and time information stored in the joint position index storage unit 232, and dislocation risk level information at the same time stored in the estimated result storage unit 234, and transmits it to the management server 3. This joint position index history notification information also includes patient identification information to identify the patient being monitored. As patient identification information, for example, an identification number unique to the sensor unit 1 worn by the patient can be used.
[0027] When the model acquisition unit 219 receives a model update operation from the input unit 205 to update the dislocation risk estimation model, it generates model request information to the management server 3 requesting the management server 3 to send dislocation risk estimation model information indicating the newly learned dislocation risk estimation model, and sends this information to the management server 3. Subsequently, when the model acquisition unit 219 acquires model notification information from the management server 3 that includes dislocation risk estimation model information indicating the learned dislocation risk estimation model, it extracts the dislocation risk estimation model information contained in the acquired model notification information. Then, the model acquisition unit 219 updates the dislocation risk estimation model information stored in the model storage unit 233 using the extracted dislocation risk estimation model information.
[0028] Returning to Figure 2, the management server 3 comprises a CPU 301, a main memory unit 302, an auxiliary memory unit 303, a communication unit 306, a timing unit 308, and a bus 309 connecting these to each other. The CPU 301 is, for example, a multi-core processor. The main memory unit 302 has volatile memory and is used as the work area for the CPU 301. The auxiliary memory unit 303 consists of large-capacity non-volatile memory and stores programs for realizing various functions of the management server 3. The communication unit 306 is connected to a network NW. The timing unit 308 has, for example, a real-time clock.
[0029] In the management server 3, the CPU 301 reads the program stored in the auxiliary storage unit 303 into the main storage unit 302 and executes it, so that, as shown in Figure 5, it functions as a joint position index history acquisition unit 311, a high dislocation risk period identification unit 312, a high dislocation risk period notification unit 313, a model notification unit 314, a model generation unit 315, and a teacher information acquisition unit 316. Also, the auxiliary storage unit 203 shown in Figure 2 has a joint position index history storage unit 331, a high dislocation risk period storage unit 332, a model storage unit 333, and a teacher information storage unit 334, as shown in Figure 5. The joint position index history storage unit 331 stores a combination of at least one joint position index information, dislocation risk level information, and time information included in the joint position index history notification information transmitted from the terminal device 2 in the past, associating it with date information representing the date the joint position index history notification information was acquired and patient identification information.
[0030] The high-dislocation risk period memory unit 332 stores high-dislocation risk period information, which indicates the time period during a day when the risk of dislocation is higher, based on the combination of past dislocation risk level information, time information, and date information stored in the joint position index history memory unit 331, in association with date information and patient identification information.
[0031] The model storage unit 333 stores dislocation risk estimation model information generated based on the joint position index information and dislocation risk information stored in the joint position index history storage unit 331, in association with patient identification information. The teacher information storage unit 334 stores teacher information including joint position index and dislocation risk level information when there is a high risk of joint dislocation. Here, the teacher information is generated, for example, based on the accumulation of joint position index when dislocation of the artificial joint occurs after total hip replacement surgery. Alternatively, the teacher information may include joint position index when performing movements for which there is evidence that are generally particularly high in the risk of artificial hip dislocation, such as complex joint movements of the hip.
[0032] When the joint position index history acquisition unit 311 acquires joint position index history notification information transmitted from the terminal device 2, it extracts joint position index information, dislocation risk level information, time information, and patient identification information contained in the acquired joint position index history notification information. The joint position index history acquisition unit 311 then stores this extracted information in the joint position index history storage unit 331, associating it with date information indicating the date on which the joint position index history notification information was acquired.
[0033] The high-dislocation risk period identification unit 312 identifies at least one second unit period in which the risk of dislocation is relatively high, based on the changes in the dislocation risk level within each first unit period indicated by the dislocation risk level information for each past first unit period stored in the joint position index history storage unit 331, whenever a preset high-dislocation risk period identification period occurs. Here, the high-dislocation risk period identification period can be set to occur, for example, every one to several months. The high-dislocation risk period identification unit 312 also calculates, for example, the average value of the dislocation risk level indicated by the dislocation risk level information for each of the multiple past first unit periods stored in the joint position index history storage unit 331. The high-dislocation risk period identification unit 312 then identifies a time period consisting of at least one second unit period in which the calculated average value of the dislocation risk level is equal to or greater than a preset dislocation risk level threshold, and stores the high-dislocation risk period information indicating the identified time period in the high-dislocation risk period storage unit 332. Here, the high-risk period information consists of time information that indicates a representative point in time for the second unit period, which includes, for example, the start date and cycle of the specified time period.
[0034] The high-dislocation risk period notification unit 313 refers to the high-dislocation risk period information stored in the high-dislocation risk period storage unit 332, and when the start of a high-dislocation risk period arrives, it generates high-dislocation risk period start notification information, which includes high-dislocation risk period start information indicating the start of the high-dislocation risk period and high-dislocation risk period cycle information indicating the cycle of the high-dislocation risk period, and transmits it to the terminal device 2.
[0035] When the model notification unit 314 receives the aforementioned model request information transmitted from the terminal device 2, it extracts the patient identification information contained in the acquired model request information and notifies the model generation unit 315. Subsequently, when the model storage unit 333 stores dislocation risk estimation model information indicating a newly learned dislocation risk estimation model, the model notification unit 314 generates model notification information including the dislocation risk estimation model information and transmits it to the terminal device 2.
[0036] The model generation unit 315 generates a dislocation risk estimation model using training information, which includes joint position indices and dislocation risk level information for cases with a high risk of dislocation, stored in the training information storage unit 334 beforehand. The model generation unit 315 then stores the dislocation risk estimation model information, which represents the generated dislocation risk estimation model, in the model storage unit 333. Subsequently, when the model generation unit 315 receives patient identification information from the model notification unit 314, it identifies past joint position index information and dislocation risk level information, which are stored in the joint position index history storage unit 331 and correspond to the notified patient identification information. The model generation unit 315 then generates a new dislocation risk estimation model using this identified information. Finally, the model generation unit 315 stores the dislocation risk estimation model information, which represents the newly generated dislocation risk estimation model, in the model storage unit 333. First, the model generation unit 315 uses the dislocation risk estimation model indicated by the dislocation risk estimation model information already stored in the model storage unit 333 to calculate the expected value of the probability of each of the multiple dislocation risk levels from past joint position index information. Next, the model generation unit 315 calculates the error between the expected value indicated by the corresponding past dislocation risk level information stored in the joint position index history storage unit 331 and the expected value of each dislocation risk level calculated from the past joint position index information stored in the joint position index history storage unit 331. Then, the model generation unit 315 determines the weight coefficients of the dislocation risk estimation model, which is a neural network, by backpropagation based on the calculated error, and stores the information indicating the determined weight coefficients in the model storage unit 333.
[0037] The teacher information acquisition unit 316 acquires teacher information generated by the user and stores the acquired teacher information in the teacher information storage unit 334. Here, when the teacher information acquisition unit 316 acquires model setting instruction information that instructs the setting of a dislocation risk estimation model, including teacher information transmitted from a terminal device (not shown) managed by, for example, a physical therapist who performs rehabilitation on a patient, it extracts the teacher information contained in the acquired model setting instruction information. The teacher information acquisition unit 316 then stores the extracted teacher information in the teacher information storage unit 334.
[0038] Next, the patient monitoring process performed by the terminal device 2 according to this embodiment will be described with reference to Figure 6. This patient monitoring process is started, for example, after power is supplied to the terminal device 2, when a program for performing the patient monitoring process is started in the terminal device 2. First, the physical quantity acquisition unit 211 determines whether or not the user has performed the aforementioned reference joint position setting operation on the input unit 205 (step S101). This reference joint position setting operation is an operation to acquire physical quantities at the zero position, which is the so-called reference posture of the sensor unit 1. If the physical quantity acquisition unit 211 determines that the reference setting operation has not been received (step S101: No), the process in step S104, which will be described later, is executed. On the other hand, if the physical quantity acquisition unit 211 determines that the reference setting operation has been received (step S101: Yes), it acquires physical quantity notification frames that are periodically transmitted from the sensor unit 1. The physical quantity acquisition unit 211 then extracts three types of angular velocity information and three types of acceleration information contained in each acquired physical quantity notification frame, and stores these extracted pieces of information in the physical quantity buffer 221 in chronological order, associating them with each other (step 102). Next, the reference setting unit 217 calculates the roll angle, pitch angle, and yaw angle in the absolute coordinate system when the user wearing the sensor unit 1 is in a pre-specified reference posture, based on the angular velocity indicated by each of the multiple angular velocity pieces of information stored in the physical quantity buffer 221 and the acceleration indicated by the corresponding acceleration information. Here, for example, an upright posture can be adopted as the reference posture. The reference setting unit 217 then stores the reference information indicating the calculated roll angle, pitch angle, and yaw angle in the reference information storage unit 235 (step S103).
[0039] Next, the physical quantity acquisition unit 211 determines whether or not it has received the dislocation risk determination start operation performed by the user to the input unit 205 (step S104). If the physical quantity acquisition unit 211 determines that it has not received the dislocation risk determination start operation (step S104: No), the process of step S115 described later is executed. On the other hand, if the physical quantity acquisition unit 211 determines that it has received the dislocation risk determination start operation (step S104: Yes), it determines whether or not the pre-set physical quantity acquisition time has arrived (step S105). If the physical quantity acquisition unit 211 determines that the physical quantity acquisition time has not yet arrived (step S105: No), the process of step S112 described later is executed. On the other hand, if the physical quantity acquisition unit 211 determines that the physical quantity acquisition time has arrived (step S105: Yes), it acquires the aforementioned physical quantity notification frame that is periodically transmitted from the sensor unit 1. The physical quantity acquisition unit 211 then extracts three types of angular velocity information and three types of acceleration information contained in each of the acquired physical quantity notification frames, and stores these extracted pieces of information in the physical quantity buffer 221 in chronological order, associating them with each other (step S106).
[0040] Subsequently, the joint position index calculation unit 212 calculates the roll angle, pitch angle, and yaw angle in the absolute coordinate system of the sensor unit 1 based on the angular velocity indicated by each of the multiple angular velocity information stored in the physical quantity buffer 221 and the acceleration indicated by the corresponding acceleration information. The joint position index calculation unit 212 also calculates the roll angle, pitch angle, and yaw angle relative to a reference posture, obtained by subtracting the calculated roll angle, pitch angle, and yaw angle of the sensor unit 1 from the roll angle, pitch angle, and yaw angle indicated by the reference information stored in the reference information storage unit 235. The joint position index calculation unit 212 then generates joint position index information representing a joint position index vector whose elements are the calculated roll angle, pitch angle, and yaw angle relative to the reference posture, and stores the generated joint position index information in the joint position index storage unit 232 in association with time information representing the time measured by the timing unit 208 (step S107).
[0041] Next, the dislocation risk estimation unit 213 uses the dislocation risk estimation model indicated by the dislocation risk estimation model information stored in the model storage unit 233 to estimate the expected values of the aforementioned multiple dislocation risk levels from the joint position index vector indicated by the joint position index information stored in the joint position index storage unit 232. Then, the dislocation risk estimation unit 213 selects the dislocation risk level with the highest expected value from among the estimated multiple dislocation risk levels, and stores the dislocation risk level information indicating the selected dislocation risk level in the estimation result storage unit 234 in association with the aforementioned time information (step S108).
[0042] Next, the joint position index history information notification unit 218 generates joint position index history notification information including the joint position index information and time information stored in the joint position index storage unit 232 and the dislocation risk level information at the same time stored in the estimated result storage unit 234, and transmits it to the management server 3 (step S109). Subsequently, the dislocation risk notification unit 214 refers to the dislocation risk level information stored in the estimated result storage unit 234 and determines whether the dislocation risk level indicated by the most recent dislocation risk information is equal to or greater than a preset dislocation risk threshold (step S110). If the dislocation risk notification unit 214 determines that the dislocation risk level indicated by the most recent dislocation risk information is less than the dislocation risk threshold (step S110: No), the process in step S112 described later is executed. On the other hand, when the dislocation risk notification unit 214 determines that the dislocation risk level indicated by the most recent dislocation risk information is equal to or greater than the dislocation risk threshold (step S110: Yes), it executes the aforementioned dislocation risk notification process (step 111). Here, the dislocation risk notification unit 214 notifies the display control unit 216 of the aforementioned dislocation risk notification command information and generates a dislocation risk notification command signal, which it outputs to the audio signal generation unit 244 and the vibration generator 243. When the display control unit 216 receives the dislocation risk notification command information, it forms a dislocation risk notification image and displays it on the display unit 241. Furthermore, when the audio signal generation unit 244 receives the dislocation risk notification command signal, it generates an audio signal and outputs it to the speaker 242, thereby outputting sound from the speaker 242. When the vibration generator 243 receives the dislocation risk notification command signal, it vibrates the entire terminal device 2 in response.
[0043] Subsequently, the high-dislocation risk period notification acquisition unit 220 determines whether or not it has acquired the high-dislocation risk period start notification information transmitted from the management server 3 (step S112). If the high-dislocation risk period notification acquisition unit 220 determines that it has not acquired the high-dislocation risk period start notification information (step S112: No), the process in step S114, described later, is executed. On the other hand, if the high-dislocation risk period notification acquisition unit 220 determines that it has acquired the high-dislocation risk period start notification information (step S112: Yes), it extracts the high-dislocation risk period start information and the high-dislocation risk period cycle information contained in the acquired high-dislocation risk period start notification information. The high-dislocation risk period notification acquisition unit 220 then notifies the dislocation risk notification unit 214 of this extracted information. Meanwhile, when the dislocation risk notification unit 214 receives information on the start date of the high dislocation risk period and the cycle of the high dislocation risk period from the high dislocation risk period notification acquisition unit 220, it notifies the display control unit 216 of the notified information on the start date of the high dislocation risk period and the cycle of the high dislocation risk period. The display control unit 216 then forms the aforementioned high dislocation risk period notification image based on this information and displays it on the display unit 241 (step S113).
[0044] Next, the physical quantity acquisition unit 211 determines whether a dislocation risk determination end operation performed by the user on the input unit 205 to terminate the dislocation risk determination processing has been received (step S114). Here, if the physical quantity acquisition unit 211 determines that the dislocation risk determination end operation has not been received (step S114: No), it executes the processing of step S105 again. On the other hand, if the physical quantity acquisition unit 211 determines that the dislocation risk determination end operation has been received (step S114: Yes), the model acquisition unit 219 determines whether the aforementioned model update operation performed by the user on the input unit 205 has been received (step S115). Here, if the model acquisition unit 219 determines that the model update operation has not been received (step S115: No), the processing of step S101 is executed again. On the other hand, if the model acquisition unit 219 determines that the model update operation has been received (step S115: Yes), it generates the aforementioned model request information and transmits it to the management server 3 (step S116). Subsequently, upon acquiring the model notification information transmitted from the management server 3, the model acquisition unit 219 extracts the dislocation risk estimation model information included in the acquired model notification information. Then, the model acquisition unit 219 uses the extracted dislocation risk estimation model information to update the dislocation risk estimation model information stored in the model storage unit 233 (step S117). Thereafter, the processing of step S101 is executed again.
[0045] Next, the patient monitoring information management process executed by the management server 3 according to this embodiment will be described with reference to Figure 7. This patient monitoring information management process is started, for example, when a program for executing the patient monitoring information management process is started on the management server 3. Furthermore, the model generation unit 315 has previously generated a dislocation risk estimation model using the teacher information stored in the teacher information storage unit 334, and has stored the dislocation risk estimation model information indicating the generated dislocation risk estimation model in the model storage unit 333. First, the joint position index history acquisition unit 311 determines whether or not it has acquired the joint position index history notification information transmitted from the terminal device 2 (step S201). If the joint position index history acquisition unit 311 determines that it has not acquired the joint position index history notification information (step S201: No), the process of step S203, which will be described later, is executed. On the other hand, when the joint position index history acquisition unit 311 determines that it has acquired joint position index history notification information (step S201: Yes), it extracts joint position index information, dislocation risk level information, time information, and patient identification information contained in the acquired joint position index history notification information. The joint position index history acquisition unit 311 then stores this extracted information in the joint position index history storage unit 331, associating it with date information indicating the date on which the joint position index history notification information was acquired (step S202).
[0046] Next, the high-dislocation risk period identification unit 312 determines whether or not the aforementioned high-dislocation risk period has arrived (step S203). If the high-dislocation risk period identification unit 312 determines that the high-dislocation risk period has not yet arrived (step S203: No), the process in step S205 described later is executed. On the other hand, if the high-dislocation risk period identification unit 312 determines that the high-dislocation risk period has arrived (step S203: Yes), it identifies at least one second unit period in which the dislocation risk is relatively high, based on the changes in the dislocation risk level within each first unit period indicated by the past dislocation risk level information within each first unit period stored in the joint position index history storage unit 331. Then, the high-dislocation risk period identification unit 312 stores the high-dislocation risk period information indicating the identified time period in the high-dislocation risk period storage unit 332 (step S204).
[0047] Next, the high-dislocation risk period notification unit 313 refers to the high-dislocation risk period information stored in the high-dislocation risk period storage unit 332 and determines whether or not the start of the high-dislocation risk period has arrived (step S205). If the high-dislocation risk period notification unit 313 determines that the start of the high-dislocation risk period has not yet arrived (step S205: No), the process in step S207 described later is executed. On the other hand, if the high-dislocation risk period notification unit 313 determines that the start of the high-dislocation risk period has arrived (step S205: Yes), it generates the aforementioned high-dislocation risk period start notification information and transmits it to the terminal device 2 (step S206).
[0048] Thereafter, the model notification unit 314 determines whether or not the aforementioned model request information transmitted from the terminal device 2 has been acquired (step S207). Here, if the model notification unit 314 determines that the model request information has not been acquired (step S207: No), the processing of step S201 is executed again. On the other hand, if the model notification unit 314 determines that the model request information has been acquired (step S207: Yes), it extracts the patient identification information included in the acquired model request information and notifies the model generation unit 315 of the extracted information. Then, when the patient identification information is notified from the model notification unit 314, the model generation unit 315 identifies the past joint limb position index information and dislocation risk level information corresponding to the notified patient identification information, which are stored in the joint limb position index history storage unit 331. Then, the model generation unit 315 newly generates a dislocation risk estimation model using these pieces of identified information. Then, the model generation unit 315 causes the model storage unit 333 to store dislocation risk estimation model information indicating the newly generated dislocation risk estimation model (step S208). Next, the model notification unit 314 generates model notification information including dislocation risk estimation model information indicating the learned dislocation risk estimation model newly stored in the model storage unit 333, and transmits the generated model notification information to the terminal device 2 (step S209). Next, the processing of step S201 is executed again.
[0049] As described above, according to the present embodiment, the sensor unit 1 is mounted on the body surface of the patient's body, so that no surgery is required even when the sensor unit 1 malfunctions, and the burden on the patient can be reduced accordingly. In addition, the joint limb position index calculation unit 212 calculates, based on the angular velocity and acceleration detected by the motion sensor 107, a joint limb position index that reflects the relative posture of one of a plurality of bones connected to each other via the joint present in the joint portion Po1 of the patient HU1 with respect to any one other bone. Then, the dislocation risk estimation unit 213 estimates the dislocation risk of the joint portion based on the calculated joint limb position index, and the dislocation risk notification unit 214 notifies the user of the estimated dislocation risk. As a result, the state of the patient's joint portion Po1 can be appropriately monitored, so that the patient's dislocation risk can be reduced.
[0050] Furthermore, the patient monitoring system according to this embodiment can provide patients with feedback on movement patterns that reduce the risk of dislocation of the artificial joint after, for example, total hip arthroplasty, thereby promoting behavioral change in the patient. Since it can reduce the rate of dislocation of the artificial joint after total hip arthroplasty, it can reduce the rate of revision hip arthroplasty. Moreover, it can reduce movement risks in cases where postoperative joint movement patterns are restricted, such as after rotator cuff repair and total elbow arthroplasty.
[0051] Although embodiments of the present invention have been described above, the present invention is not limited to the configuration of the embodiments described above. For example, the dislocation risk estimation unit may identify at least one second unit period in which the risk of dislocation is high by clustering the joint position index indicated by the joint position index information or the dislocation risk level indicated by the dislocation risk level information into a plurality of groups. Specifically, after clustering the joint position index for each of the plurality of second unit periods within the first unit period to identify a plurality of groups, the average value of the dislocation risk level corresponding to each of the plurality of groups is calculated. Then, based on the calculated average value of the dislocation risk level, at least one group with a high risk of dislocation is identified, and the second unit period belonging to each of the identified at least one group is identified as a high dislocation risk period.
[0052] In the embodiment described, an example was given in which the sensor unit 1 is equipped with a motion sensor 107. However, the embodiment is not limited to this, and for example, the sensor unit 1 may be equipped with a geomagnetic sensor or the like in addition to the motion sensor 107. In this case, the reference setting unit 217 and the joint position index calculation unit 212 may use, for example, the 3-axis gyro sensor and 3-axis acceleration sensor, which are the motion sensor 107, and, for example, the 3-axis geomagnetic sensor to calculate the Euler angles (roll angle, pitch angle and yaw angle) that reflect the attitude of the sensor unit 1.
[0053] In this embodiment, an example was described in which the dislocation risk estimation model outputs an expected value vector whose elements are the expected probabilities of each of several dislocation risk levels, from a three-dimensional joint position index vector showing the roll angle, pitch angle, and yaw angle relative to the reference posture of the sensor unit 1. However, the joint position index vector is not limited to this, and for example, in addition to the roll angle, pitch angle, and yaw angle relative to the reference posture, it may also include at least one element selected from angular velocity around the three axes in the absolute coordinate system and acceleration in the three axial directions. Furthermore, if the sensor unit 1 is equipped with a geomagnetic sensor or the like, the joint position index vector may also include an element indicating the orientation of the patient.
[0054] In the embodiments described, an example was given in which the sensor unit 1 and the terminal device 2 are separate components. However, the present invention is not limited to this, and a patient monitoring device that combines the hardware configuration of the sensor unit 1 and the hardware functions of the terminal device 2 described in the embodiments is also included in the scope of the present invention. In this case, the patient monitoring device has the functions of the sensor unit 1 and the terminal device 2 described in the embodiments.
[0055] In this embodiment, an example was described in which the terminal device 2 is equipped with a vibration generator 243, but the invention is not limited to this, and for example, the sensor unit 1 may be equipped with a vibration generator.
[0056] In this embodiment, the high-dislocation-risk period identification unit 312 may identify movement patterns of patients with a relatively high risk of dislocation based on the changes in the dislocation risk level within each first unit period indicated by the past dislocation risk level information for each first unit period stored in the joint position index history storage unit 331.
[0057] Furthermore, the various functions of the terminal device 2 and management server 3 according to the present invention can be implemented using a normal computer system, without requiring a dedicated system. For example, the terminal device 2 and management server 3 that perform the above-mentioned operations may be configured by distributing a program for performing the above operations to computers connected to a network, stored on a non-temporary recording medium (such as a CD-ROM (Compact Disc Read Only Memory)) that can be read by the computer system, and then installing the program into the computer system.
[0058] Furthermore, the method of providing the program to the computer is arbitrary. For example, the program may be uploaded to a bulletin board system (BBS) on a communication line and distributed to the computer via the communication line. The computer then launches this program and executes it under the control of the operating system (OS), just like any other application. In this way, the computer functions as a terminal device 2 and a management server 3 that perform the aforementioned processing.
[0059] This invention allows for various embodiments and modifications without departing from the broad spirit and scope of the invention. Furthermore, the embodiments described above are for illustrative purposes only and do not limit the scope of the invention. In other words, the scope of the invention is indicated by the claims, not by the embodiments. Various modifications made within the scope of the claims and the equivalent meaning of the invention are considered to be within the scope of the invention.
[0060] This application is based on Japanese Patent Application No. 2025-088370, filed on 9 May 2025. The entire specification, claims, and drawings of Japanese Patent Application No. 2025-088370 are incorporated herein by reference.
[0061] This invention is suitable as a system for monitoring patients with implanted artificial joints.
[0062] 1: Sensor unit, 2: Terminal device, 3: Management server, 8: Access point, 9: Switch, 101: Processing circuit, 106: Wireless module, 107: Motion sensor, 201, 301: CPU, 202, 302: Main memory unit, 203, 303: Auxiliary memory unit, 208, 308: Timing unit, 241: Display unit, 205: Input unit, 261: First wireless module, 262: Second wireless module, 209, 309: Bus, 211: Physical quantity acquisition unit, 212: Joint position index calculation unit, 213: Dislocation risk estimation unit, 214: Dislocation risk notification unit, 216: Display control unit, 217: Reference setting unit, 218 : Joint position index history information notification unit, 219: Model acquisition unit, 220: High dislocation risk period notification acquisition unit, 221: Physical quantity buffer, 232: Joint position index storage unit, 233, 333: Model storage unit, 234: Estimation result storage unit, 242: Speaker, 243: Vibration generator, 244: Audio signal generation unit, 311: Joint position index history information acquisition unit, 312: High dislocation risk period identification unit, 313: High dislocation risk period notification unit, 314: Model notification unit, 315: Model generation unit, 331: Joint position index history storage unit, 332: High dislocation risk period storage unit, HU1: Human body, NW: Network, Po1: Joint part
Claims
1. A patient monitoring system comprising: a motion sensor attached to the body surface of at least one of two body parts aligned via a joint of a patient; a joint position index calculation unit that calculates a joint position index that reflects the relative positional relationship of one of a plurality of bones connected to each other via a joint present in the joint, based on a physical quantity detected by the motion sensor; a dislocation risk estimation unit that estimates the dislocation risk of the joint based on the joint position index; and a dislocation risk notification unit that notifies the user of the estimated dislocation risk.
2. The patient monitoring system according to claim 1, wherein the dislocation risk estimation unit estimates the dislocation risk using a dislocation risk estimation model that determines the dislocation risk from the joint position index.
3. The patient monitoring system according to claim 1 or 2, wherein there are multiple types of physical quantities.
4. The patient monitoring system according to claim 1 or 2, further comprising a physical quantity acquisition unit that repeatedly acquires a physical quantity detected by the motion sensor from the motion sensor at a preset sampling period.
5. A patient monitoring system according to claim 1 or 2, comprising: a sensor unit having the motion sensor and a first wireless module; and a terminal device having a second wireless module, wherein the sensor unit generates a physical quantity notification frame indicating the physical quantity detected by the motion sensor, and transmits the generated physical quantity notification frame to the terminal device via the first wireless module; and the terminal device comprises a physical quantity acquisition unit that acquires the physical quantity notification frame transmitted from the sensor unit via the second wireless module, a joint position index calculation unit, a dislocation risk estimation unit, and a dislocation risk notification unit.
6. The patient monitoring system according to claim 2, further comprising a model generation unit that generates the dislocation risk estimation model using training information including joint position indicators for cases where the joint portion is at high risk of dislocation.
7. The patient monitoring system according to claim 6, further comprising a teacher information acquisition unit that acquires the teacher information generated by the user.
8. The patient monitoring system according to claim 1 or 2, wherein the motion sensor is attached to the body surface of either one of two adjacent body parts, at least via the patient's hip joint, and the joint position index reflects at least one movement selected from hip flexion, hip adduction, and hip internal rotation.
9. A patient monitoring system according to claim 1 or 2, further comprising: a joint position index history storage unit that stores the joint position index calculated from each of the physical quantities detected in each of the multiple second unit periods obtained by dividing a preset first unit period into a plurality of second unit periods shorter than the first unit period, in association with time information indicating a representative point in time within the second unit period; and a high dislocation risk period identification unit that identifies at least one second unit period in which the dislocation risk is high, based on the progression of the dislocation risk level indicating the degree of dislocation risk within the first unit period.
10. A patient monitoring system according to claim 1 or 2, further comprising: a joint position index history storage unit that stores the joint position index calculated from each of the physical quantities detected in each of the multiple second unit periods obtained by dividing a preset first unit period into a plurality of second unit periods shorter than the first unit period, in association with time information indicating a representative point in time within the second unit period; and a high dislocation risk period identification unit that identifies at least one of the second unit periods in which the risk of dislocation is high by clustering the joint position index into a plurality of groups.
11. The patient monitoring system according to claim 2, further comprising: a model storage unit that stores dislocation risk estimation model information, which indicates a dislocation risk estimation model corresponding to each patient, in association with patient identification information that identifies the patient; and a physical quantity acquisition unit that acquires the patient identification information that identifies the patient to whom the motion sensor is to be attached, along with a physical quantity detected by the motion sensor, wherein the dislocation risk estimation unit selects the dislocation risk estimation model information corresponding to the acquired patient identification information from the dislocation risk estimation model information stored in the model storage unit, and determines the dislocation risk using the dislocation risk estimation model indicated by the selected dislocation risk estimation model information.
12. The patient monitoring system according to claim 1 or 2, wherein the dislocation risk notification unit comprises: an audio signal generation unit that generates an audio signal indicating a warning sound when the estimated dislocation risk is high; and a speaker that outputs the warning sound based on the audio signal.
13. The patient monitoring system according to claim 1 or 2, wherein the dislocation risk notification unit has a vibration generator that generates vibrations when the estimated dislocation risk is high.
14. A patient monitoring device comprising: a joint position index calculation unit that calculates a joint position index that reflects the relative positional relationship of one of a plurality of bones connected to each other via a joint present in the joint, based on a physical quantity detected by a motion sensor attached to the surface of at least one of two body parts aligned via a joint of the patient; a dislocation risk estimation unit that estimates the dislocation risk of the joint based on the joint position index; and a dislocation risk notification unit that notifies the user of the estimated dislocation risk.
15. A patient monitoring method comprising: a step of a patient monitoring device calculating a joint position index that reflects the relative positional relationship of any one of a plurality of bones connected to each other via a joint present in the joint, based on a physical quantity detected by a motion sensor attached to the surface of at least one of two body parts aligned via a joint of the patient; a step of the patient monitoring device estimating the risk of dislocation of the joint based on the joint position index; and a step of the patient monitoring device informing the user of the estimated risk of dislocation.
16. A program to cause a computer to function as: a joint position index calculation unit that calculates a joint position index that reflects the relative positional relationship of one of a plurality of bones connected to each other via a joint present in the joint, based on physical quantities detected by a motion sensor attached to the surface of at least one of two body parts aligned via a joint of a patient; a dislocation risk estimation unit that estimates the risk of dislocation of the joint based on the joint position index; and a dislocation risk notification unit that notifies the user of the estimated dislocation risk.
