JP7224731B2 - Shunt sound management system, shunt sound management computer, shunt sound management application program, and shunt sound management method - Google Patents

Description

The present invention relates to a method of outputting information indicating whether the shunt sound is normal or abnormal to a shunt sound management system for managing the shunt sound of a dialysis patient.

In order for dialysis patients to replace blood, bypass surgery is performed to connect veins and arteries called shunts. This bypass may become constricted, and the degree of constriction can be determined by listening to the sound of the shunt (shunt sound) with a stethoscope. Patients can identify their own abnormalities by listening with a stethoscope, but since it is often difficult for the elderly and those with hearing disabilities to hear the sound of a stethoscope, a system that can be easily used by patients themselves can detect the shunt sound. should be able to manage

As a device for assisting a doctor's diagnosis of shunt stenosis or the like by using shunt sounds, for example, a method using an array sound pickup sensor is disclosed (Patent Document 1).

JP 2018-149274 A

However, with the technique of Patent Document 1, it is necessary to attach an arrayed sound pickup sensor to the shunt portion, and it is difficult for patients such as the elderly to use the technique to manage the shunt sound on a daily basis.

To solve this problem, the inventor of the present invention acquires the shunt sound with an electronic stethoscope, wearable device, etc., receives the acquired shunt sound with an application program on the terminal side, and outputs the waveform of the received shunt sound to the terminal. By doing so, the patient can easily manage the shunt sound using their own terminal. In addition, by comparing the shunt sound with the normal shunt sound of the patient himself/herself, if the waveform of the shunt sound is different from normal, or if there is an abnormality such as when the upper limit of the frequency is exceeded, the terminal will notify the medical staff. It is possible.

According to the present invention, a shunt sound management system for managing the shunt sound of a dialysis patient outputs information indicating whether the shunt sound is normal or abnormal, thereby prompting the patient to self-manage the stenosis. It is an object of the present invention to provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method capable of preventing or treating stenosis at an early stage by detecting a sign at an early stage.

The present invention provides the following solutions.

The invention according to the first feature is
A shunt sound management system for managing the shunt sound of a dialysis patient,
a shunt sound acquiring means for acquiring the shunt sound of the hemodialysis patient;
data receiving means for wirelessly receiving shunt sound data from the shunt sound obtaining means;
Waveform output means for outputting the received shunt sound data as a waveform;
a shunt sound comparing means for comparing the received data of the shunt sound with the normal shunt sound of the hemodialysis patient and determining whether the received data of the shunt sound is abnormal;
User attribute acquisition means for acquiring user attributes of the dialysis patient;
with
A shunt sound management system is provided in which the shunt sound comparison means performs machine learning by associating the user attribute of the hemodialysis patient with the shunt sound to improve the accuracy of comparison.

According to the first aspect of the invention, in a shunt sound management system for managing shunt sounds of hemodialysis patients, shunt sound acquisition means for acquiring the shunt sounds of the hemodialysis patients; Data receiving means for wirelessly receiving shunt sound data; Waveform output means for outputting the received shunt sound data as a waveform; shunt sound comparison means for comparing and determining whether the received shunt sound data is abnormal; and user attribute acquisition means for acquiring user attributes of the dialysis patient, wherein the shunt sound comparison means improves the accuracy of comparison by associating the user attribute of the hemodialysis patient with the shunt sound and performing machine learning.

Although the invention according to the first feature belongs to the category of shunt sound management system, the shunt sound management computer, the shunt sound management application program, and the shunt sound management method have similar functions and effects.

The invention according to a second characteristic is the shunt sound management system according to the invention according to the first characteristic,
and notification means for notifying that the dialysis patient is abnormal when the received shunt sound is determined to be abnormal.

According to the second aspect of the invention, in the shunt sound management system of the first aspect of the invention, when the received shunt sound is determined to be abnormal, the dialysis patient is notified of the abnormality. and notification means for performing .

The invention according to a third characteristic is the shunt sound management system according to the invention according to the first characteristic,
A shunt sound management system is provided, wherein the user attributes include any item of sex, height, age, and type of shunt.

According to the third aspect of the invention, in the shunt sound management system of the first aspect of the invention, the user attribute includes any one of sex, height, age, and type of shunt.

An invention according to a fourth feature is a shunt sound management system according to any one of the first to third features,
The shunt sound acquisition means is an electronic stethoscope to provide a shunt sound management system.

According to a fourth aspect of the invention, in the shunt sound management system according to any one of the first to third aspects, the shunt sound acquiring means is an electronic stethoscope.

The invention according to the fifth feature is
A shunt sound management computer for managing the shunt sound of a dialysis patient acquired from a terminal,
Data acquisition means for acquiring the shunt sound of the dialysis patient;
Waveform output means for outputting the acquired shunt sound data as a waveform to the terminal;
a shunt sound comparing means for comparing the acquired shunt sound data with the normal shunt sound of the hemodialysis patient and determining whether the acquired shunt sound data is abnormal ;
notification means for notifying an abnormality of the dialysis patient when the acquired shunt sound is determined to be abnormal ;
User attribute acquisition means for acquiring user attributes of the dialysis patient;
with
Provided is a shunt sound management computer characterized in that the shunt sound comparison means performs machine learning by associating the user attribute of the dialysis patient with the shunt sound to improve the accuracy of comparison.

According to the fifth aspect of the invention, in a shunt sound management computer for managing a shunt sound of a dialysis patient acquired from a terminal, data acquisition means for acquiring the shunt sound of the dialysis patient; Waveform output means for outputting shunt sound data as a waveform to the terminal, and comparing the acquired shunt sound data with the normal shunt sound of the hemodialysis patient to determine if the acquired shunt sound data is abnormal. shunt sound comparing means for determining whether or not there is a shunt sound; notification means for notifying an abnormality of the dialysis patient when the acquired shunt sound is determined to be abnormal; and acquiring a user attribute of the dialysis patient. the shunt sound comparison means performs machine learning by associating the user attributes of the hemodialysis patient with the shunt sound, and notification means for improving the accuracy of the comparison. , provided.

The invention according to the sixth feature is
For the shunt sound management system for managing the shunt sound of dialysis patients,
obtaining a shunt sound of the dialysis patient;
a step of outputting the obtained shunt sound data as a waveform;
comparing the acquired shunt sound data with the normal shunt sound of the dialysis patient to determine whether the acquired shunt sound data is abnormal ;
a step of notifying the dialysis patient of an abnormality when the acquired shunt sound is determined to be abnormal;
obtaining user attributes of the dialysis patient;
and
The obtaining step provides an application program for improving comparison accuracy by performing machine learning by associating the user attribute of the dialysis patient with the shunt sound.

The invention according to the seventh feature is
In the shunt sound management method for managing the shunt sound of a dialysis patient,
obtaining a shunt sound of the dialysis patient;
wirelessly receiving the acquired shunt sound data;
a step of outputting the received shunt sound data as a waveform;
comparing the received shunt sound data with the normal shunt sound of the dialysis patient to determine whether the received shunt sound data is abnormal;
obtaining user attributes of the dialysis patient;
with
The determining step provides a shunt sound management method, wherein machine learning is performed by associating the user attribute of the dialysis patient with the shunt sound, thereby improving accuracy of comparison.

According to the present invention, a shunt sound management system for managing the shunt sound of a dialysis patient outputs information indicating whether the shunt sound is normal or abnormal, thereby encouraging self-management by the patient, Early detection of signs of stenosis makes it possible to provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method that enable prevention and early treatment of stenosis.

1 is a schematic diagram of a preferred embodiment of the present invention; FIG. 1 is an outline of a shunt sound management system and a functional block diagram of a terminal 100 and an electronic stethoscope 200; FIG. 4 is a flowchart of waveform output processing; FIG. 1 is a functional block diagram of a terminal 100 and an electronic stethoscope 200, and an overview of a shunt sound management system in the case of notifying when a shunt sound is abnormal; FIG. It is a flowchart figure of an abnormality notification process. 1 is a functional block diagram of a terminal 100 and an electronic stethoscope 200 and an overview of a shunt sound management system when performing machine learning processing based on user attribute acquisition; FIG. It is a flowchart figure of the machine-learning process by user attribute acquisition. 1 is an outline of a shunt sound management system when processing is performed by a computer 300, and a functional block diagram of a terminal 100 and a computer 300; FIG. FIG. 3 is a flow chart diagram when processing is performed by the computer 300. FIG. 1 is an outline of a shunt sound management system when processing is performed by the terminal 100 alone, and a functional block diagram of the terminal 100. FIG. FIG. 10 is a flowchart for the case where the terminal 100 alone performs processing; It is an example of a screen outputting a waveform to the terminal 100 . It is an example of a screen when a notification is received from the shunt sound management system.

BEST MODE FOR CARRYING OUT THE INVENTION The best mode for carrying out the present invention will be described below with reference to the drawings. This is just an example, and the technical scope of the present invention is not limited to this.

[Outline of shunt sound management system]
FIG. 1 is a schematic diagram of a preferred embodiment of the present invention. Based on this FIG. 1, the outline of the present invention will be explained. The shunt sound management system may consist of a terminal 100, an electronic stethoscope 200, and a communication network 400, as shown in FIG. It is assumed that the terminal 100 is a device capable of operating the shunt sound management system. Although a smartphone is illustrated here, it may be a laptop computer, tablet terminal, wearable device, smart watch, or the like. The number is not limited to one and may be plural. The terminal 100 includes a control section 110, a communication section 120, a storage section 130, an input section 140, and an output section 150. FIG. The control unit 110 includes a data acquisition unit 111 . The control unit 110 cooperates with the communication unit 120, the storage unit 130, the input unit 140, and the output unit 150 as necessary to implement each means. Also, the output section 150 implements a waveform output section 151 in cooperation with the control section 110 . The electronic stethoscope 200 is a stethoscope for electrically acquiring shunt sounds. A device similar to an analog stethoscope is illustrated here, but it may be a smart phone, tablet terminal, wearable device, smart watch, smart speaker, or the like. The electronic stethoscope 200 is assumed to have a function capable of clearly acquiring the shunt sound near the shunt portion. The electronic stethoscope 200 comprises a control section 210 , a communication section 220 , a storage section 230 , an input section 240 and an output section 250 . The input unit 240 includes the shunt sound acquisition unit 21 . Communication network 400 is a wireless communication network that enables communication between terminal 100 and electronic stethoscope 200 .

In the shunt sound management system of FIG. 1, first, the shunt sound acquisition unit 241 of the electronic stethoscope 200 acquires the patient's shunt sound (step S101). The shunt sound obtained here needs to be a clear sound that can determine whether the shunt section is normal or abnormal. The electronic stethoscope 200 may alert the user if an appropriate shunt sound cannot be obtained. The user of the shunt sound management system here is assumed to be the patient himself or a person who has the terminal 100, such as the patient's family member or caregiver.

Next, the data receiving unit 111 of the terminal 100 receives the patient's shunt sound data from the electronic stethoscope 200 (step S102). When receiving shunt sound data, the patient's sex, height, age, type of shunt, etc. may be received together as user attributes. When the user attributes are acquired from the electronic stethoscope 200 each time, the user attributes may be stored in the storage unit 230 of the electronic stethoscope 200 or may be stored in the storage unit 130 on the terminal 100 side.

Finally, the waveform output unit 151 of the terminal 100 causes the output unit 150 of the terminal 100 to output the waveform of the shunt sound based on the shunt sound data received in step S102 (step S103). As in the example displayed on the screen of the terminal 100 at the bottom of FIG. 1, the time when the shunt sound was acquired may be displayed together with the waveform. Also, the name of the patient who is the user and the user attributes may be displayed together. Also, by repeating steps S101 to S103, the waveform output in step S103 may be updated in real time. Furthermore, the data received in step S102 and the waveform output in step S103 may be stored in the storage unit 130 of the terminal 100 as a log. By checking the output waveform, the user can confirm whether the shunt sound is normal or abnormal. As early treatment, it is possible to widen the stenosis with a balloon catheter.

As described above, according to the present invention, the waveform of the shunt sound is output to the shunt sound management system for managing the shunt sound of the hemodialysis patient, thereby encouraging the patient to self-manage the stenosis and detecting signs of stenosis. Early detection makes it possible to provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method that enable prevention and early treatment of stenosis.

[Description of each function]
FIG. 2 is an outline of the shunt sound management system and a functional block diagram of the terminal 100 and the electronic stethoscope 200. As shown in FIG. The shunt sound management system comprises a terminal 100, an electronic stethoscope 200, and a communication network 400. FIG.

It is assumed that the terminal 100 is a device capable of operating the shunt sound management system. Although a smartphone is illustrated here, it may be a laptop computer, tablet terminal, wearable device, smart watch, or the like. The number is not limited to one and may be plural. The terminal 100 includes a control section 110, a communication section 120, a storage section 130, an input section 140, and an output section 150. FIG.

The terminal 100 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), etc. as a control unit 110 . The control unit 110 includes a data acquisition unit 111 . The control unit 110 cooperates with the communication unit 120, the storage unit 130, the input unit 140, and the output unit 150 as necessary to implement each means.

The terminal 100 includes a device or the like that enables communication with the electronic stethoscope 200 as a communication unit 120 . Assume that the communication method is wireless.

The terminal 100 includes a data storage unit as the storage unit 130, and stores necessary data such as shunt sound data, waveform data, user attributes, etc. acquired from the electronic stethoscope 200. FIG.

Terminal 100 includes a device for realizing input as input unit 140 . Examples include touch panels, pen tablets, microphones, keyboards, and mice.

Terminal 100 includes a device for realizing output as output unit 150 . Examples are displays, speakers, and the like. Also, the output section 150 implements a waveform output section 151 in cooperation with the control section 110 .

The electronic stethoscope 200 is a stethoscope for electrically acquiring shunt sounds. A device similar to an analog stethoscope is illustrated here, but it may be a smart phone, tablet terminal, wearable device, smart watch, smart speaker, or the like. The electronic stethoscope 200 is assumed to have a function capable of clearly acquiring the shunt sound near the shunt portion. The electronic stethoscope 200 comprises a control section 210 , a communication section 220 , a storage section 230 , an input section 240 and an output section 250 .

The electronic stethoscope 200 includes a CPU, RAM, ROM, etc. as a control unit 210 .

The electronic stethoscope 200 includes a device or the like for enabling communication with the terminal 100 as a communication unit 220 . Assume that the communication method is wireless.

The electronic stethoscope 200 includes a data storage unit as a storage unit 230, and stores necessary data such as acquired shunt sounds and user attributes.

Electronic stethoscope 200 includes a device for realizing input as input unit 240 . An example is a microphone or the like. The input unit 240 includes a shunt sound acquisition unit 241 .

Electronic stethoscope 200 includes a device for realizing output as output section 250 . An example is a speaker or the like.

Communication network 400 is a wireless communication network that enables communication between terminal 100 and electronic stethoscope 200 .

[Waveform output processing]
FIG. 3 is a flowchart of waveform output processing. Processing executed by each unit described above will be described with reference to this flowchart.

First, the user starts the shunt sound management system using the input unit 140 of the terminal 100 (step S301). The user of the shunt sound management system here is assumed to be the patient himself or a person who has the terminal 100, such as the patient's family member or caregiver. A login process may be performed for the user to start the shunt sound management system.

Next, the electronic stethoscope 200 connects to the terminal 100 (step S302). This is to confirm whether a connection is possible in order to transmit data from the electronic stethoscope 200 to the terminal 100 . Here, if the connection to the terminal 100 does not go well, it is desirable to notify the user of the error via the output section 250 of the electronic stethoscope 200 and instruct the user on how to improve the situation.

Next, the shunt sound acquisition unit 241 of the electronic stethoscope 200 acquires the patient's shunt sound (step S303). The shunt sound obtained here needs to be a clear sound that can determine whether the shunt section is normal or abnormal. The user may be warned via the output section 250 of the electronic stethoscope 200 if an appropriate shunt sound cannot be obtained. Examples of cases in which an appropriate shunt sound cannot be obtained include the case where the electric stethoscope 200 is floating from the shunt portion, and the case where the electronic stethoscope 200 is considered to be in contact with a portion other than the shunt joint. At this time, it is desirable to warn the user by voice, alert sound, or the like from the output unit 250 of the stethoscope 200 .

Next, the electronic stethoscope 200 transmits the patient's shunt sound data to the terminal 100 via the communication unit 220 (step S304).

The data receiving unit 111 of the terminal 100 receives the patient's shunt sound data from the electronic stethoscope 200 (step S305). When receiving shunt sound data, the patient's sex, height, age, type of shunt, etc. may be received together as user attributes. When the user attributes are acquired from the electronic stethoscope 200 each time, the user attributes may be stored in the storage unit 230 of the electronic stethoscope 200 or may be stored in the storage unit 130 on the terminal 100 side.

Finally, the waveform output unit 151 of the terminal 100 causes the output unit 150 of the terminal 100 to output the waveform of the shunt sound based on the shunt sound data received in step S305 (step S306). Although not shown here, the waveform output in step S306 may be updated in real time by repeating steps S303 to S306. Furthermore, the data received in step S305 and the waveform output in step S306 may be stored in the storage unit 130 of the terminal 100 as a log.

FIG. 12 is an example of a screen on which waveforms are output to the terminal 100. FIG. As in the example of FIG. 12, the time when the shunt sound was acquired may be displayed together with the waveform. Also, the name of the patient who is the user and the user attributes may be displayed together. By checking the output waveform, the user can confirm whether the shunt sound is normal or abnormal. As early treatment, it is possible to widen the stenosis with a balloon catheter.

As described above, according to the present invention, the waveform of the shunt sound is output to the shunt sound management system for managing the shunt sound of the hemodialysis patient, thereby encouraging the patient to self-manage the stenosis and detecting signs of stenosis. Early detection makes it possible to provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method that enable prevention and early treatment of stenosis.

[Error notification process]
FIG. 4 is a functional block diagram of the terminal 100 and the electronic stethoscope 200 and an outline of the shunt sound management system in the case of notifying when the shunt sound is abnormal. In addition to the configuration of FIG.

FIG. 5 is a flowchart of abnormality notification processing. Steps S501 to S506 in FIG. 5 correspond to steps S301 to S306 in the flow in FIG. 3, so steps after step S507 will be described in detail here.

After outputting the waveform in step S506, the shunt sound comparison unit 112 of the terminal 100 compares the normal shunt sound with the shunt sound received in step S505 (step S507). It is assumed that normal shunt sounds are stored in the storage unit 130 of the terminal 100 in advance. In the comparison, attention is paid to whether the waveform is significantly different from the normal shunt sound and whether the frequency exceeds the upper limit value calculated from the normal shunt sound.

Next, the shunt sound comparison unit 112 of the terminal 100 determines whether or not the shunt sound received in step S505 is abnormal. (step S508). The case where there is an abnormality is the case where the waveform is significantly different from the normal shunt sound in the comparison in step S507, or the case where the frequency exceeds the upper limit value calculated from the normal shunt sound.

If the shunt sound is determined to be abnormal, the notification unit 113 of the terminal 100 notifies the dialysis patient of abnormality (step S509). The notification here may be performed by display or sound via the output unit 150 of the terminal 100, or via the communication unit 120 by a person registered in advance such as a medical worker, a caregiver, or a family member. may be sent by e-mail or telephone.

FIG. 13 is an example of a screen when a notification is received from the shunt sound management system. It notifies the user (Mr. **) that there is an abnormality in the shunt sound together with the date and time. By displaying the waveform of the abnormal shunt sound next to the waveform of the normal shunt sound, it is possible to make it easier to understand the abnormal part and to judge the degree of urgency when the medical staff sees it. becomes easier.

As described above, according to the present invention, the shunt sound management system for managing the shunt sound of the hemodialysis patient outputs information indicating whether the shunt sound is normal or abnormal, so that the patient can determine whether the shunt sound is normal or abnormal. To provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method capable of preventing and early treatment of stenosis by promoting self-management and early detection of signs of stenosis. becomes possible.

[Machine learning processing by user attribute acquisition]
FIG. 6 is a functional block diagram of the terminal 100 and the electronic stethoscope 200 and an overview of the shunt sound management system when performing machine learning processing based on user attribute acquisition. In addition to the configuration of FIG. 4 , the control unit 110 of the terminal 100 is provided with a user attribute acquisition unit 114 .

FIG. 7 is a flowchart of machine learning processing based on user attribute acquisition.

First, the user starts the shunt sound management system using the input unit 140 of the terminal 100 (step S701). The user of the shunt sound management system here is assumed to be the patient himself or a person who has the terminal 100, such as the patient's family member or caregiver. A login process may be performed for the user to start the shunt sound management system.

Next, the electronic stethoscope 200 connects to the terminal 100 (step S702). This is to confirm whether a connection is possible in order to transmit data from the electronic stethoscope 200 to the terminal 100 . Here, if the connection to the terminal 100 does not go well, it is desirable to notify the user of the error via the output section 250 of the electronic stethoscope 200 and instruct the user on how to improve the situation.

Next, the user attribute acquisition unit 114 of the terminal 100 acquires user attributes (step S703). The user attributes may be obtained from the electronic stethoscope 200 or from the storage unit 130 of the terminal 100 . If you are using the system for the first time, you may be asked to register your user attributes here. The user attributes here are the patient's sex, height, age, type of shunt, etc., and are physical characteristics that affect the waveform of the shunt sound.

Next, the shunt sound acquisition unit 241 of the electronic stethoscope 200 acquires the patient's shunt sound (step S704). The shunt sound obtained here needs to be a clear sound that can determine whether the shunt section is normal or abnormal. The user may be warned via the output section 250 of the electronic stethoscope 200 if an appropriate shunt sound cannot be obtained. Examples of cases in which an appropriate shunt sound cannot be obtained include the case where the electric stethoscope 200 is floating from the shunt portion, and the case where the electronic stethoscope 200 is considered to be in contact with a portion other than the shunt joint. At this time, it is desirable to warn the user by voice, alert sound, or the like from the output unit 250 of the stethoscope 200 .

Next, the electronic stethoscope 200 transmits the patient's shunt sound data to the terminal 100 via the communication unit 220 (step S705).

The data receiving unit 111 of the terminal 100 receives the patient's shunt sound data from the electronic stethoscope 200 (step S706).

Next, the waveform output unit 151 of the terminal 100 causes the output unit 150 of the terminal 100 to output the waveform of the shunt sound based on the shunt sound data received in step S706 (step S707). Although not shown here, by repeating steps S704 to S707, the waveform output in step S707 may be updated in real time. Furthermore, the data received in step S706 and the waveform output in step S707 may be stored in the storage unit 130 of the terminal 100 as a log.

FIG. 12 is an example of a screen on which waveforms are output to the terminal 100. FIG. As in the example of FIG. 12, the time when the shunt sound was acquired may be displayed together with the waveform. Also, the name of the patient who is the user and the user attributes may be displayed together. By checking the output waveform, the user can confirm whether the shunt sound is normal or abnormal. As early treatment, it is possible to widen the stenosis with a balloon catheter.

Next, the shunt sound comparison unit 112 of the terminal 100 compares the normal shunt sound with the shunt sound received in step S706 (step S708). It is assumed that normal shunt sounds are stored in the storage unit 130 of the terminal 100 in advance. In the comparison, attention is paid to whether the waveform is significantly different from the normal shunt sound and whether the frequency exceeds the upper limit value calculated from the normal shunt sound.

Next, the shunt sound comparison unit 112 of the terminal 100 determines whether or not the shunt sound received in step S706 is abnormal. , to step 710 (step S709). The case where there is an abnormality is the case where the waveform is significantly different from the normal shunt sound in the comparison in step S708, or the case where the frequency exceeds the upper limit value calculated from the normal shunt sound.

If the shunt sound is determined to be abnormal, the notification unit 113 of the terminal 100 notifies the dialysis patient of the abnormality (step S710). The notification here may be performed by display or sound via the output unit 150 of the terminal 100, or via the communication unit 120 by a person registered in advance such as a medical worker, a caregiver, or a family member. may be sent by e-mail or telephone.

FIG. 13 is an example of a screen when a notification is received from the shunt sound management system. It notifies the user (Mr. **) that there is an abnormality in the shunt sound together with the date and time. By displaying the waveform of the abnormal shunt sound next to the waveform of the normal shunt sound, it is possible to make it easier to understand the abnormal part and to judge the degree of urgency when the medical staff sees it. becomes easier.

Returning to FIG. 7, finally, the user attribute acquisition unit 114 of the terminal 100 performs machine learning based on the user attribute acquired in step S703 and the shunt sound comparison result in step S708 (step S711). The machine learning here is performed by associating user attributes that affect the waveform of the shunt sound, such as the patient's gender, height, age, and type of shunt, with the waveform. This is for increasing the accuracy of shunt noise abnormality determination. Further, machine learning may be used to determine how the action was taken in response to the notification made in step S710. Actions in response to the notification include, for example, "judgment that there is no abnormality", "wait and see", "respond with preventive measures", "respond with early treatment", "emergency hospitalization", and the like. By confirming the action in response to this notification, it is possible to confirm whether or not the shunt sound comparison result in step S708 was correct. Therefore, by performing machine learning, it is possible to improve the accuracy of shunt sound abnormality determination.

As described above, according to the present invention, by introducing machine learning into the shunt sound management system for managing the shunt sound of hemodialysis patients, information on whether the shunt sound is normal or abnormal can be obtained. Shunt sound management system, shunt sound management computer, shunt It is possible to provide a sound management application program and a shunt sound management method.

[Shunting sound management system when processing is performed by computer 300]
FIG. 8 is an outline of the shunt sound management system when processing is performed by the computer 300, and a functional block diagram of the terminal 100 and the computer 300. As shown in FIG. FIG. 8 describes a case where the terminal 100 acquires the shunt sound without using the electronic stethoscope 200, but as described above, the electronic stethoscope 200 acquires the shunt sound, You may perform wireless communication with.

It is assumed that the terminal 100 is a device capable of using the shunt sound management system. Although a smartphone is illustrated here, it may be a laptop computer, tablet terminal, wearable device, smart watch, or the like. The number is not limited to one and may be plural. The terminal 100 includes a control section 110, a communication section 120, a storage section 130, an input section 140, and an output section 150. FIG.

The terminal 100 includes a CPU, RAM, ROM, etc. as a control unit 110 . The control unit 110 cooperates with the communication unit 120, the storage unit 130, the input unit 140, and the output unit 150 as necessary to implement each means.

The terminal 100 includes a device or the like for enabling communication with the computer 300 as the communication unit 120 .

The terminal 100 includes a data storage unit as the storage unit 130, and stores necessary data such as acquired shunt sound data, waveform data, user attributes, and the like.

Terminal 100 includes a device for realizing input as input unit 140 . Examples include touch panels, pen tablets, microphones, keyboards, and mice. The input unit 140 includes a shunt sound acquisition unit 141 . The shunt sound acquisition section 141 is assumed to have a function capable of clearly acquiring the shunt sound near the shunt section.

Terminal 100 includes a device for realizing output as output unit 150 . Examples are displays, speakers, and the like. The output section 150 also includes a waveform output section 151 .

Assume that the computer 300 is a device capable of operating the shunt sound management system. Although a desktop computer is illustrated here, a notebook computer, a tablet terminal, or the like may be used. It may be a virtual computer, and the number is not limited to one and may be plural. The computer 300 includes a control section 310 , a communication section 320 , a storage section 330 , an input section 340 and an output section 350 .

The computer 300 includes a CPU, RAM, ROM, etc. as a control unit 310 . The control unit 310 includes a data acquisition unit 311 , a waveform output unit 312 , a shunt sound comparison unit 313 , a notification unit 314 and a user attribute acquisition unit 315 . The control unit 310 cooperates with the communication unit 320, the storage unit 330, the input unit 340, and the output unit 350 as necessary to implement each means.

The computer 300 includes a device or the like for enabling communication with the terminal 100 as a communication unit 320 .

The computer 300 includes a data storage unit as a storage unit 330 and stores necessary data such as shunt sound data, waveform data, user attributes, etc. acquired from the terminal 100 .

Computer 300 includes a device for realizing input as input unit 340 . Examples include keyboards, mice, touch panels, pen tablets, and microphones.

Terminal 100 includes a device for realizing output as output section 350 . Examples are displays, speakers, and the like.

FIG. 9 is a flow chart for processing in the computer 300. As shown in FIG.

First, the user starts the shunt sound management system using the input unit 140 of the terminal 100 (step S901). The user of the shunt sound management system here is assumed to be the patient himself or a person who has the terminal 100, such as the patient's family member or caregiver. In order for the user to start the shunt sound management system, a login process or notification to the computer 300 may be performed.

The terminal 100 then transmits the user attributes to the computer 300 (step S902). The user attribute may be input by the user via the input unit 140, or may be stored in the storage unit 130 of the terminal 100. FIG. If you are using the system for the first time, you may be asked to register your user attributes here. The user attributes here are the patient's sex, height, age, type of shunt, etc., and are physical characteristics that affect the waveform of the shunt sound.

The user attribute acquisition unit 314 of the computer 300 acquires user attributes from the terminal 100 (step S903).

Next, the shunt sound acquisition unit 141 of the terminal 100 acquires the patient's shunt sound (step S904). The shunt sound obtained here needs to be a clear sound that can determine whether the shunt section is normal or abnormal. If an appropriate shunt sound cannot be obtained, the user may be warned via the output unit 150 of the terminal 100 . At this time, it is desirable to display on the output unit 150 of the terminal 100, or to warn the user by voice, alert sound, or the like.

Next, the terminal 100 transmits the patient's shunt sound data to the computer 300 via the communication unit 120 (step S905).

The data acquisition unit 311 of the computer 300 acquires the patient's shunt sound data from the terminal 100 (step S906).

Next, the waveform output unit 312 of the computer 300 creates a shunt sound waveform based on the shunt sound data acquired in step S906 (step S907). The data received in step S906 and the waveform created in step S907 may be stored in the storage unit 330 of the computer 300 as a log.

Next, the waveform output unit 312 of the computer 300 transmits the waveform of the shunt sound acquired in step S907 to the terminal 100 (step S908).

The terminal 100 receives the waveform of the shunt sound (step S909).

Next, the waveform output unit 151 of the terminal 100 outputs the waveform of the shunt sound (step S910). Although not shown here, the waveform output in step S910 may be updated in real time by repeating steps S904 to S910.

FIG. 12 is an example of a screen on which waveforms are output to the terminal 100. FIG. As in the example of FIG. 12, the time when the shunt sound was acquired may be displayed together with the waveform. Also, the name of the patient who is the user and the user attributes may be displayed together. By checking the output waveform, the user can confirm whether the shunt sound is normal or abnormal. As early treatment, it is possible to widen the stenosis with a balloon catheter.

Returning to FIG. 9, next, the shunt sound comparison unit 313 of the computer 300 compares the normal shunt sound with the shunt sound acquired in step S906 (step S911). It is assumed that the normal shunt sound is stored in the storage unit 330 of the computer 300 in advance. In the comparison, attention is paid to whether the waveform is significantly different from the normal shunt sound and whether the frequency exceeds the upper limit value calculated from the normal shunt sound.

Next, the shunt sound comparison unit 313 of the computer 300 determines whether or not the shunt sound acquired in step S906 is abnormal. , to step 913 (step S912). The case where there is an abnormality is the case where the waveform is significantly different from the normal shunt sound in the comparison in step S911, or the case where the frequency exceeds the upper limit value calculated from the normal shunt sound.

If the shunt sound is determined to be abnormal, the notification unit 314 of the computer 300 notifies the dialysis patient of abnormality (step S913). The notification here may be performed by display or sound via the output unit 150 of the terminal 100, or via the communication unit 320 of the computer 300, by pre-registering medical personnel, caregivers, family members, etc. It may be done by e-mail, telephone, or the like to the person who left it.

FIG. 13 is an example of a screen when a notification is received from the shunt sound management system. It notifies the user (Mr. **) that there is an abnormality in the shunt sound together with the date and time. By displaying the waveform of the abnormal shunt sound next to the waveform of the normal shunt sound, it is possible to make it easier to understand the abnormal part and to judge the degree of urgency when the medical staff sees it. becomes easier.

Returning to FIG. 9, finally, the user attribute acquisition unit 315 of the terminal 100 performs machine learning based on the user attribute acquired in step S903 and the shunt sound comparison result in step S911 (step S914). The machine learning here is performed by associating user attributes that affect the waveform of the shunt sound, such as the patient's gender, height, age, and type of shunt, with the waveform. This is for increasing the accuracy of shunt noise abnormality determination. Furthermore, machine learning may use how the action was taken in response to the notification performed in step S913. Actions in response to the notification include, for example, "judgment that there is no abnormality", "wait and see", "respond with preventive measures", "respond with early treatment", "emergency hospitalization", and the like. By confirming the action in response to this notification, it is possible to confirm whether or not the shunt sound comparison result in step S911 was correct. Therefore, by performing machine learning, it is possible to improve the accuracy of the abnormality determination of the shunt sound. Step S914 can also be omitted.

According to the present invention, a shunt sound management system for managing the shunt sound of a dialysis patient outputs information indicating whether the shunt sound is normal or abnormal, thereby encouraging self-management by the patient, Early detection of signs of stenosis makes it possible to provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method that enable prevention and early treatment of stenosis.

[Shunting sound management system when processing is performed by the terminal 100 alone]
FIG. 10 is an outline of a shunt sound management system when processing is performed by the terminal 100 alone, and a functional block diagram of the terminal 100. As shown in FIG. In this case, the shunt sound management system can also be expressed as a shunt sound management application program.

It is assumed that terminal 100 is a device capable of executing a shunt sound management application program. Although a smartphone is illustrated here, it may be a laptop computer, tablet terminal, wearable device, smart watch, or the like. The number is not limited to one and may be plural. The terminal 100 includes a control section 110, a communication section 120, a storage section 130, an input section 140, and an output section 150. FIG.

The terminal 100 includes a CPU, RAM, ROM, etc. as a control unit 110 . The passing unit 100 includes a shunt sound comparison unit 111 , a notification unit 112 and a user attribute acquisition unit 113 . The control unit 110 cooperates with the communication unit 120, the storage unit 130, the input unit 140, and the output unit 150 as necessary to implement each means.

The terminal 100 includes a device or the like as a communication unit 120 for enabling communication with other devices.

The terminal 100 includes a data storage unit as the storage unit 130, and stores necessary data such as acquired shunt sound data, waveform data, user attributes, and the like.

Terminal 100 includes a device for realizing input as input unit 140 . Examples include touch panels, pen tablets, microphones, keyboards, and mice. The input unit 140 includes a shunt sound acquisition unit 141 . The shunt sound acquisition section 141 is assumed to have a function capable of clearly acquiring the shunt sound near the shunt section.

Terminal 100 includes a device for realizing output as output unit 150 . Examples are displays, speakers, and the like. Also, the output section 150 implements a waveform output section 151 in cooperation with the control section 110 .

FIG. 11 is a flowchart for the case where the terminal 100 alone performs processing.

First, the user uses the input unit 140 of the terminal 100 to start the shunt sound management application program (step S1101). The user of the shunt sound management application program here is assumed to be the patient himself or a person who has the terminal 100, such as the patient's family member or caregiver.

Next, the user attribute acquisition unit 113 acquires user attributes (step S1102). The user attribute may be input by the user via the input unit 140, or may be stored in the storage unit 130 of the terminal 100. FIG. If you are using the system for the first time, you may be asked to register your user attributes here. The user attributes here are the patient's sex, height, age, type of shunt, etc., and are physical characteristics that affect the waveform of the shunt sound.

Next, the shunt sound acquisition unit 141 acquires the patient's shunt sound (step S1103). The shunt sound obtained here needs to be a clear sound that can determine whether the shunt section is normal or abnormal. If an appropriate shunt sound cannot be obtained, the user may be warned via the output unit 150 of the terminal 100 . At this time, it is desirable to display on the output unit 150 of the terminal 100, or to warn the user by voice, alert sound, or the like.

Next, the waveform output unit 151 outputs the waveform of the shunt sound (step S1104). Although not shown here, the waveform output in step S1104 may be updated in real time by repeating steps S1103 to S1104.

FIG. 12 is an example of a screen on which waveforms are output to the terminal 100. FIG. As in the example of FIG. 12, the time when the shunt sound was acquired may be displayed together with the waveform. Also, the name of the patient who is the user and the user attributes may be displayed together. By checking the output waveform, the user can confirm whether the shunt sound is normal or abnormal. As early treatment, it is possible to widen the stenosis with a balloon catheter.

Returning to FIG. 11, next, the shunt sound comparison unit 111 compares the normal shunt sound with the shunt sound acquired in step S1103 (step S1105). It is assumed that normal shunt sounds are stored in the storage unit 130 in advance. In the comparison, attention is paid to whether the waveform is significantly different from the normal shunt sound and whether the frequency exceeds the upper limit value calculated from the normal shunt sound.

Next, the shunt sound comparison unit 111 determines whether or not the shunt sound acquired in step S1103 is abnormal. (step S1106). The case where there is an abnormality is the case where the waveform is significantly different from the normal shunt sound in the comparison in step S1105, or the case where the frequency exceeds the upper limit value calculated from the normal shunt sound.

When it is determined that the shunt sound is abnormal, the notification unit 112 notifies the dialysis patient of abnormality (step S1107). The notification here may be performed by display or sound via the output unit 150, or via the communication unit 120 to a person registered in advance such as a medical worker, a caregiver, or a family member It may be done by e-mail or telephone.

FIG. 13 is an example of a screen when a notification is received from the shunt sound management system. It notifies the user (Mr. **) that there is an abnormality in the shunt sound together with the date and time. By displaying the waveform of the abnormal shunt sound next to the waveform of the normal shunt sound, it is possible to make it easier to understand the abnormal part and to judge the degree of urgency when the medical staff sees it. becomes easier.

Returning to FIG. 11, finally, the user attribute acquisition unit 315 performs machine learning based on the user attribute acquired in step S1102 and the shunt sound comparison result in step S1105 (step S1108). The machine learning here is performed by associating user attributes that affect the waveform of the shunt sound, such as the patient's gender, height, age, and type of shunt, with the waveform. This is for increasing the accuracy of shunt noise abnormality determination. Furthermore, machine learning may use how the action was taken in response to the notification performed in step S1107. Actions in response to the notification include, for example, "judgment that there is no abnormality", "wait and see", "respond with preventive measures", "respond with early treatment", "emergency hospitalization", and the like. By confirming the action in response to this notification, it is possible to confirm whether or not the shunt sound comparison result in step S1105 was correct. Therefore, by performing machine learning, it is possible to improve the accuracy of the abnormality determination of the shunt sound. Step S1108 can also be omitted.

As described above, according to the present invention, the shunt sound management system for managing the shunt sound of the hemodialysis patient outputs information indicating whether the shunt sound is normal or abnormal, so that the patient can determine whether the shunt sound is normal or abnormal. To provide a shunt sound management system, a shunt sound management computer, a shunt sound management application program, and a shunt sound management method capable of preventing and early treatment of stenosis by promoting self-management and early detection of signs of stenosis. becomes possible.

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments described above. Moreover, the effects described in the embodiments of the present invention are merely enumerations of the most suitable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. isn't it.

100 terminal, 200 electronic stethoscope, 300 computer, 400 communication network

Claims (7)

A shunt sound management system for managing the shunt sound of a dialysis patient,
a shunt sound acquiring means for acquiring the shunt sound of the hemodialysis patient;
data receiving means for wirelessly receiving shunt sound data from the shunt sound obtaining means;
Waveform output means for outputting the received shunt sound data as a waveform;
a shunt sound comparing means for comparing the received data of the shunt sound with the normal shunt sound of the hemodialysis patient and determining whether the received data of the shunt sound is abnormal;
User attribute acquisition means for acquiring user attributes of the dialysis patient;
with
The shunt sound management system, wherein the shunt sound comparison means improves accuracy of comparison by associating user attributes of the dialysis patient with the shunt sound and performing machine learning. notification means for notifying the dialysis patient of an abnormality when the received shunt sound is determined to be abnormal;
The shunt sound management system of claim 1, comprising: 2. The shunt sound management system according to claim 1 , wherein the user attributes include any item of sex, height, age, and type of shunt. The shunt sound management system according to any one of claims 1 to 3 , wherein the shunt sound acquisition means is an electronic stethoscope. A shunt sound management computer for managing the shunt sound of a dialysis patient acquired from a terminal,
Data acquisition means for acquiring the shunt sound of the dialysis patient;
Waveform output means for outputting the acquired shunt sound data as a waveform to the terminal;
a shunt sound comparing means for comparing the acquired shunt sound data with the normal shunt sound of the hemodialysis patient and determining whether the acquired shunt sound data is abnormal ;
notification means for notifying an abnormality of the dialysis patient when the acquired shunt sound is determined to be abnormal ;
User attribute acquisition means for acquiring user attributes of the dialysis patient;
with
The shunt sound management computer, wherein the shunt sound comparison means improves the accuracy of comparison by associating the user attribute of the dialysis patient with the shunt sound and performing machine learning. A shunt sound management application program for managing the shunt sound of a dialysis patient,
obtaining a shunt sound of the dialysis patient;
a step of outputting the acquired shunt sound data as a waveform;
comparing the acquired shunt sound data with the normal shunt sound of the dialysis patient to determine whether the acquired shunt sound data is abnormal ;
a step of notifying the dialysis patient of an abnormality when the acquired shunt sound is determined to be abnormal ;
obtaining user attributes of the dialysis patient;
with
The shunt sound management application program, wherein in the obtaining step, the user attribute of the dialysis patient and the shunt sound are associated with each other and machine learning is performed to improve accuracy of comparison. A shunt sound management method for managing the shunt sound of a dialysis patient, comprising:
obtaining a shunt sound of the dialysis patient;
wirelessly receiving the acquired shunt sound data;
a step of outputting the received shunt sound data as a waveform;
comparing the received shunt sound data with the normal shunt sound of the dialysis patient to determine whether the received shunt sound data is abnormal;
obtaining user attributes of the dialysis patient;
with
The shunt sound management method, wherein in the determining step, the user attribute of the dialysis patient and the shunt sound are associated with each other and machine learning is performed to improve accuracy of comparison.

Images