JP7550325B1 - Blood Purification Device - Google Patents

Abstract

To provide a blood purification device that can select staff who can respond appropriately in real time when an abnormality occurs in a patient or the blood purification device, and can quickly notify the selected staff of the abnormality.
[Solution] This blood purification device has a required skill acquisition unit 3, a staff information acquisition unit 4, a staff determination processing unit 5 that generates and stores a learning model obtained by performing machine learning to determine staff for abnormalities in a specific patient or an abnormality in the blood purification device 1 using accumulated data that has been accumulated when an abnormality occurs as teacher data, a selection unit 6 that inputs the staff information into the learning model of the staff determination processing unit 5 to select the staff to respond, and an abnormality notification unit 7 that notifies a mobile device 2 carried by the selected staff member of the abnormality.
[Selected Figure] Figure 1

Description

The present invention relates to a blood purification device for purifying a patient's blood and performing blood purification therapy.

Dialysis machines, which serve as blood purification devices used in dialysis treatments, etc., are installed in multiple dialysis rooms in medical facilities such as hospitals, and dialysis treatments (blood purification treatments) are administered to a large number of patients in the dialysis rooms. For example, the blood purification device disclosed in Patent Document 1 is configured to display information about the patient on the display unit (monitor, etc.) of the blood purification device, allowing medical staff and other staff to understand the patient's condition and to respond quickly and appropriately when an abnormality occurs.

JP 2019-311 A

However, in the above-mentioned conventional blood purification devices, although it is possible to display information about a patient on the display unit of the blood purification device, staff who can take appropriate and prompt action based on that information during treatment are not necessarily located near the blood purification device where action is required, and it is necessary to search for staff who can take appropriate action within the medical facility. Furthermore, such issues can arise when responding to abnormalities in the blood purification device as well as abnormalities in the patient.

On the other hand, it is possible to pre-select staff members who can respond to each abnormality in the patient or blood purification device, and then call the pre-selected staff members within the medical facility when an abnormality occurs in the patient or blood purification device. However, it is difficult to pre-select appropriate staff members for each abnormality, and even if they are selected, they may not be able to respond quickly to the abnormality depending on the staff members' locations or schedules when the abnormality occurs.

The present invention was made in consideration of these circumstances, and aims to provide a blood purification device that can select staff who can respond appropriately in real time when an abnormality occurs in the patient or blood purification device, and can quickly notify the selected staff of the abnormality.

A blood purification device according to one embodiment of the present invention is a blood purification device for purifying a patient's blood to perform blood purification therapy, comprising: a required skill acquisition unit that acquires skill information required to resolve an abnormality when an abnormality occurs in a patient being treated by the blood purification device or in the blood purification device; a staff information acquisition unit that acquires staff information including position information, schedule information and skill information of staff based on the required skill information acquired by the required skill acquisition unit; a staff determination processing unit that generates and stores a learning model obtained by performing machine learning to determine the staff member who should respond to an abnormality in a specific patient or abnormality in the blood purification device, using accumulated data, which is a plurality of pieces of staff information acquired by the staff information acquisition unit when an abnormality occurs in a patient being treated by the blood purification device or in the blood purification device, as teacher data; The blood purification device is provided with: a selection unit that, when an abnormality occurs in the purification device, acquires staff information using the staff information acquisition unit and inputs the staff information into a learning model of the staff judgment processing unit to select staff to respond; an abnormality notification unit that notifies the mobile terminal carried by the responding staff selected by the selection unit of the abnormality in the patient or the abnormality in the blood purification device; a prediction processing unit that generates and stores a learning model obtained by performing machine learning using accumulated data, which is a collection of multiple pieces of past information on patients being treated with the blood purification device or multiple pieces of past information on the blood purification device, as teacher data to predict the occurrence of an abnormality in a specific patient or an abnormality in the blood purification device and propose an appropriate way to deal with it; and a display control unit that displays the occurrence of the abnormality in a specific patient or an abnormality in the blood purification device predicted by the prediction processing unit and the proposal for an appropriate way to deal with it on a display unit possessed by the blood purification device .

According to the present invention, a learning model based on machine learning is used to determine the staff member who should respond when an abnormality occurs with the patient or blood purification device. Therefore, when an abnormality occurs with the patient or blood purification device, it is possible to select staff who can respond appropriately in real time and to promptly notify the selected staff member of the abnormality.

FIG. 1 is a block diagram showing the overall configuration of a blood purification device according to an embodiment of the present invention. FIG. 2 is a perspective view showing the appearance of the blood purification device and a mobile terminal. FIG. 1 is a schematic diagram illustrating blood purification treatment performed by the blood purification device. Schematic diagram showing the concept of machine learning applied to the staff determination processing unit of the blood purification device. Schematic diagram showing the concept of machine learning applied to the prediction processing unit of the blood purification device. A flowchart showing the steps for generating and storing a learning model by machine learning applied to the staff determination processing unit of the blood purification device. A flowchart showing the steps to be taken when an abnormality occurs in the patient or the blood purification device. A flowchart showing how the blood purification device responds when it predicts an abnormality in the patient's blood pressure or a sensor abnormality.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
1 and 2, the blood purification device according to this embodiment is used to purify a patient's blood for blood purification therapy, and is configured to include a blood purification device 1 for purifying a patient's blood to perform blood purification therapy, and a portable terminal 2 carried by staff including medical professionals and capable of communicating with the blood purification device 1. The staff refers to all workers involved in medical facilities, and includes doctors and medical professionals who can treat patients, as well as technical staff who can handle breakdowns and maintenance of the blood purification device 1.

As shown in FIG. 3, the blood purification device 1 includes a dialysis fluid inlet line L1 for introducing dialysis fluid and a dialysis fluid outlet line L2 for discharging dialysis fluid waste, and a duplex pump 16 is disposed across the dialysis fluid inlet line L1 and the dialysis fluid outlet line L2. A dialyzer 15 (blood purifier) is connected to the dialysis fluid inlet line L1 and the dialysis fluid outlet line L2, and a blood circuit 13 for circulating the patient's blood extracorporeally is connected to the dialyzer 15. By driving the blood pump 14, blood purification treatment can be performed by the dialyzer 15 on the patient's blood circulating extracorporeally through the blood circuit 13.

The pump chamber of the duplex pump 16 is separated by a single plunger (not shown) into a pump chamber connected to the dialysis fluid introduction line L1 and a pump chamber connected to the discharge line L2. The plunger reciprocates to supply the dialysis fluid or cleaning fluid sent to the pump chamber to the dialyzer 15, and to draw the drainage fluid from the dialyzer 15 into the pump chamber.

Furthermore, a bypass line L3 is formed on the dialysis fluid discharge line L2, bypassing the duplex pump 16, and a water removal pump 17 is disposed midway along the bypass line L3. By driving the water removal pump 17, it is possible to remove water from the patient's blood flowing through the dialyzer 15. Note that instead of the duplex pump 16, a so-called chamber type pump may be used.

As shown in FIG. 2, the blood purification device 1 is equipped with a display unit 1a, and is equipped with various treatment means (e.g., a blood pump, a fluid replacement pump, a syringe pump, etc.) related to blood purification treatment (hemodialysis treatment). Such treatment means are not limited to actuators such as pumps, but also include various general-purpose means used in blood purification treatment, such as clamping means such as solenoid valves and monitoring means for fluid pressure, etc.

As shown in FIG. 1, the blood purification device 1 according to this embodiment is configured to have a display unit 1a, as well as a required skills acquisition unit 3, a staff information acquisition unit 4, a staff judgment processing unit 5, a selection unit 6, an abnormality notification unit 7, a prediction processing unit 8, and a display control unit 9. The display unit 1a is composed of a monitor such as a touch panel, and is configured to display various information controlled by the display control unit 9 in addition to treatment information related to blood purification treatment.

The required skills acquisition unit 3 is made up of a microcomputer or the like, and when an abnormality occurs in the patient being treated by the blood purification device 1 or in the blood purification device 1, it acquires the skill information required to resolve the abnormality from information previously input into the blood purification device. This makes it possible to select in advance which skills are required, such as skills related to patient treatment or skills related to the blood purification device 1, and allows for smooth subsequent selection of staff.

The staff information acquisition unit 4 is made up of a microcomputer or the like, and acquires staff information including staff position information, schedule information, and skill information based on the necessary skill information acquired by the necessary skill acquisition unit 3. The staff position information is made up of information input in real time by radio waves from a mobile terminal 2 carried by the staff, and the position information is detected by a position detection unit 11 equipped in the mobile terminal 2.

In other words, the mobile terminal 2 uses the Global Positioning System (GPS) or other position measurement systems to detect location information in real time using the location detection unit 11, which allows the mobile terminal 2 to determine its current location. This location information is then transmitted to the blood purification device 1 via a wireless communication system such as Wi-Fi, making it possible for the staff member's location information to be acquired by the staff information acquisition unit 4 of the blood purification device 1.

Staff schedule information is input to the blood purification device 1 based on the schedule management unit 12 that manages staff work information (attendance information), and consists of information such as vacation schedules, business trip schedules, and schedules and time periods for events such as meetings and surgeries. The schedule management unit 12 may be a specific department (attendance management department) in a medical facility or a server located in a cloud system, and it is preferable that the schedule information can be updated sequentially.

Staff skill information is input to the blood purification device 1 based on the staff's expertise, skill, or proficiency (length of experience or number of experiences), and includes information such as doctors who can decide to change treatment conditions, specialist doctors who can specialize in specific treatments, medical professionals who can handle patients other than deciding to change treatment conditions (including changing settings based on the doctor's judgment), and technicians who can inspect and maintain the blood purification device 1 (including peripheral devices, etc.). It also includes the time available to perform the same work.

The staff determination processing unit 5 uses accumulated data of multiple pieces of staff information acquired by the staff information acquisition unit 4 when an abnormality occurs in a patient being treated by the blood purification device 1 or in the blood purification device 1 (including peripheral devices) as training data, and generates and stores a learning model obtained by performing machine learning to determine the staff member who should respond to a specific patient abnormality or abnormality in the blood purification device 1.

That is, the staff determination processing unit 5 is capable of generating a learning model using artificial intelligence (AI), and as shown in FIG. 4, when an abnormality occurs in a patient being treated by the blood purification device 1 or in the blood purification device 1, multiple (large number) pieces of staff information acquired by the staff information acquisition unit 4 are accumulated, and this vast amount of accumulated data is input in advance as training data. Then, based on the input training data, machine learning can be performed to determine the staff member who should (is suitable to) respond to a specific patient's abnormality or an abnormality in the blood purification device 1, and a learning model can be generated and stored.

Here, machine learning refers to a technology in which a computer learns large amounts of data and automatically constructs algorithms and models to perform tasks such as classification and prediction. As the technology and algorithms that make machine learning work, well-known machine learning algorithms such as neural networks can be used.

When an abnormality occurs in a patient being treated by the blood purification device 1 or in the blood purification device 1, the selection unit 6 acquires staff information using the staff information acquisition unit 4 and inputs the staff information into the learning model of the staff judgment processing unit 5 to select staff to respond, as shown in FIG. 4. In other words, when an abnormality actually occurs in a patient being treated by the blood purification device 1 or in the blood purification device 1, the selection unit 6 acquires staff information at that time using the staff information acquisition unit 4 and then inputs the staff information into the learning model of the artificial intelligence possessed by the staff judgment processing unit 5, which outputs information on staff suitable for responding and a priority order (staff available for first priority response, staff available for second priority response, etc.).

For example, when an abnormality occurs in the blood purification device 1, if the necessary skill information acquired by the necessary skill acquisition unit 3 for the abnormality is skill information capable of inspecting and maintaining the blood purification device 1, the number of staff members who have the skill information capable of inspecting and maintaining the blood purification device and can respond based on the current working status and location information from the pre-input staff information is narrowed down to two, and the learning model outputs the staff member who is currently performing another inspection but closest to the blood purification device 1 in terms of distance as the staff member who can respond with the highest priority (the staff member who should respond), and the staff member who is currently waiting but a little farther away from the blood purification device 1 in terms of distance as the second most capable staff member. In addition, the judgment may be made based not only on the distance but also on transportation means and traffic congestion information. In addition, if the time available to respond to the abnormality, including the travel time to the blood purification device 1, is within the allowable time range for responding to the abnormality, the staff member currently waiting at a location far from the blood purification device 1 may be determined as the staff member who can respond with the highest priority, and the staff member currently performing another inspection work close to the blood purification device 1 may be determined as the staff member who can respond with the second most priority.

The abnormality notification unit 7 notifies the mobile terminal 2 carried by the staff member selected by the selection unit 6 of an abnormality in the patient or the blood purification device 1. In other words, when the selection unit 6 selects the staff member to handle the case, the abnormality notification unit 7 can notify the selected staff member by displaying the occurrence of an abnormality on the display unit 10 of the mobile terminal 2, etc., of the selected staff member via a wireless communication system such as Wi-Fi.

If the staff member responds to the notified information by saying that they are not available (or does not respond within a certain time), the information may be notified to the mobile device 2 carried by the second available staff member. If the second available staff member responds by saying that they are not available (or does not respond within a certain time), the information may be notified to other available staff members in order of priority. Furthermore, the abnormality notification unit 7 may provide information on the staff member who should respond and the available staff member selected by the selection unit 6 and priority ranking information to other medical personnel via a notification means such as a screen display on the blood purification device 1 or a mobile terminal, and make a proposal to determine the staff member who should respond. Furthermore, the abnormality notification unit 7 may notify the mobile device 2 carried by all available staff members selected by the selection unit 6 of the abnormality of the patient or the abnormality of the blood purification device 1, and provide the content of the response from each staff member and the response time to other medical personnel via a notification means such as a screen display on the blood purification device 1 or a mobile terminal, and make a proposal to determine the staff member who should respond.

The mobile terminal 2, on the other hand, is made up of a smartphone or tablet that can be carried by staff, and is capable of continuous wireless communication with the blood purification device 1. It also has a display unit 10 made up of a touch panel or the like, and the position detection unit 11 described above. When an abnormality in the patient or blood purification device 1 is wirelessly transmitted from the abnormality notification unit 7 of the blood purification device 1, the occurrence of the abnormality is displayed on the display unit 10, and instructions can be given to the patient or blood purification device 1 in which the abnormality is occurring to take action.

The prediction processing unit 8 is composed of a microcomputer or the like, and uses accumulated data that includes multiple pieces of past information on patients being treated with the blood purification device 1 or past information on the blood purification device 1 (including past information on peripheral devices) as training data, and performs machine learning to predict the occurrence of abnormalities in a specific patient or the blood purification device 1 and to propose appropriate ways to deal with the abnormalities, thereby generating and storing a learning model.

That is, the prediction processing unit 8 is capable of generating a learning model using artificial intelligence (AI), and as shown in FIG. 5, accumulates multiple (large) pieces of past information on patients being treated with the blood purification device 1 (e.g., the patient's blood pressure over the course of treatment) or past information on the blood purification device 1 (sensor failures, pump malfunctions, etc.), and this vast amount of accumulated data is input in advance as training data. Then, based on the input training data, machine learning is performed to predict the occurrence of abnormalities in a specific patient or in the blood purification device 1 and propose appropriate countermeasures, and a learning model obtained can be generated and stored.

For example, when blood purification treatment is started, the treatment information at that time is sent to the prediction processing unit 8, and the received treatment information is input into an artificial intelligence learning model, which outputs a prediction of the occurrence of an abnormality in the patient or blood purification device 1 that is expected in the future if treatment is continued, and a proposal for how to deal with the predicted abnormality. The prediction of the occurrence of an abnormality in the patient or blood purification device 1 output in this manner, and the proposal for how to deal with the predicted abnormality, may be single or multiple.

The display control unit 9 displays the occurrence of an abnormality in a specific patient or an abnormality in the blood purification device 1 predicted by the prediction processing unit 8 and a proposal for an appropriate method of dealing with the abnormality on the display unit 1a of the blood purification device 1. The prediction processing unit 8 may also propose multiple methods of dealing with the predicted abnormality in a specific patient or an abnormality in the blood purification device 1, and the display control unit 9 may display the multiple methods of dealing with the abnormality proposed by the prediction processing unit 8 on the display unit 1a.

The occurrence of an abnormality in a specific patient or an abnormality in the blood purification device 1 predicted by the prediction processing unit 8 and the proposal of an appropriate way to deal with it may be displayed on the display unit 1a by the display control unit 9, or may be transmitted as data to the mobile terminal 2 and displayed on the display unit 10 of the mobile terminal 2 to notify staff.

Next, a method for generating and storing a learning model by the staff determination processor 5 according to this embodiment will be described with reference to the flowchart of FIG.
First, in S1, the staff skills and the skills required for treatment are acquired, and then in S2, the staff location information is acquired. Then, in S3, the staff schedule information is acquired, and in S4, the time when treatment actually started is acquired, and then the process proceeds to S5. In S5, the information acquired in S1 to S4 is accumulated as training data, and machine learning is performed to determine the staff member who should respond to an abnormality in a specific patient or an abnormality in the blood purification device 1, and a learning model is generated and stored.

Next, a method for selecting and notifying staff when an abnormality occurs in a patient being treated by the blood purification apparatus 1 or in the blood purification apparatus 1 will be described with reference to the flowchart of FIG.
First, when an abnormality occurs in a patient being treated by the blood purification device 1 or in the blood purification device 1, the skills required to deal with the abnormality (resolve the abnormality) are acquired by the required skill acquisition unit 3 in S1, and staff skill information is acquired by the staff information acquisition unit 4 in S2.

Then, in S3, the position of the blood purification device 1 that needs to be addressed for an abnormality (the position of the blood purification device 1 where the abnormality has occurred is identified because the position of each blood purification device 1 is set in advance). Then, in S4, staff position information and in S5, staff schedule information are acquired by the staff information acquisition unit 4, and the information acquired in the series of information acquisition steps Sa (S1 to S5) is input into the learning model in S6.

In this way, by inputting the information obtained in the series of information acquisition steps Sa into the learning model in S6, it is possible to output the staff member who should respond to the abnormality of a specific patient or an abnormality in the blood purification device 1, and the selection unit 6 selects the output staff member in S7 and notifies the mobile terminal 2 carried by that staff member of the abnormality. This abnormality notification includes the location information of the blood purification device 1 obtained in S3, so that the staff member who receives the notification can quickly head to the blood purification device 1.

Then, after selecting a staff member in S7 and notifying them of the abnormality, data after the abnormality occurs is acquired in S8, and the acquired data is input into the learning model of the staff determination processing unit 5 to update the learning model, completing the series of steps. In this way, each time an abnormality occurs, the associated data is input into the learning model and updated, so that the learning model can be updated continuously.

Next, a method for predicting the occurrence of an abnormality in a specific patient or in the blood purification apparatus 1 by the prediction processing unit 8 and proposing an appropriate countermeasure will be described with reference to the flowchart of FIG.
First, when blood purification treatment is started, various information related to the treatment is input into the prediction processing unit 8, and in S1 a learning model is used to predict the occurrence of an abnormality in a specific patient (e.g., a drop in the patient's blood flow) or an abnormality in the blood purification device 1 (e.g., a sensor abnormality), and in S2 the predicted abnormality and a proposal for an appropriate countermeasure are displayed on the display unit 1a.

Then, in S3, staff information is acquired. This staff information can be acquired, for example, by a process similar to the information acquisition process Sa shown in FIG. 7. The information acquired in S3 is then input to the learning model of the staff determination processing unit 5 in S4, and in S5, the selection unit 6 selects staff to respond to the predicted abnormality of the patient or the abnormality of the blood purification device 1, and notifies the mobile terminal 2 carried by the staff of the predicted abnormality.

In this way, after a staff member is selected in S5 and notified of the abnormality, data after the abnormality is predicted is acquired in S6, and the acquired data is input into the learning model of the prediction processing unit 8 to update the learning model, completing the series of steps. In this way, each time an abnormality is predicted, the associated data is input into the learning model and updated, so that the learning model can be updated continuously.

Although the present embodiment has been described above, the present invention is not limited to this, and may, for example, not have a prediction processing unit 8 and may not predict the occurrence of an abnormality in a specific patient or an abnormality in the blood purification device 1 and propose an appropriate method of dealing with it. In addition, the method of acquiring staff information and the communication means may be in various forms different from the above embodiment. Note that, although a dialysis device for dialysis treatment is applied in this embodiment, other devices that can purify the patient's blood while circulating it extracorporeally (e.g., a blood purification device used in hemodiafiltration, hemofiltration, AFBF, a plasma adsorption device, etc.) may also be applied.

The first embodiment of the present invention is a blood purification device 1 for purifying the blood of a patient to perform blood purification treatment, which includes a necessary skill acquisition unit 3 that acquires skill information required to resolve an abnormality when an abnormality occurs in the patient being treated by the blood purification device 1 or in the blood purification device 1, a staff information acquisition unit 4 that acquires staff information including staff position information, schedule information, and skill information based on the necessary skill information acquired by the necessary skill acquisition unit 3, and a teacher data that stores accumulated data of multiple pieces of staff information acquired by the staff information acquisition unit 4 when an abnormality occurs in the patient being treated by the blood purification device 1 or in the blood purification device 1. The system has a staff determination processing unit 5 that generates and stores a learning model obtained by performing machine learning to determine the staff member who should respond to an abnormality in a specific patient or an abnormality in the blood purification device 1 as data, a selection unit 6 that acquires staff information using a staff information acquisition unit 4 when an abnormality occurs in a patient being treated by the blood purification device 1 or an abnormality in the blood purification device 1, and inputs the staff information into the learning model of the staff determination processing unit 5 to select the staff member who should respond, and an abnormality notification unit 7 that notifies the mobile terminal 2 carried by the staff member who should respond selected by the selection unit 6 of the abnormality in the patient or the abnormality in the blood purification device 1.

As a result, a learning model based on machine learning is used to determine the staff member who should respond when an abnormality occurs with the patient or the blood purification device 1. Therefore, when an abnormality occurs with the patient or the blood purification device 1, staff who can respond appropriately can be selected in real time, and the selected staff member can be promptly notified of the abnormality.

In the second embodiment of the present invention, the staff position information acquired by the staff information acquisition unit 4 in the first embodiment is detected by the position detection unit 11 provided in the mobile terminal 2 and input in real time. This makes it possible to acquire staff position information with high accuracy and in real time.

In a third embodiment of the present invention, in the first embodiment, the staff schedule information acquired by the staff information acquisition unit 4 is input based on the schedule management unit 12 that manages the work information of the staff. This makes it possible to acquire detailed and accurate staff schedule information.

In the fourth embodiment of the present invention, in the first embodiment, the selection unit 6 determines the available staff, including the staff who should respond based on the learning model, and the priority order of the available staff, and the abnormality notification unit 7 notifies the mobile terminal 2 carried by the available staff of the abnormality in the patient or the abnormality in the blood purification device 1, and then notifies other available staff in accordance with the staff's response to the notification, according to the priority order. This makes it possible to notify other available staff in accordance with the staff's response, according to the priority order.

In a fifth embodiment of the present invention, in the first embodiment, the staff skill information acquired by the staff information acquisition unit 4 is input based on the staff's expertise, skill, or period or number of experiences. This makes it possible to acquire detailed and accurate staff skill information.

The sixth embodiment of the present invention is the first embodiment, and further includes a prediction processing unit 8 that generates and stores a learning model obtained by performing machine learning using accumulated data, which is a collection of multiple pieces of past information on patients being treated with the blood purification device 1 or past information on the blood purification device 1, as teacher data to predict the occurrence of abnormalities in a specific patient or abnormalities in the blood purification device and propose an appropriate method of dealing with the abnormalities, and a display control unit 9 that displays the occurrence of abnormalities in a specific patient or abnormalities in the blood purification device 1 predicted by the prediction processing unit 8 and the proposal of an appropriate method of dealing with the abnormalities on the display unit 1a of the blood purification device 1. As a result, it is possible to predict the occurrence of abnormalities in a specific patient or abnormalities in the blood purification device 1 and propose an appropriate method of dealing with the abnormalities, using the learning model that has been subjected to machine learning.

In the seventh embodiment of the present invention, in the sixth embodiment, the prediction processing unit 8 proposes multiple ways of dealing with a predicted abnormality of a specific patient or an abnormality of the blood purification device 1, and the display control unit 9 causes the display unit 1a to display the multiple ways of dealing with the abnormality proposed by the prediction processing unit 8. This allows the staff to select the most appropriate one of the multiple proposed ways of dealing with the abnormality.

As long as the purpose is the same as that of the present invention, it can also be applied to products with different external shapes or products with added functions.

REFERENCE SIGNS LIST 1 Blood purification device 1a Display unit 2 Portable terminal 3 Required skill acquisition unit 4 Staff information acquisition unit 5 Staff judgment processing unit 6 Selection unit 7 Abnormality notification unit 8 Prediction processing unit 9 Display control unit 10 Display unit 11 Position detection unit 12 Schedule management unit 13 Blood circuit 14 Blood pump 15 Dialyzer 16 Duplex pump 17 Water removal pump

Claims (6)

A blood purification device for purifying a patient's blood and performing blood purification therapy, comprising:
a required skill acquisition unit that acquires skill information required for solving an abnormality when an abnormality occurs in a patient being treated by the blood purification apparatus or in the blood purification apparatus;
a staff information acquisition unit that acquires staff information including location information, schedule information, and skill information of the staff based on the necessary skill information acquired by the necessary skill acquisition unit;
a staff determination processing unit that generates and stores a learning model obtained by performing machine learning to determine the staff member who should respond to an abnormality in a specific patient or an abnormality in the blood purification device, using accumulated data, which is a plurality of pieces of staff information acquired by the staff information acquisition unit when an abnormality occurs in a patient being treated by the blood purification device or in the blood purification device, as training data;
a selection unit which, when an abnormality occurs in a patient being treated by the blood purification apparatus or in the blood purification apparatus, acquires the staff information by the staff information acquisition unit and inputs the staff information into a learning model of the staff determination processing unit to select a staff member to handle the abnormality;
an abnormality notification unit that notifies a mobile terminal carried by the staff member selected by the selection unit of the abnormality of the patient or the abnormality of the blood purification apparatus;
a prediction processing unit that generates and stores a learning model obtained by performing machine learning using past information of patients being treated with the blood purification device or accumulated data of multiple pieces of past information of the blood purification device as teacher data to predict the occurrence of an abnormality in a specific patient or an abnormality in the blood purification device and to propose an appropriate method of dealing with the abnormality;
a display control unit that displays, on a display unit of the blood purification apparatus, the occurrence of an abnormality in a specific patient or an abnormality in the blood purification apparatus predicted by the prediction processing unit and a proposal of an appropriate method of dealing with the abnormality;
A blood purification device equipped with the above . The blood purification device according to claim 1, wherein the staff position information acquired by the staff information acquisition unit is detected by a position detection unit provided in the mobile terminal and input in real time. The blood purification device according to claim 1, wherein the staff schedule information acquired by the staff information acquisition unit is input based on a schedule management unit that manages staff work information. The blood purification device according to claim 1, wherein the selection unit determines the staff members who can respond, including the staff member who should respond, in the learning model, and the priority order of the staff members who can respond, and the abnormality notification unit notifies the mobile device carried by the staff member who should respond of the abnormality of the patient or the abnormality of the blood purification device, and then notifies other staff members who can respond in order of priority depending on the response status of the staff member to the notification. The blood purification device according to claim 1, wherein the staff skill information acquired by the staff information acquisition unit is input based on the staff's expertise, skill, or length of experience or number of experiences. The blood purification apparatus of claim 1, wherein the prediction processing unit proposes a plurality of countermeasures for the predicted abnormality of a specific patient or the abnormality of the blood purification apparatus, and the display control unit causes the display unit to display the plurality of countermeasures proposed by the prediction processing unit.

Images