Program, computer, and information processing method

Abstract

To provide a new program, a computer, and an information processing method for increasing convenience for a medical practitioner.SOLUTION: Storage means 31 stores patient information 120 in a storage section 40. Reception means 32 periodically receives a patient's biological information 130 from a measuring instrument 2 at a predetermined interval. Display instruction means 33 transmits a display instruction signal to a biological information monitor 4 so as to display the received biological information 130. Prediction means 35 predicts a patient sudden change symptom from the patient's biological information 130 that the reception means 32 receives and the patient information 120 stored in the storage section 40 by using an algorithm 110 learned on the basis of teacher data 100 including patient information, biological information, and sudden change symptom information. Notification means 36 notifies a medical practitioner of information regarding a patient's predicted sudden change symptom.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a program, a computer, and an information processing method.

Background Art

[0002] Conventionally, in the medical field, in order for medical staff such as doctors and medical technicians to smoothly proceed with examinations and tests, etc., an integrated medical support device that shares the process and results of medical treatment, etc. among medical staff or between medical departments has been used. Here, Patent Documents 1 and 2 disclose a medical support device that can predict a sudden change in a patient and take countermeasures in advance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] When performing treatment such as surgery on a patient, if a sudden change in symptoms such as a sudden change in arterial oxygen saturation, a sudden change in blood pressure or pulse, or an impending cardiac arrest can be predicted and notified to medical staff before the sudden change symptoms occur, there is a possibility of preventing the patient's symptoms from deteriorating, but conventionally, such a device has not existed.

[0005] The present disclosure has been made in consideration of such points, and an object thereof is to provide a new program, computer, and information processing method that improve convenience for medical staff.

Means for Solving the Problems

[0006] The program of the present disclosure is A program that causes a computer to function as a storage means, a reception means, a display instruction means, a prediction means, and a notification means, The storage means stores patient information in a storage unit in advance, The reception means periodically receives biometric information of a patient from a measuring instrument at predetermined intervals, The display instruction means transmits a display instruction signal to the biometric information monitor so as to display the biometric information of the patient received by the reception means on the biometric information monitor, The prediction means uses an algorithm learned based on teacher data including patient information stored in the storage unit, biometric information of the patient, and critical symptom information of the patient, and the reception means periodically receives the biometric information of the patient at predetermined intervals and the patient information stored in the storage unit, predicts a critical symptom, The notification means notifies medical staff of information regarding the critical symptom of the patient predicted by the prediction means, The critical symptom includes a critical condition that requires admission to the intensive care unit, a sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, a sudden change in blood pressure or pulse that requires vasopressors, massive fluid infusion or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion or aortic occlusion, and the initiation of cardiopulmonary resuscitation or in-hospital death as a result of any of these.

[0007] In the program of the present disclosure, The critical symptom may be a sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management.

[0008] In this case, in the program of the present disclosure, It can be predicted that the sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management as the critical symptom will occur within the range of 30 minutes to 6 hours from when it is notified to the medical staff.

[0009] In the program of the present disclosure, The acute symptoms may be sudden changes in blood pressure or pulse rate that require vasopressors, massive fluid infusion, or blood transfusion.

[0010] In this case, in the program of the present disclosure, it is possible to predict that the sudden changes in blood pressure or pulse rate that require vasopressors, massive fluid infusion, or blood transfusion as the acute symptoms will occur within the range of 30 minutes to 6 hours from the time when the medical staff is notified.

[0011] In the program of the present disclosure, the acute symptoms may be the occurrence of disseminated intravascular coagulation that requires treatment.

[0012] In this case, in the program of the present disclosure, it is possible to predict that the disseminated intravascular coagulation that requires treatment as the acute symptoms will occur within the range of 12 hours to 48 hours from the time when the medical staff is notified.

[0013] In the program of the present disclosure, the acute symptoms may be impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion.

[0014] In this case, in the program of the present disclosure, it is possible to predict that the impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion as the acute symptoms will occur within the range of 1 minute to 10 minutes from the time when the medical staff is notified.

[0015] The program of the present disclosure further functions the computer as algorithm generation means, the algorithm generation means generates an algorithm by learning based on teacher data including patient information stored in the storage unit, biometric information of the patient, and acute symptom information of the patient, The prediction means may predict a sudden change symptom from the biological information of the patient periodically received by the reception means at a predetermined interval and the patient information stored in the storage unit, using the algorithm generated by the algorithm generation means.

[0016] In the program of the present disclosure, The reception means receives at least any one of vital signs, an electrocardiogram waveform, an arterial waveform diagram, and an end-tidal carbon dioxide concentration waveform as the biological information of the patient from the measuring instrument. The biological information of the patient included in the teacher data may be at least any one of vital signs, an electrocardiogram waveform, an arterial waveform diagram, and an end-tidal carbon dioxide concentration waveform.

[0017] In the program of the present disclosure, The patient information stored in the storage unit may be collected from an electronic medical record.

[0018] In the program of the present disclosure, The patient information stored in the storage unit may include information regarding the content of the treatment performed on the patient.

[0019] In the program of the present disclosure, The notification means may transmit a second display instruction signal to the biological information monitor so as to display information regarding the sudden change symptom of the patient predicted by the prediction means on the biological information monitor.

[0020] The computer of the present disclosure, is a computer that functions as a storage means, a reception means, a display instruction means, a prediction means, and a notification means by executing a program. The storage means stores patient information in a storage unit in advance. The reception means periodically receives the biological information of the patient from the measuring instrument at a predetermined interval. The display instruction means transmits a display instruction signal to the biological information monitor so as to display the biological information of the patient received by the reception means on the biological information monitor. The prediction means uses an algorithm learned based on teacher data including patient information stored in the storage unit, the patient's biological information, and the patient's acute change symptom information, and predicts acute change symptoms from the patient's biological information regularly received by the reception means at predetermined intervals and the patient information stored in the storage unit. The notification means notifies medical staff of information regarding the acute change symptoms of the patient predicted by the prediction means. The acute change symptoms include an acute change that requires admission to the intensive care unit, an acute change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, an acute change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion, and the initiation of cardiopulmonary resuscitation or in-hospital death as a result of any of these.

[0021] The information processing method of the present disclosure is an information processing method executed by a computer having a control unit, a step in which the control unit stores patient information in a storage unit in advance; a step in which the control unit regularly receives the patient's biological information from a measuring instrument at predetermined intervals; a step in which the control unit transmits a display instruction signal to the biological information monitor so as to display the received patient's biological information on the biological information monitor; a step in which the control unit predicts acute change symptoms from the patient's biological information regularly received at predetermined intervals and the patient information stored in the storage unit, using an algorithm learned based on teacher data including the patient information stored in the storage unit, the patient's biological information, and the patient's acute change symptom information; a step in which the control unit notifies medical staff of information regarding the predicted acute change symptoms of the patient; and includes The acute symptoms include an acute change that requires admission to the intensive care unit, an acute change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, an acute change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion, and the initiation of cardiopulmonary resuscitation or in-hospital death as a result of any of these, and are characterized by including such.

Advantages of the Invention

[0022] According to the present disclosure, a new program, computer, and information processing method that improve convenience for medical staff are provided.

Brief Description of the Drawings

[0023]

Figure 1

Figure 2

Figure 3

Figure 4

Modes for Carrying Out the Invention

[0024] Patients receiving treatment in a hospital (including general wards, intensive care units, and emergency departments) often experience a sudden change in their condition during the process. From previous reports, it has become clear that a delay in the initial response to an acute change has an adverse effect on the prognosis of the patient. However, there is no clear indicator for predicting such an acute change in advance, and it depends on the judgment based on the experience and knowledge of medical staff.

[0025] As a result of intensive studies on the above problems, the present inventor has generated an algorithm learned based on teacher data including patient information, the patient's biological information, and the patient's critical symptom information, and using this algorithm, it has been found that critical symptoms can be predicted (output) from the biological information and patient information (input) of patients in a medical facility.

[0026] In addition, based on this finding, the present inventor has conceived of assisting immediate and time-series monitoring by medical staff engaged in the treatment and management of patients, and improving the convenience for medical staff, thereby completing the present invention. Here, "immediate" means that the occurrence risk information of the outcome (critical symptom) calculated from the input data is immediately displayed via a biological monitor or the like, and medical staff engaged in the treatment and management of patients can recognize the critical risk at the treatment site. Also, "time-series" means that the occurrence risk information of the outcome calculated from the input data is continuously displayed via a biological monitor or the like, so that medical staff engaged in the treatment and management of patients can always monitor the critical risk. Note that biological information is collected by the biological information monitor in units of about 0.1 milliseconds (0.0001 seconds). It is normal for the critical risk to be updated using this biological information.

[0027] Hereinafter, embodiments of the present disclosure will be described, but the invention according to the present disclosure is not limited thereto. FIG. 1 is a diagram schematically showing the configuration of a prediction system according to an embodiment of the present disclosure. FIG. 2 is a diagram showing an exemplary flow of information processing in a prediction system and an information processing method according to an embodiment of the present disclosure. FIG. 3 is a diagram showing an exemplary flow of information processing in an information processing method according to an embodiment of the present disclosure. FIG. 4 is an exemplary diagram showing the relationship between critical symptoms (critical changes due to circulatory failure) and their parameters (biological information or integrated circulatory failure risk) in a prediction system and an information processing method according to an embodiment of the present disclosure.

[0028] [Prediction System 1] FIG. 1 shows a prediction system 1 according to the present disclosure. As shown in FIG. 1, the prediction system 1 includes a measuring device 2 that periodically measures a patient's biological information 130 at predetermined intervals, a computer (prediction device) 3 that periodically receives the patient's biological information 130 from the measuring device 2 at predetermined intervals, and a biological information monitor 4 that displays the patient's biological information 130 and the like received by the computer 3. The measuring device 2, the computer 3, and the biological information monitor 4 are communicably connected. Such a prediction system 1 supports immediate and time-dependent monitoring of a patient's critical symptoms and enables prompt medical intervention for the patient.

[0029] <Measuring device 2> Next, the configuration of the measuring device 2 will be described. In the prediction system 1, the measuring device 2 periodically measures the patient's biological information 130 at predetermined intervals. As the "predetermined interval", it can be measured every 0.1 milliseconds. As shown in FIG. 1, the measuring device 2 includes a measuring unit 21 and a communication unit 22. The measuring unit 21 measures the biological information 130, and the communication unit 22 transmits this biological information 130 to the computer 3.

[0030] The measuring unit 21 is not particularly limited as long as it can acquire predetermined biological information 130. One measuring unit 21 may be provided for one biological information 130, or a plurality of measuring units 21 may be provided in the measuring device 2. Also, in the prediction system 1, there may be a plurality of measuring devices 2.

[0031] The communication unit 22 includes a communication interface for transmitting and receiving signals to and from an external device wirelessly or by wire. The biological information 130 measured by the measuring unit 21 is transmitted by the communication unit 22 to the computer 3 (receiving means 32) (FIG. 2).

[0032] <Computer (prediction device) 3> The configuration of the computer 3 of the present disclosure will be described with reference to FIG. 1. The computer 3 of the present embodiment is composed of an industrial computer, a tablet terminal, etc. In the example shown in FIG. 1, it includes a control unit 30, a storage unit 40, a communication unit 50, a display unit 60, and an operation unit 70.

[0033] The control unit 30 is composed of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an AI inference device, etc., and controls the operation of the computer 3. Specifically, the control unit 30 functions as a storage means 31, a reception means 32, a display instruction means 33, an algorithm generation means 34, a prediction means 35, and a notification means 36 by executing a program stored in a storage unit 40 described later (Fig. 1). Each of these means will be described later.

[0034] The storage unit 40 is composed of, for example, an HDD (Hard Disk Drive), a RAM (Random Access Memory), a ROM (Read Only Memory), an SSD (Solid State Drive), etc. Further, the storage unit 40 is not limited to being built into the computer 3, and may be a storage medium (for example, a USB memory) that can be detachably attached to the computer 3. In the present embodiment, the storage unit 40 stores a program executed by the control unit 30, the generated algorithm 110, patient information 120, biological information 130, acute change symptom information, etc. Note that these programs, algorithms 110, various information such as patient information 120 and biological information 130 may be stored not in the storage unit 40 but in other storage means (such as a cloud server).

[0035] The communication unit 50 includes a communication interface for transmitting and receiving signals to and from an external device wirelessly or by wire. The control unit 30 transmits and receives signals to and from the measuring instrument 2 and the biological information monitor 4 through the communication unit 50.

[0036] The display unit 60 is, for example, a monitor or the like, and displays various screens by receiving a display command signal from the control unit 30. The operation unit 70 is, for example, a keyboard or the like, and an administrator of the computer 3 can give various commands to the control unit 30. In an embodiment, a display operation unit such as a touch panel in which these display unit 60 and operation unit 70 are integrated may be used. Further, in another embodiment, the computer 3 and the biological information monitor 4 may be integrated. In this case, the display unit 60 can also function as the biological information monitor 4.

[0037] (Details of the control unit 30) (Storage means 31) The storage means 31 causes the storage unit 40 to store the patient information 120 prior to the prediction of acute change symptoms. In this specification, "patient information" refers to specific information of a certain patient, and may include information collected from medical records such as electronic medical records and information regarding the content of treatment performed on the patient. Specific examples of patient information include patient basic information (gender, age, height, weight, etc.), presence or absence of disease and disease name, past history, allergy, blood sampling result, oral or injection drug (presence or absence, drug name), infusion information (type of infusion preparation, type of administered drug, administration rate, administration amount), blood transfusion information, culture test result, etc. Examples of the described content of the "medical record" include physical examination results, record information including evaluations by doctors and nurses, etc. Examples of the "treatment content" include artificial respiration, hemodialysis, peritoneal dialysis, extracorporeal circulation, etc.

[0038] (Receiving means 32) The receiving means 32 periodically receives the biological information 130 of the patient from the measuring instrument 2 at a predetermined interval (every 0.1 milliseconds). In this specification, "biological information" is not particularly limited as long as it can obtain acute change symptom information of the patient (including sudden changes in arterial oxygen saturation, sudden changes in blood pressure or pulse, or impending cardiac arrest, etc.). In particular, as biological information, it is preferable to measure vital signs (pulse, blood pressure, shock index (heart rate / systolic blood pressure), respiratory rate), electrocardiogram waveform, arterial pressure waveform, arterial oxygen saturation waveform, end-tidal carbon dioxide concentration waveform, etc. and use them for the prediction of acute change symptoms.

[0039] (Indicating means 33) The indicating means 33 transmits a display instruction signal to the biological information monitor 4 (communication unit 41) so that the biological information monitor 4 (display unit 42) displays the biological information 130 of the patient received by the receiving means 32 (Figs. 4(A) and (B)). Thereby, medical staff can grasp the individual biological information 130 of the patient over time.

[0040] (Algorithm generation means 34) The algorithm generation means 34 generates an algorithm 110 by learning based on teacher data 100 including patient information 120 stored in the storage unit 40, the biological information 130 of the patient, and the acute change symptom information of the patient. This learning is, for example, so-called supervised learning. The "patient" in the teacher data 100 may or may not include the patient for whom acute change symptoms are to be predicted. Also, the "teacher data 100" for generating the algorithm 110 used for predicting acute change symptoms may be data collected as an integrated database of past data on acute change symptoms in various parts of the world including Japan and anonymized. If necessary, preprocessing may be appropriately performed on the teacher data 100. Examples of "acute change symptoms" include an acute change that requires admission to the intensive care unit, an acute change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, an acute change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion, and the initiation of cardiopulmonary resuscitation or in-hospital death as a result of any of these. However, it is not necessarily limited to these. "Acute change symptoms" typically include an acute change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, an acute change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, and an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion.

[0041] Note that the algorithm 110 used in the computer 3 may be generated by an external device. In this case, the algorithm 110 generated by another device is transmitted to the computer 3 and stored in the storage unit 40 of this computer 3. Further, the algorithm 110 may be configured to function when a device other than the computer 3 executes a predetermined program.

[0042] Here, the "algorithm" generally means the processing procedure of a program executed to predict the acute change symptoms of a patient. For example, this algorithm also includes a so-called "model" that outputs a prediction result by inputting information such as patient information.

[0043] (Prediction means 35) The prediction means 35 uses the algorithm 110 generated by the algorithm generation means 34 to predict acute change symptoms from the biological information 130 of the patient that the reception means 32 periodically receives at a predetermined interval and the patient information 120 stored in the storage unit 40. When the algorithm 110 is generated by another device as described above, acute change symptoms can be predicted from the biological information 130 of the patient that the reception means 32 periodically receives at a predetermined interval and the patient information 120 stored in the storage unit 40 using the algorithm 110.

[0044] Furthermore, according to the present invention (particularly from the teacher data 100), it is possible to predict when a critical symptom will occur by the algorithm 110. For example, a critical cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion as a critical symptom can be predicted to occur within the range of 1 minute to 10 minutes from when it is notified to medical staff (i.e., from the arithmetic processing by the algorithm 110. The same applies hereinafter). Similarly, it can be considered that the initiation of cardiopulmonary resuscitation for these cardiac arrests and the measures of vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion are required within the range of 1 minute to 10 minutes after notification. Also, a sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management as a critical symptom can be predicted to occur within the range of 30 minutes to 6 hours from when it is notified to medical staff. Similarly, it can be considered that oxygen administration or artificial respiration management due to a sudden change in arterial oxygen saturation is required within the range of 30 minutes to 6 hours after notification. Also, a sudden change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion as a critical symptom can be predicted to occur within the range of 30 minutes to 6 hours from when it is notified to medical staff. Similarly, it can be considered that the initiation or increase of vasopressor administration and the initiation of massive fluid infusion or blood transfusion due to a sudden change in blood pressure or pulse are required within the range of 30 minutes to 6 hours after notification. Also, disseminated intravascular coagulation that requires treatment as a critical symptom can be predicted to occur within the range of 12 hours to 48 hours from when it is notified to medical staff. Furthermore, it can be predicted that an unexpected admission to the intensive care unit as a critical symptom should be carried out within the range of 12 hours to 48 hours from when it is notified to medical staff, and in-hospital death as a critical symptom can be predicted to occur within the range of 1 day to 7 days from when it is notified to medical staff.

[0045] As an output of the prediction result, if the risk can be quantified by integrating the circulatory dynamics parameters, it is suitable for medical staff to grasp the prediction result immediately and over time. That is, as shown in the examples of FIGS. 4(A) and (B), simply displaying the biological information 130 individually on the biological information monitor 4 in the ICU, even if the biological information 130 is collected in units of 0.1 milliseconds, it is difficult for medical staff to integrate information such as blood tests and treatment interventions and immediately (in seconds) grasp the deterioration of the circulatory dynamics. In contrast, as shown in the example of FIG. 4(C), if the integrated circulatory failure risk can be utilized and the algorithm 110 can be learned to show the risk value of rapid change as the prediction result, it will be easier for the medical staff in the ICU to grasp the prediction result. That is, if the algorithm 110 can integrate the patient information 120 and the biological information 130 of a certain patient and output the integrated circulatory failure risk, which is one factor (parameter), the deterioration of the circulatory dynamics can be easily recognized. Also, if the contribution degree (ratio) of each individual factor important for prediction can be displayed, medical staff can immediately grasp the factors of rapid change. In the example of FIG. 4(C), a blood test has been performed at the time of the arrow part. The lactic acid value (patient information) is high in this blood test, which is the main factor for the increase in the integrated circulatory failure risk (the value is 0.8), and it is shown that the rapid change of the patient can be predicted prior to the range where the deterioration of the circulatory dynamics accompanied by a decrease in blood pressure is recognized (the shaded part in the center of FIG. 4(B)). Note that the "time from ICU admission" in FIG. 4 is in minutes.

[0046] (Notification means 36) The notification means 36 notifies medical staff of information regarding the predicted rapid change symptoms of the patient by the prediction means 35. As a "notification" method, when rapid change symptoms are predicted, or when a predetermined numerical value is obtained as the prediction result, a buzzer sound or the like may be output, the light may be blinked, or the sound and light may be output simultaneously. Also, the notification means 36 may transmit a second display instruction signal to the biological information monitor 4 to display on the biological information monitor 4 information regarding the predicted rapid change symptoms of the patient by the prediction means 35.

[0047] In addition to the prediction system 1, computer (prediction device) 3, and program according to the present disclosure, the integrated database serving as the teacher data 100 and the algorithm 110 can also assist medical personnel in immediate and time-series monitoring, and improve the convenience for medical staff.

[0048] [Information Processing Method] Next, an information processing method by such a computer 3 (prediction system 1) will be described. Components with the same reference numerals are the same as the above-described components, and duplicate descriptions will be omitted as appropriate. The following processing is performed by the control unit 30 executing the program stored in the storage unit 40.

[0049] First, the control unit 30 (storage means 31) stores the patient information 120 in the storage unit 40 (FIG. 3, step S1). The storage unit 40 may also store other patient information that serves as a basis for generating the algorithm 110.

[0050] Next, the control unit 30 (reception means 32) periodically receives the biological information 130 of the patient from the measuring instrument 2 at a predetermined interval (FIG. 3, step S2).

[0051] Next, the control unit 30 (display instruction means 33) transmits a display instruction signal to the biological information monitor 4 so as to display the received biological information 130 of the patient on the biological information monitor 4 (FIG. 3, step S3).

[0052] Next, the control unit 30 (algorithm generation means 34) generates the algorithm 110 through learning based on the teacher data 100 including the patient information stored in the storage unit 40, the biological information of the patient, and the acute change symptom information of the patient (FIG. 3, step S4). Note that the acute change symptom is as described above.

[0053] Next, the control unit 30 (prediction means 35) predicts acute change symptoms from the biological information 130 of the patient received periodically at a predetermined interval and the patient information 120 stored in the storage unit 40 using the generated algorithm 110 (FIG. 3, step S5).

[0054] Next, the control unit 30 (notification means 36) notifies medical staff of information regarding the predicted acute change symptoms of the patient (FIG. 3, step S6). Here, in order to notify the medical staff, the control unit 30 (notification means 36) may display information regarding the predicted acute change symptoms of the patient by the prediction means 35 on the biological information monitor 4.

[0055] In addition, the prediction accuracy can be improved with more patient information 120 and biological information 130. Therefore, the algorithm 110 may be further learned with this information. Also, the superiority or inferiority of the algorithm 110 may be confirmed using a validation data set for verification or the like, and the algorithm 110 may be updated as appropriate.

[0056] In the program, computer 3, and information processing method of the present embodiment having the above-described configuration, by executing the program, computer 3 functions as a storage unit 31, a reception unit 32, a display instruction unit 33, a prediction unit 35, and a notification unit 36. The storage unit 31 stores patient information 120 in the storage unit 40 in advance. The reception unit 32 periodically receives biometric information 130 of the patient from the measuring device 2 at a predetermined interval. The display instruction unit 33 transmits a display instruction signal to the biometric information monitor 4 so as to display the biometric information 130 of the patient received by the reception unit 32 on the biometric information monitor 4. The prediction unit 35 uses an algorithm 110 learned based on teacher data 100 including patient information stored in the storage unit 40, the patient's biometric information, and the patient's critical symptom information, and the biometric information 130 of the patient periodically received by the reception unit 32 at a predetermined interval and the patient information 120 stored in the storage unit 40 to predict critical symptoms. The notification unit 36 notifies medical staff of information regarding the critical symptoms of the patient predicted by the prediction unit 35. The critical symptoms include a critical change that requires admission to the intensive care unit, a sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management, a sudden change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion, the occurrence of disseminated intravascular coagulation that requires treatment, an impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion, and the initiation of cardiopulmonary resuscitation or in-hospital death as a result of any of these. According to the program, computer 3, and information processing method of the present embodiment as described above, immediate and time-dependent monitoring can be supported so as to cope with critical symptoms including a sudden change in the patient's arterial oxygen saturation, a sudden change in blood pressure or pulse, or an impending cardiac arrest, and the convenience for medical staff can be improved. It is also possible to immediately notify the prediction result by this AI.

[0057] More specifically, the program, computer 3, and information processing method of the present embodiment assist in predicting the risk of sudden changes in general ward patients, ICU patients, and emergency department patients. For general ward patients, it is possible to predict sudden changes that require admission to the ICU and prompt medical staff to consider the necessity of oxygen administration and artificial respiration management accompanying a decrease in oxygen saturation, the necessity of vasopressors, massive fluid infusion, and blood transfusion accompanying a decrease in blood pressure, etc. For ICU patients, it is possible to predict sudden changes that require treatment intervention and prompt medical staff to consider the necessity of oxygen administration and artificial respiration management accompanying a decrease in oxygen saturation, the necessity of vasopressors, massive fluid infusion, and blood transfusion accompanying a sudden or persistent decrease in blood pressure, etc. Furthermore, for ICU patients, it is possible to predict the occurrence of disseminated intravascular coagulation that requires treatment (decrease in platelet count, prolongation of prothrombin time, increase in fibrin / fibrinogen degradation products) and prompt medical staff to consider countermeasures. For emergency department patients, it is possible to predict reversible cardiac arrest and prompt medical staff to consider countermeasures. One of the points of the present invention is not only to predict an element of physiological states such as hemodynamics and respiratory states, but also to predict clinically important sudden changes. By predicting the occurrence of important future patient sudden changes and having medical staff involved in the treatment and management of the patient monitor the occurrence risk immediately and over time, it is possible to support the decision-making of medical staff that directly leads to an improvement in the patient's prognosis.

[0058] Also, in the program, computer 3, and information processing method of the present embodiment, it can be predicted that a sudden change in arterial oxygen saturation that requires oxygen administration or artificial respiration management as a critical symptom will occur within the range of 30 minutes to 6 hours from when it is notified to medical staff. It can be predicted that a sudden change in blood pressure or pulse that requires vasopressors, massive fluid infusion, or blood transfusion as a critical symptom will occur within the range of 30 minutes to 6 hours from when it is notified to medical staff. It can be predicted that disseminated intravascular coagulation that requires treatment as a critical symptom will occur within the range of 12 hours to 48 hours from when it is notified to medical staff. It can be predicted that impending cardiac arrest or reversible cardiac arrest that requires vasopressors, massive fluid infusion, blood transfusion, or aortic occlusion as a critical symptom will occur within the range of 1 minute to 10 minutes from when it is notified to medical staff. Thus, since the occurrence time of critical symptoms can be predicted, the convenience for medical staff can be improved.

[0059] Also, in the program, computer 3, and information processing method of the present embodiment, by executing the program, computer 3 (control unit 30) may further function as algorithm generation means 34, and algorithm generation means 34 generates algorithm 110 through learning based on teacher data 100 including patient information stored in storage unit 40, the patient's biological information, and the patient's critical symptom information. Prediction means 35 may predict critical symptoms from the biological information 130 of the patient that reception means 32 periodically receives at a predetermined interval and the patient information 120 stored in storage unit 40 using the algorithm generated by algorithm generation means 34. Thus, algorithm 110 may be generated within computer 3, and the administrator of computer 3 can appropriately adjust teacher data 100 according to the symptoms.

[0060] Further, in the program, computer 3, and information processing method of the present embodiment, the reception means 32 receives at least any one of vital signs, electrocardiogram waveforms, arterial waveform diagrams, and end-tidal carbon dioxide concentration waveforms as the biological information 130 of the patient from the measuring instrument 2, and the biological information of the patient included in the teacher data 100 may also be at least any one of vital signs, electrocardiogram waveforms, arterial waveform diagrams, and end-tidal carbon dioxide concentration waveforms. With such biological information 130, sudden change symptoms can be predicted particularly accurately.

[0061] Further, in the program, computer 3, and information processing method of the present embodiment, the patient information 120 stored in the storage unit 40 may be collected from an electronic medical record and may also include information regarding the content of the treatment performed on the patient. In this way, by using various patient information 120 and biological information 130, the prediction accuracy can be further improved. There are reports that predict the occurrence of hemodynamics and events using each piece of information as an input. However, there are no known reports that predict the occurrence of clinically important sudden changes by using all these inputs in combination.

[0062] Further, in the program, computer 3, and information processing method of the present embodiment, the notification means 36 may transmit a second display instruction signal to the biological information monitor 4 to cause the biological information monitor 4 to display information regarding the sudden change symptoms of the patient predicted by the prediction means 35. Thereby, medical staff can easily grasp the sudden change symptom information in addition to the biological information 130.

[0063] Note that the program, computer 3, and information processing method according to the present embodiment are not limited to the modes and combinations as described above, and various changes can be made.

Explanation of Signs

[0064] 1 Prediction system 2 Measuring instrument 21 Measuring unit 22 Communication unit 3 Computer 30 Control Unit 31 Memory Means 32 Reception Means 33 Display Instruction Means 34 Algorithm Generation Means 35 Prediction Means 36 Notification Means 40 Memory Section 50 Communication Section 60 Display Section 70 Operation Section 4 Biosignal Monitor 41 Communication Section 42 Display Section 100 Teacher Data 110 Algorithm 120 Patient Information 130 Biosignal

Claims

1. A program for causing a computer to function as a storage means, a receiving means, a display instruction means, a prediction means, and a notification means, The storage means stores patient information in a storage unit in advance, the receiving means periodically receives the biological information of the patient from the measuring device at a predetermined interval; the display instruction means transmits a display instruction signal to the vital sign monitor to cause the vital sign monitor to display the vital sign of the patient accepted by the acceptance means; The prediction means predicts a sudden symptom from the patient's biological information periodically received at a predetermined interval by the reception means and the patient information stored in the storage unit, using an algorithm learned based on teacher data including the patient information stored in the storage unit, the patient's biological information, and the patient's sudden symptom information; The notification means notifies a medical professional of information regarding the sudden change in symptoms of the patient predicted by the prediction means, The above-mentioned sudden changes in symptoms include a sudden change requiring admission to an intensive care unit, a sudden change in arterial oxygen saturation requiring oxygen administration or artificial respiration management, a sudden change in blood pressure or pulse requiring vasopressors, massive fluid infusion or blood transfusion, the occurrence of disseminated intravascular coagulation requiring treatment, impending or reversible cardiac arrest requiring vasopressors, massive fluid infusion, blood transfusion or aortic cross-clamping, and the initiation of cardiopulmonary resuscitation or death in hospital as a result of any of these, in the program.

2. The program according to claim 1 , wherein the sudden symptom is a sudden change in arterial blood oxygen saturation that requires oxygen administration or artificial respiration management.

3. The program of claim 2, which predicts that a sudden change in arterial blood oxygen saturation that requires oxygen administration or artificial respiration management as an acute symptom will occur within a range of 30 minutes to 6 hours from the time the medical professional is notified.

4. The program according to claim 1 , wherein the sudden symptom is a sudden change in blood pressure or pulse rate that requires the use of a vasopressor, large volume fluid infusion or blood transfusion.

5. The program of claim 4, which predicts that a sudden change in blood pressure or pulse rate requiring a vasopressor, large volume fluid infusion or blood transfusion as the sudden symptom will occur within a range of 30 minutes to 6 hours from the time the medical professional is notified.

6. The program according to claim 1 , wherein the acute symptom is the occurrence of disseminated intravascular coagulation requiring treatment.

7. The program according to claim 6, which predicts that disseminated intravascular coagulation requiring treatment as the sudden symptom will occur within a range of 12 to 48 hours from the time the medical professional is notified.

8. The program according to claim 1, wherein the sudden change in symptoms is impending cardiac arrest or reversible cardiac arrest requiring the administration of a vasopressor, massive fluid infusion, blood transfusion or aortic clamping.

9. The program of claim 8, which predicts that impending cardiac arrest or reversible cardiac arrest requiring the administration of a vasopressor, massive fluid infusion, blood transfusion or aortic clamping as the sudden symptom will occur within 1 to 10 minutes from the time the medical professional is notified.

10. The computer is further caused to function as an algorithm generating means, The algorithm generating means generates an algorithm by learning based on teacher data including patient information, biological information of the patient, and information on a sudden symptom of the patient stored in the storage unit; 2. The program according to claim 1, wherein the prediction means predicts a sudden change in symptoms from the patient's biological information periodically received by the reception means at predetermined intervals and the patient information stored in the memory unit, using the algorithm generated by the algorithm generation means.

11. the receiving means receives at least one of vital signs, an electrocardiogram waveform, an arterial waveform diagram, and an end-tidal carbon dioxide concentration waveform as biological information of the patient from the measuring device; 2. The program according to claim 1, wherein the patient's biological information included in the teaching data is at least one of vital signs, an electrocardiogram waveform, an arterial waveform diagram, and an end-tidal carbon dioxide concentration waveform.

12. 2. The program according to claim 1, wherein the patient information stored in the storage unit is collected from an electronic medical record.

13. 2. The program according to claim 1, wherein the patient information stored in the storage unit includes information regarding the content of treatment to be performed on the patient.

14. 2. The program according to claim 1, wherein the notification means transmits a second display instruction signal to the biological information monitor to cause the biological information monitor to display information regarding the patient's sudden symptom predicted by the prediction means.

15. A computer that functions as a storage means, a receiving means, a display instruction means, a prediction means, and a notification means by executing a program, The storage means stores patient information in a storage unit in advance, the receiving means periodically receives the biological information of the patient from the measuring device at a predetermined interval; the display instruction means transmits a display instruction signal to the vital sign monitor to cause the vital sign monitor to display the vital sign of the patient accepted by the acceptance means; The prediction means predicts a sudden symptom from the patient's biological information periodically received at a predetermined interval by the reception means and the patient information stored in the storage unit, using an algorithm learned based on teacher data including the patient information stored in the storage unit, the patient's biological information, and the patient's sudden symptom information; The notification means notifies a medical professional of information regarding the sudden change in symptoms of the patient predicted by the prediction means, The sudden change in symptoms includes a sudden change requiring admission to an intensive care unit, a sudden change in arterial blood oxygen saturation requiring oxygen administration or artificial respiration management, a sudden change in blood pressure or pulse requiring vasopressors, massive fluid infusion or blood transfusion, the occurrence of disseminated intravascular coagulation requiring treatment, impending or reversible cardiac arrest requiring vasopressors, massive fluid infusion, blood transfusion or aortic cross-clamping, and the initiation of cardiopulmonary resuscitation or death in hospital as a result of any of these, as detected by the computer.

16. An information processing method executed by a computer having a control unit, A step of causing the control unit to previously store patient information in a storage unit; a step of the control unit periodically receiving biological information of a patient from a measuring device at a predetermined interval; a step of transmitting a display instruction signal to the vital sign monitor by the control unit to cause the vital sign monitor to display the received vital sign of the patient; The control unit predicts a sudden symptom from the patient's biological information periodically received at a predetermined interval and the patient information stored in the storage unit, using an algorithm learned based on teacher data including the patient information stored in the storage unit, the patient's biological information, and the patient's sudden symptom information; The control unit notifies a medical professional of information regarding the predicted sudden change in the patient's condition; Equipped with The information processing method includes the sudden change in symptoms requiring admission to an intensive care unit, a sudden change in arterial blood oxygen saturation requiring oxygen administration or artificial respiration management, a sudden change in blood pressure or pulse rate requiring vasopressors, massive fluid infusion or blood transfusion, the occurrence of disseminated intravascular coagulation requiring treatment, imminent cardiac arrest or reversible cardiac arrest requiring vasopressors, massive fluid infusion, blood transfusion or aortic cross-clamping, and the initiation of cardiopulmonary resuscitation or death in hospital as a result of any of these.

Images