JP2021111373A - Emergency transfer support system - Google Patents

Abstract

To provide an emergency transfer support system used for an emergency transfer.SOLUTION: The emergency transfer support system used for an emergency transfer includes: an input information reception processing unit for receiving input information including symptoms of a patient from an emergency team terminal used by an emergency team; a prediction processing unit for calculating the possibility of hospitalization and/or the emergency degree of an emergency transfer on the basis of a part or all of the input information; and an output processing unit for outputting the possibility of hospitalization of the patient calculated by the prediction processing unit and/or the emergency degree of an emergency transfer.SELECTED DRAWING: Figure 1

Description

The present invention relates to an emergency transport support system used for emergency transport and the like.

When a patient is transported by ambulance, an ambulance crew will search for a medical institution to accept the patient while listening to the patient's condition. In this case, the medical institution will be asked whether it is acceptable or not by telling the patient's symptoms and condition by telephone or the like.

On the other hand, the medical institution that receives the request for acceptance of emergency transportation will judge whether or not it is acceptable at the medical institution and will reply to the ambulance crew whether or not it is acceptable. In this case, the acceptance is answered by comprehensively considering the number of vacant beds at the medical institution, the urgency of emergency transportation of the patient, and the possibility of the patient's expected illness. The current state of the emergency medical care system is disclosed in Non-Patent Document 1 and the like below.

Ministry of Health, Labor and Welfare, "Current Status and Issues of Emergency Medical System", [online], Internet <URL: https://www.mhlw.go.jp/content/10802000/000328610.pdf>

Conventionally, medical institutions that have received requests for acceptance of emergency transport include the number of vacant beds mentioned above, the degree of urgency of patients, the possibility of illness expected for patients, etc., based on the contents heard by telephone from the ambulance crew. Will be judged and the acceptance or rejection will be answered. However, the patient's symptoms and conditions are based on the experience of paramedics who are not doctors and are not specialized.

As a medical institution, whether a patient who accepts emergency transportation can return home or be hospitalized is a very important issue because it depends on the number of vacant beds. Therefore, the possibility of hospitalization of a patient is important from the viewpoint of deciding whether or not to request acceptance, but it may be delicate whether a patient who has been transported by emergency will be hospitalized, and the accuracy is based only on information from emergency personnel who are not doctors. It is difficult to make a high judgment.

In addition, regardless of the number of vacant beds, if the patient has a high degree of urgency, it may be life-threatening, so it is equally important from the viewpoint of determining whether or not to accept the request. For example, even if the medical institution does not have an empty bed, it is possible to accept a patient with a high degree of urgency, perform the treatment, and then transfer the patient to another medical institution. ..

In addition, some diseases such as stroke, myocardial infarction, and sepsis are particularly specialized, and if the patient is likely to have those diseases, early treatment by a specialist improves the patient's prognosis. It is preferable because it increases the possibility of doing so. However, medical institutions do not always have specialists and may be absent. Therefore, the possibility of highly specialized diseases is also an important point from the viewpoint of whether or not to accept emergency transportation.

As described above, in the conventional emergency medical care as in Non-Patent Document 1, the medical institution must judge whether or not to accept the emergency transportation based on uncertain information such as the symptoms transmitted by the ambulance crew. Therefore, as a medical institution, it is required to obtain objective information, but such a thing has not been done yet at this time.

Therefore, in view of the above problems, the present inventors have invented an emergency transport support system capable of transmitting objective information to a medical institution.

The first invention is an emergency transportation support system used for emergency transportation, which includes an input information reception processing unit that receives input information including a patient's symptom from an emergency services terminal used by the emergency services, and a part of the input information or a part of the input information. The predictive processing unit that calculates the possibility of hospitalization and / or the urgency of emergency transportation of the patient based on all of them, and the possibility of hospitalization and / or the urgency of emergency transportation of the patient calculated by the prediction processing unit. It is an emergency transport support system that has an output processing unit that outputs.

By using the emergency transport support system of the present invention, the possibility of hospitalization and the degree of urgency of the patient to be transported by emergency can be output as objective information, and these can be transmitted to the medical institution. As a result, the medical institution can make a judgment based on objective information. For example, for patients who are unlikely to be hospitalized, even if there is no vacant bed, they will accept emergency transportation or if there is no vacant bed. Even if there is a high degree of urgency, it is conceivable to take measures such as accepting the patient once, treating it, and then transferring the patient to another medical institution. In this way, by using the present invention, it is possible to improve the efficiency of finite emergency medical resources.

In the above-described invention, the prediction processing unit calculates the hospitalization possibility and / or urgency of the patient by using the information of the patient's symptom, vital signs and medical history in the input information, and is an emergency transportation support system. It can be configured as follows.

Various factors can be used to calculate the hospitalization possibility and urgency of a patient, but it can be calculated accurately by considering factors such as those of the present invention.

In the above-described invention, the prediction processing unit can be configured like an emergency transport support system that calculates the possibility of a predetermined disease based on a part or all of the input information.

It is also preferable for the prediction processing department to calculate the possibility of hospitalization, the degree of urgency, and the possibility of a predetermined disease. As a result, it is preferable for a specialist to deal with a highly specialized disease, but the medical institution can decide whether or not to accept emergency transportation in consideration of the presence or absence of the disease.

In the above-mentioned invention, it can be configured as an emergency transport support system including any one or more of stroke, myocardial infarction, and sepsis as the predetermined diseases.

As a highly specialized disease, there is a disease like the present invention. These are particularly highly specialized, and it is preferable that they be handled by a specialist. Early treatment is also a disease that may improve the prognosis of patients.

In the above-described invention, the output processing unit can be configured like an emergency transportation support system that sends the calculation result in the prediction processing unit to the emergency services terminal and / or the computer of the medical institution.

The calculation result in the prediction processing unit should be sent to the emergency services terminal or the computer of the medical institution. In particular, by sending to the computer of a medical institution, it is not necessary to intervene in the ambulance crew, so that the ambulance crew can concentrate on the treatment of the patient and prevent miscommunication of information.

In the above-described invention, the emergency transport support system identifies a medical institution that can handle emergency transport in the vicinity based on the position information of the emergency services or the emergency services terminal, and the specified one or more medical institutions. It can be configured as an emergency transportation support system that sends a request for acceptance of emergency transportation to the computer and the result of calculation by the prediction processing unit by the output processing unit.

The result of the calculation in the above-mentioned prediction processing unit may be performed together with the request for acceptance of the emergency transport. At that time, the burden on the ambulance crew can be reduced by automatically identifying the surrounding medical institutions.

In the above-described invention, when the emergency transport support system receives a notification of acceptance from the computer of the medical institution that sent the acceptance request, the reception request is sent to the computer of the other medical institution that sent the acceptance request. It can be configured like an emergency transport support system that sends completion information and prevents acceptance notifications.

When request for acceptance of emergency transportation is sent to multiple medical institutions, there is a possibility that multiple medical institutions will notify the acceptance. That would be inefficient. Therefore, when one medical institution gives an acceptance notice, the inefficiency can be eliminated by configuring the other medical institutions so that the acceptance notice cannot be sent.

In the above-described invention, the output processing unit causes the emergency services terminal or the computer of the medical institution to highlight when the calculation result in the prediction processing unit is equal to or higher than a predetermined value or a predetermined rank or higher. It can be configured like a support system.

By highlighting as in the present invention, it becomes easier for emergency services and medical institutions to recognize that there is a high possibility of hospitalization and that the degree of urgency is high.

In the above-described invention, when the calculation result of the possibility of the disease is equal to or higher than the predetermined threshold value, the emergency services terminal has no input item among the input items corresponding to the calculation of the possibility of the disease. It can be configured like an emergency transport support system that prompts for input.

If there are input items that have not been entered, it cannot be said that the calculated possibility of the disease is highly accurate. Therefore, as in the present invention, by prompting input for an input item that has not been input and inputting the input item, the accuracy can be improved.

In the above-described invention, the ambulance crew terminal highlights the non-input items, additionally displays the non-input items as input items, and superimposes the non-input items as a display prompting the input of the non-input items. It can be configured like an emergency transport support system, which is performed by either displaying or switching the display to the display position of uninput items.

There are various processing methods for prompting input, but it is better to execute as in the present invention.

In the above-described invention, the ambulance crew terminal is used to calculate the possibility of the disease from the first input screen when the calculation result of the possibility of the disease is equal to or more than a predetermined threshold value. It can be configured like an emergency transport support system that can switch to a second input screen that includes input items.

In the above-described invention, the ambulance crew terminal is used to calculate the possibility of the disease from the first input screen when the calculation result of the possibility of the disease is equal to or more than a predetermined threshold value. It can be configured like an emergency transport support system that automatically switches the display to a second input screen that includes input items.

Normally, input is performed from the first input screen where general input is performed, and when the possibility of a predetermined disease increases, input items corresponding to the disease are used to improve the calculation accuracy of the possibility of the disease. Input can be prompted by switching to the second input screen including.

In the above-described invention, the non-input item can be configured like an emergency transport support system specified by using an input item corresponding to the non-input parameter among the parameters used for calculating the possibility of the disease. can.

There are various methods for specifying uninput parameters, but by specifying the uninput items based on the parameters used to calculate the disease possibility as in the present invention, the accuracy of the disease possibility is accurate. It can lead to improvement.

In the above-described invention, the emergency transport support system has a meteorological information processing unit that acquires meteorological information, and the prediction processing unit uses a part or all of the meteorological information to hospitalize the patient. It can be configured like an emergency transport support system that calculates the likelihood and / or the urgency of emergency transport.

A fifteenth invention is an emergency transport support system used for emergency transport, which includes an input information reception processing unit that receives input information including a patient's symptom from an emergency services terminal used by the emergency services, and a part of the input information or a part of the input information. Calculated by the prediction processing unit and the prediction processing unit that calculates one or more of the possibility of the patient's disease, the possibility of hospitalization, or the urgency of emergency transportation using all and part or all of the weather information. It is an emergency transport support system that has an output processing unit that outputs the results.

In the emergency transport support system of the present invention, the prediction accuracy can be further improved by using the weather information in the prediction process in addition to the input information including the patient's symptom input by the emergency services. Then, as in the first invention, the ambulance crew can obtain the prediction result with improved prediction accuracy and can transmit it to the medical institution. As a result, the medical institution can make a judgment based on objective information. For example, for patients who are unlikely to be hospitalized, even if there is no vacant bed, they will accept emergency transportation or if there is no vacant bed. Even if there is a high degree of urgency, it is conceivable to take measures such as accepting the patient once, treating it, and then transferring the patient to another medical institution. In this way, by using the present invention, it is possible to improve the efficiency of finite emergency medical resources.

In the invention described above, the predictive processing unit can be configured like an emergency transport support system that calculates the likelihood of the patient's disease subcategory.

The seventeenth invention is an ambulance crew terminal used by an ambulance crew for emergency transportation, which accepts input of input information including a patient's symptom, and hospitalization of the patient calculated based on a part or all of the input information. It is an ambulance crew terminal that displays the possibility of and / or the urgency of ambulance transportation.

Similar to the first invention, by using the ambulance crew terminal of the present invention, the ambulance crew can obtain the possibility of hospitalization and the degree of urgency of the patient to be transported by emergency as objective information. Can be communicated to the institution. As a result, the medical institution can make a judgment based on objective information. For example, for patients who are unlikely to be hospitalized, even if there is no vacant bed, they will accept emergency transportation or if there is no vacant bed. Even if there is a high degree of urgency, it is conceivable to take measures such as accepting the patient once, treating it, and then transferring the patient to another medical institution. In this way, by using the present invention, it is possible to improve the efficiency of finite emergency medical resources.

The eighteenth invention is an ambulance crew terminal used by an ambulance crew for emergency transportation, which accepts input of input information including a patient's symptom and is a predetermined disease calculated based on a part or all of the input information. When the possibility of the disease is equal to or higher than a predetermined threshold, it is an ambulance crew terminal that prompts the input of the uninput items among the input items corresponding to the calculation of the possibility of the disease.

If there are input items that have not been entered, it cannot be said that the accuracy of the possibility of the disease obtained by the ambulance crew terminal is high. However, although the ambulance crew is a medical worker, they are not medical specialists, so it may not be possible to know which input items should be given priority to obtain highly accurate information on the possibility of the disease. Therefore, by using the ambulance crew terminal of the present invention, it is possible to make the ambulance crew recognize which input item should be preferentially input among the input items that have not been input, and as a result, the possibility of illness is possible. It leads to obtaining highly accurate information about sex.

The nineteenth invention is an ambulance crew terminal used by an ambulance crew for emergency transportation, which accepts input of input information including a patient's symptom, and includes a part or all of the input information and a part or all of weather information. It is an ambulance crew terminal that displays one or more of the possibility of the patient's illness, the possibility of hospitalization, or the urgency of emergency transportation calculated using.

Similar to the fifteenth invention, by using the ambulance crew terminal of the present invention, the ambulance crew can obtain a prediction result with improved prediction accuracy and can transmit it to a medical institution. As a result, the medical institution can make a judgment based on objective information. For example, for patients who are unlikely to be hospitalized, even if there is no vacant bed, they will accept emergency transportation or if there is no vacant bed. Even if there is a high degree of urgency, it is conceivable to take measures such as accepting the patient once, treating it, and then transferring the patient to another medical institution. In this way, by using the present invention, it is possible to improve the efficiency of finite emergency medical resources.

The emergency transport support system of the first invention can be realized by loading and executing the program of the present invention in a computer. That is, the computer is used as an input information reception processing unit that receives input information including the patient's symptoms from the emergency services terminal used by the emergency services, and the possibility of hospitalization of the patient and / or based on a part or all of the input information. Like an emergency transport support program that functions as a predictive processing unit that calculates the urgency of emergency transport, an output processing unit that outputs the possibility of hospitalization of the patient and / or the urgency of emergency transport calculated by the predictive processing unit. Can be configured in.

The emergency services terminal of the seventeenth invention can be realized by loading and executing the program of the present invention in a computer. That is, the ambulance crew terminal used by the ambulance crew for emergency transport is a processing unit that accepts input of input information including the patient's symptoms, the possibility of hospitalization of the patient calculated based on a part or all of the input information, and / Or a computer program that functions as a processing unit that displays the urgency of emergency transportation.

The emergency services terminal of the eighteenth invention can be realized by loading and executing the program of the present invention in a computer. That is, there is a possibility that the emergency services terminal used by the emergency services for emergency transportation is a processing unit that accepts input of input information including the patient's symptoms, and a predetermined disease calculated based on a part or all of the input information. When the value is equal to or higher than a predetermined threshold, it is a computer program that functions as a processing unit that prompts input for uninput items among the input items corresponding to the calculation of the possibility of the disease.

The ambulance crew terminal of the nineteenth invention can be realized by loading and executing the program of the present invention in a computer. That is, the ambulance crew terminal used by the ambulance crew for emergency transport is calculated using a processing unit that accepts input of input information including the patient's symptoms, a part or all of the input information, and a part or all of the weather information. A computer program that functions as a processing unit that displays one or more of the patient's illness potential, hospitalization potential, and emergency transport urgency.

By using the emergency transport support system of the present invention, objective information can be transmitted to a medical institution. Therefore, even in medical institutions, it is possible to obtain objective information on the diseases of patients to be transported, the possibility of hospitalization, the degree of urgency, and the like. This will lead to the optimization of emergency medical care.

It is a conceptual diagram which shows typically an example of the whole structure of the emergency transport support system of this invention. It is a figure which shows typically an example of the hardware composition of the computer used in the emergency transport support system of this invention. It is a flowchart which shows an example of the whole processing process in the emergency transport support system of this invention. It is a figure which shows an example of the input screen of vital signs. It is a figure which shows an example of the input screen of an electrocardiogram, a symptom, an inquiry item, and an observation item. It is a figure which shows an example of the input screen of the medical history. It is a figure which shows the part of the input screen of the treatment content. It is a figure which shows an example of the mathematical formula which calculates the possibility of a stroke in a prediction processing part. It is a figure which shows an example of the mathematical formula which calculates the possibility of myocardial infarction in a prediction processing part. It is a figure which shows an example of the mathematical formula which calculates the possibility of sepsis in a prediction processing part. It is a figure which shows an example of the formula which calculates the possibility of hospitalization in a prediction processing part. It is a figure which shows an example of the formula which calculates the possibility of urgency in a prediction processing part. It is a figure which shows an example of the screen which outputs the result in the prediction processing unit. It is a figure which shows an example of the non-input item and the screen when highlighting it in Example 2. FIG. It is a figure which shows an example of the non-input item and the screen when highlighting it in Example 2. FIG. It is a figure which shows an example of the screen in the case of superimposing the non-input item on the input screen in Example 2 and prompting the input. FIG. 5 is a diagram showing an example of a screen in which an uninput item of FIG. 16 is input in the second embodiment and the next uninput item is superimposed and displayed to prompt the input. It is a figure which shows an example of the screen in the case of superimposing and displaying a plurality of uninput items on the input screen in Example 2, and prompting the input. It is a figure which shows an example of the screen at the time of prompting the input by sequentially switching the display position to the position of the input item which has an uninput item in Example 2. FIG. When using the input screen for general input in Example 2 and the input screen for necessary items when a predetermined disease is equal to or higher than a predetermined threshold value, when inputting general input is performed. It is a figure which shows an example of a screen. When the input screen for performing general input in Example 2 and the input screen for the necessary items when the predetermined disease is equal to or higher than the predetermined threshold value are used, the predetermined disease is equal to or higher than the predetermined threshold value. It is a figure which shows an example of the screen at the time of inputting the necessary item in the case of. FIG. 5 is a conceptual diagram schematically showing an example of the overall configuration of the emergency transport support system according to the fourth embodiment. It is a flowchart which shows an example of the whole processing process in the emergency transport support system in Example 4. It is a table which shows the improvement of the prediction accuracy in Example 4, and the figure which shows the ROC curve. It is a table which shows the prediction accuracy in the stroke subcategory in Example 5, and the figure which shows the ROC curve.

FIG. 1 shows an example of the overall configuration of the emergency transport support system 1 of the present invention. Further, FIG. 2 shows an example of the hardware configuration of the computer used in the emergency transport support system 1 of the present invention.

The emergency transport support system 1 uses a management computer 2 and an ambulance crew terminal 3. The management computer 2 and the ambulance crew terminal 3 are realized by a computer such as a server, a personal computer, or a portable communication terminal. The computer includes an arithmetic unit 70 such as a CPU that executes arithmetic processing of a program, a storage device 71 such as a RAM or a hard disk that stores information, a display device 72 such as a display, and an input device 73 that inputs information. It has a communication device 74 that communicates the processing result of the arithmetic unit 70 and the information stored in the storage device 71. When the computer is provided with a touch panel display, the input device 73 and the display device 72 may be integrally configured. Touch panel displays are often used, for example, in portable communication terminals such as tablet computers and smartphones used by paramedics, but are not limited to this.

The touch panel display is a device in which the functions of the display device 72 and the input device 73 are integrated in that input can be performed directly on the display with a predetermined input device (such as a pen for a touch panel) or a finger.

The function may be realized by one computer, or the function may be realized by a plurality of computers. The computer in that case may be, for example, a cloud server.

Further, each means in the emergency transport support system 1 of the present invention may form the same area physically or substantially, only the functions thereof are logically distinguished.

The ambulance crew terminal 3 is a computer used by ambulance crew members, and is preferably a tablet computer or the like. Information such as the symptoms of the patient transported by the ambulance crew is input from the ambulance crew terminal 3, and the input is accepted by the management computer 2, and the patient's illness possibility, hospitalization possibility, emergency transportation urgency Accepts the output of the result such as.

As the information (input information) that receives input from the ambulance crew terminal 3, the possibility of a predetermined disease (a disease that is predetermined to be treated by a specialist in particular), the possibility of hospitalization, and the degree of urgency are predicted processing described later. Information for use in the processing in unit 21 is sufficient. In addition to the processing in the prediction processing unit 21, the patient's attribute information (name, age, gender, blood type, telephone number, address, family medical institution, chronic disease, patient's own information such as the medicine being taken) and the patient. You may accept input of related information (such as family information).

When the ambulance crew terminal 3 is, for example, a tablet computer, information can be input by tapping each input item, or voice input may be performed. In this case, the voice recognition engine is started by selecting the button for voice input in the input item, and the utterance voice of the ambulance crew is accepted as the utterance voice data by the microphone of the ambulance crew terminal 3. Then, when the voice recognition engine executes the text conversion process on the spoken voice data, it is reflected as the information of the input item as the text data. The voice recognition engine may be provided in the first aid corps terminal 3, or the utterance voice data is sent from the first aid corps terminal 3 to a server equipped with the voice recognition engine, the text is converted into text on the server, and then the text is displayed. Data may be accepted and reflected in the input items.

Further, the emergency transport support system 1 of the present invention may be realized by mounting a part or all the functions of the management computer 2 on the emergency services terminal 3 without distinguishing between the management computer 2 and the emergency services terminal 3.

The management computer 2 has an input information reception processing unit 20, a prediction processing unit 21, and an output processing unit 22.

The input information reception processing unit 20 receives input information such as a patient's symptom to be transported from the emergency services terminal 3. In addition, a part or all of the input information that has received the input is passed to the prediction processing unit 21.

The prediction processing unit 21 executes a process of predicting the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency based on the input information such as the symptom of the patient who received the input by the input information reception processing unit 20. Various methods can be used for this prediction processing. For example, input information such as the symptoms of the patient who received the input is used as a variable (parameter), and a method of executing multivariate analysis by a predetermined mathematical formula based on the parameter is used. be.

The prediction processing unit 21 may execute processing by using machine learning, deep learning, or the like based on input information such as a symptom of a patient who has received the input. In this case, input information such as the symptoms of the patient who received the above input is input to the learning model in which the weighting coefficient between the neurons of each layer of the neural network consisting of many layers is optimized, and the output thereof is output. Based on the values, the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency may be obtained as output values. As the learning model, a part or all of the input information such as the patient's symptom and the one in which the patient's disease, the presence or absence of hospitalization, and the degree of urgency are given as correct answer data can be used.

When machine learning and deep learning are used, the prediction processing unit 21 sends a part or all of the input information to a predetermined server that executes the machine learning and deep learning processing, and the possibility of a predetermined disease is sent from the server. ， The possibility of hospitalization and the degree of urgency may be accepted as output values.

The output processing unit 22 executes a process of outputting the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency obtained by the prediction processing unit 21. For example, the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency obtained by the prediction processing unit 21 are sent to the emergency services terminal 3 and displayed on the screen.

By visually recognizing the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency displayed on the emergency services terminal 3, the ambulance crew can find the possibility of a predetermined disease and hospitalization when searching for a medical institution to be transported. It is possible to convey the possibility and urgency of the patient, and the medical institution can appropriately judge whether or not to accept the patient. For example, if the degree of urgency is high but the possibility of hospitalization is low, it will be possible to make a decision on acceptance even if there is no vacant bed, which will lead to the efficiency of limited emergency medical resources. can.

Next, an example of the processing process of the emergency transport support system 1 of the present invention will be described with reference to the flowchart of FIG.

When the ambulance crew is dispatched by calling 119 or the like to carry out emergency transportation of the patient, the ambulance crew members perform a predetermined operation to input information (input information) about the patient from the ambulance crew terminal 3. At this time, for example, if the patient can talk, the information is directly heard from the patient, and the information collected by the ambulance crew during the interview and observation, the information collected by the medical device, etc. are input as input information. As the input information, if it is information to be used in the processing in the prediction processing unit 21, the disease that is predetermined to be treated by a specialist, for example, the possibility of stroke, myocardial infarction, sepsis, the possibility of hospitalization, and the degree of urgency. good.

The paramedics perform a predetermined operation on the paramedics terminal 3, for example, display the input screens shown in FIGS. 4 to 7 on the paramedics terminal 3 and input the input information.

First, on the emergency services terminal 3, the age and gender of the patient are input from the input screen shown in FIG. 4, and vital sign information is input. Information such as the patient's age and gender may be displayed by being sent to the ambulance crew terminal 3 from the computer of the headquarters such as the disaster emergency information center that receives the 119 call and receiving it at the ambulance crew terminal 3. .. In addition to age and gender, other attribute information such as name and address may be accepted.

Vital signs include whether or not there is cardiac arrest, and if there is cardiac arrest, "VF" (ventricular fibrillation), "pulseless VT" (pulseless ventricular tachycardia), "PEA" (pulseless) Enter the state such as "electrical activity" or "Asystole".

In addition, as vital signs, the consciousness level ("0", "I-1", "I-2", "I-3", "II-10", "II-20", "II-30", " III-100 ”,“ III-200 ”,“ III-300 ”), whether blood pressure (systolic blood pressure, diastolic blood pressure) and blood pressure cannot be measured, whether pulse and pulse cannot be measured, and whether or not there is arrhythmia , Body temperature, respiratory rate and respiratory rate not measurable, SpO ₂ (percutaneous arterial oxygen saturation) and unmeasurable, room air (with natural breathing), oxygen administration If so, the administration method and dose, whether breathing is difficult ("inability to speak", "word unit", "interrupted or choppy or phrase unit", "mild or sentence unit", "none"), cold feeling of the hand Information such as the presence / absence of wetness of the skin of the hand, the presence / absence of cardiac arrest during transportation, the presence / absence of left and right pupils and light reflection is input as input information.

Information that can be measured by medical devices such as blood pressure, pulse, SpO ₂ , respiratory rate, and oxygen dose among the patient's vital signs can be received from each medical device by the emergency services terminal 3 by wireless communication or the like. It may be reflected as input information in the corresponding input item on the input screen of.

Further, from the input screen shown in FIG. 5, information such as an electrocardiogram, a symptom, an interview item, and an observation item of the patient is input. Electrocardiograms include limb leads ("I", "II", "III"), ST values ("rise", "decrease", "no change"), RR intervals ("arranged", "irregular"), etc. Enter the information as input information. As for the information related to the electrocardiogram, as with the vital signs, the information measured by the medical device such as the electrocardiogram monitor is received by the emergency services terminal 3 by wireless communication or the like, and is reflected as input information in the corresponding input items on the input screen of FIG. May be done.

For information such as symptoms, interview items, and observation items, in addition to listening to the information by talking with the patient, input the information observed by the ambulance crew. If the patient cannot perform all because he / she cannot talk due to impaired consciousness, enter "All cannot be performed". And the presence or absence of pain, the type of pain (“thrilling”, “tingling”, “changing with movement”, “changing with cough / inspiration”, “changing with touch”), presence / absence of nausea, presence / absence of vomiting, Presence / absence of dizziness, presence / absence of spasm, presence / absence of cold sweat, presence / absence of cough, presence / absence of sputum, presence / absence of numbness, presence / absence of abdominal pain, presence / absence of diarrhea, presence / absence of headache, presence / absence of back pain and its site ("full surface", "one side" ”), Acute chest, epigastric, neck, jaw, arm pain, discomfort, pressure, sudden consciousness disorder or headache, paralysis and its site (“face ”,“ upper limbs ”, "Lower limbs"), presence or absence of limb deformity and its site, presence or absence of wounds, presence or absence of bleeding, presence or absence of vomiting / melena, presence or absence of active bleeding, presence or absence of walking, presence or absence of airway foreign body, presence or absence of sitting breathing ， Presence or absence of high-energy trauma ， Whether or not the symptom worsens when walking ， Whether or not the symptom gradually worsens ， Whether or not the symptom is present even at rest ， Whether or not the symptom continues even now ， 24 hours Enter the information as input information by selecting the relevant item for information such as whether or not the patient has symptoms more than once within the period and whether or not the patient has had symptoms for half a year.

Further, from the input screen shown in FIG. 6, information such as the patient's medical history is input by listening to the patient or an attendant. If the patient is unable to have a conversation due to impaired consciousness, etc., and there is no attendant, and all cannot be performed, enter "All cannot be performed". And diabetes, hypertension, hyperlipidemia, angina, myocardial infarction, PCI (percutaneous coronary angioplasty), CABG (coronary artery bypass grafting), cerebral hemorrhage, cerebral infarction, malignant tumor, atrial fibrillation, epilepsy, psychiatry. Enter as input information by selecting the relevant item for information such as disease, oral administration of anti-platelet / anticoagulant, history of hospital visits at least once a month, history of hospitalization within half a year, unknown, unconfirmed, etc. ..

Further, from the input screen shown in FIG. 7, the information on the treatment contents performed by the ambulance crew on the patient is input. Airway management ("hands", "laryngoscope", "forcers", "nasal AW" (nasal airway management), "LM" (larin dial mask), "esophageal closure", "tracheotomy", " Video laryngoscope "), with or without artificial breathing, thoracic compression ("hand", "automatic"), cardiopulmonary resuscitation ("hand", "automatic"), airway management before and after CPA (cardiopulmonary arrest) Results (“implemented”, “not performed”, “discontinued”, “cannot be secured”), defibrillation performed, drug administration (“adrenaline”, “dextrose”, “epipen”), blood glucose Presence or absence of measurement, presence or absence of bleeding ("compression bleeding", "hemostatic band", "tightening bleeding", "others"), presence or absence of fixation ("cervical collar", "total spinal column fixation", "negative pressure type fixation mat" , "Defibrillation", "Other"), with or without coating ("Wound coating", "Burning wound coating", "Prolapsed intestinal tract coating", "Penetrating foreign body fixation", "Other"), Continuation of home medical care Presence / absence (“intravenous nutrition / drip”, “tracheostomy / external fistula, etc.”, “others (home oxygen / laryngoscope, etc.)”), presence / absence of heat retention, presence / absence of cooling, presence / absence of obstetric treatment, psychiatry For information such as whether or not there is a response and whether or not there is another action, enter it as input information by selecting the relevant item.

The input information is not limited to the above, but is not limited to the above, but is patient attribute information (name, age, gender, blood type, telephone number, address, family medical institution, chronic illness, patient's own information such as the medicine being taken). And the patient's relevant information (such as family information) may be accepted. Further, the input information can be appropriately changed according to the information used in the prediction processing in the prediction processing unit 21.

Then, when the ambulance crew inputs each of the above input information on the ambulance crew terminal 3 and performs a predetermined operation, the input information is sent from the ambulance crew terminal 3 to the management computer 2, and the input information reception processing unit in the management computer 2 The input information is accepted at 20 (S100).

The input information reception processing unit 20 passes a part or all of the input information received from the emergency team terminal 3 to the prediction processing unit 21, and the prediction processing unit 21 determines in advance based on a part or all of the input information. Perform a process to predict the disease, such as stroke, myocardial infarction, sepsis, hospitalization, and urgency (S110).

The possibility of stroke is calculated using, for example, the mathematical formula shown in FIG.

The parameters "history of cerebral bleeding" and "history of blood pressure" in the formula of FIG. 8 are "1" when there is "cerebral bleeding" and "diabetes" in the history of FIG. , And the parameters "dazzle", "convulsions", and "arrhythmia" are input information. 1 ”,“ 0 ”if not present, and the parameters“ systolic blood pressure ≧ 160 ”and“ systolic blood pressure ≦ 110 ”are conditions based on the value of systolic blood pressure on the vital sign input screen of FIG. 4 as input information. If is satisfied, "1" is satisfied, if not satisfied, "0" is set, and the parameter "pupil disparity" is the input information of the size of the pupil and the light reflection on the vital sign input screen of FIG. If the condition such as no condition is satisfied, "1" is satisfied, if not satisfied, "0" is set, and the parameter "change of consciousness" is input information. The consciousness level on the vital sign input screen in Fig. 4 is "II-20". "1" if the prescribed conditions are met, "0" if the conditions are not met, and the parameters "upper limb paralysis" and "face paralysis" are the symptoms, interview items, and observation items in Fig. 5. On the input screen of, if "paralysis" is selected and "upper limbs" and "face" are selected, "1" is selected, if not selected, "0" is set as the parameter "age". Is the input screen of FIG. 4, in which "1" is substituted as an argument when the "age" is 40 years or older, and "0" is substituted when the "age" is less than 40 years.

The possibility of myocardial infarction is calculated using, for example, the mathematical formula shown in FIG.

What is the parameter "chest symptom that lasts for 5 minutes or more" in the formula of FIG. 9? Chest pain, discomfort, and oppressive feeling continue for 5 minutes or more on the input screen of the symptom, interview item, and observation item of FIG. "1" in the case, "0" in the absence, the parameters "pulse rate ≥ 100 times / minute" and "systolic blood pressure ≥ 140" are input information, and the pulse rate and contraction on the vital sign input screen of FIG. Enter "1" if the condition is satisfied based on the value of the systolic blood pressure, "0" if the condition is not satisfied, and enter the vital sign shown in FIG. 4 as input information for the parameter "difficulty in breathing and unable to speak". If the dyspnea on the screen is "mild or sentence unit" or more, "1", if "none", "0", the parameters "movement changes pain", "cold sweat", "walking worsens symptoms" "Symptoms continue" means that there is pain on the input screen of the symptoms, interview items, and observation items in Fig. 5, and "changes when moving", cold sweat, worsens when walking, and persistent symptoms are selected. If yes, "1" if not selected, "0" if not selected, and "1" if PCI is selected on the history input screen of FIG. 6 as input information for the parameter "PCI". , And if it is not selected, "0" is substituted as an argument.

The possibility of sepsis is calculated using, for example, the mathematical formula shown in FIG.

The parameters "diabetes in history" and "malignant tumor in history" in the formula of FIG. 10 are "1" when there are "diabetes" and "malignant tumor" in the input screen of FIG. 6 as input information, and when there is no malignant tumor. "0" is set to the parameters "systolic blood pressure ≤ 110", "mean blood pressure ≤ 65", "pulse rate ≥ 90 times / minute", respiratory rate ≥ 22 times / minute "," body temperature ≥ 38.0 degrees "," "Body temperature ≤ 36.0 degrees" is "1" when the conditions are satisfied based on the values of systolic blood pressure, diastolic blood pressure, pulse rate, respiratory rate, and body temperature on the vital sign input screen of FIG. 4 as input information. , If not satisfied, "0" is satisfied, and the parameter "change of consciousness" is the input information, such as the consciousness level on the vital sign input screen in Fig. 4 is "II-20" or higher. In case of "1", if not satisfied, "0", and the parameters "cough or sputum", "abdominal pain or diarrhea", and "one-sided back pain" are input information of the symptoms, interview items, and observations in Fig. 5. "1" when "one side" is selected from "cough", "sputum", "abdominal pain", "diarrhea", and "back pain" on the item input screen, and "0" when it is not selected. , And the parameter “age” is the input screen of FIG. 4, and “1” is substituted as an argument when the “age” is 40 years or older, and “0” is substituted when the “age” is less than 40 years.

The possibility of hospitalization is calculated using, for example, the mathematical formula shown in FIG.

The parameters "myocardial infarction in history", "history of hospital visits at least once a month", and "history of hospitalization within half a year" in the formula of FIG. 11 are input information such as "myocardial infarction" and "month 1" on the input screen of FIG. "1" if there is "history of hospital visits more than once" and "history of hospitalization within half a year", "0" if not, parameters "mean blood pressure ≤ 65", "pulse rate ≥ 90 times / minute", "breathing" "Number ≥ 22 times / minute", "body temperature ≥ 38.0 degrees", "body temperature ≤ 36.0 degrees", and "oxygen saturation ≤ 95" are input information such as systolic blood pressure on the vital sign input screen of FIG. If the conditions are satisfied based on the values of diastolic blood pressure, pulse rate, respiratory rate, and body temperature, "1" is satisfied, if not satisfied, "0" is used, and the parameter "change of consciousness" is input information in FIG. The figure shows "1" when the condition above the specified value is satisfied, such as the consciousness level on the vital sign input screen is "II-20" or higher, "0" when the condition is not satisfied, and the parameter "oxygen administration" is shown in the figure. If "room air" is not selected and "oxygen administration" is selected as the "administration method" on the vital sign input screen of 4, "1" is set, and if it is not satisfied, "0" is set, and the parameter " "Incapable of oral intake" means "1" when "possible" is selected for "oral intake" on the input screen for symptoms, interview items, and observation items in Fig. 5, and "1" when "possible" is not selected. “0”, parameter “age” is the input screen of FIG. 4, “1” when “age” is 40 years or older, “0” when under 40 years, parameter “high energy trauma” , "Chest pain", "Blood pressure or melena", and "Impossible to walk" are input information such as "High energy trauma", "Chest pain", and "Vomiting" on the input screen for symptoms, interview items, and observation items in Fig. 5. Substitute "1" as an argument when "impossible" is selected from "/ melena" and "walking", and "0" when "impossible" is not selected.

The degree of urgency is calculated using, for example, the mathematical formula shown in FIG.

Parameters in the formula of FIG. 12, "mean blood pressure ≤ 65", "pulse rate ≥ 90 times / minute", "respiratory rate ≥ 22 times / minute", "body temperature ≥ 38.0 degrees", "oxygen saturation ≤ 95", "Body temperature ≤ 36.0 degrees" is "body temperature ≤ 36.0 degrees" when the conditions are satisfied based on the values of systolic blood pressure, diastolic blood pressure, pulse rate, respiratory rate, and body temperature on the vital sign input screen of FIG. "1", "0" if not satisfied, and the parameter "change of consciousness" as input information, such as the consciousness level on the vital sign input screen of Fig. 4 is "II-20" or higher, and the condition of a predetermined value or higher is satisfied. If this is the case, "1" is selected, if it is not satisfied, "0" is selected. What is the parameter "oxygen administration"? "Room air" is not selected on the vital sign input screen shown in FIG. "1" if "oxygen administration" is selected, "0" if not satisfied, parameters "high energy trauma", "inability to take orally", "foreign body of airway", "sitting breath", " What is "active bleeding"? "High energy trauma", "oral intake" is "impossible", "foreign body in the airway", "sitting breath", etc. "1" when "active bleeding" is selected, "0" when not selected, parameter "age" is the input screen of Fig. 4, and "age" is 40 years or older. If "1" is selected, "0" is selected if the person is under 40 years old, and "arrhythmia" is the parameter "arrhythmia" if "with arrhythmia" is selected on the vital sign input screen of FIG. 4 of the input information. Substitute 1 ”and“ 0 ”if not selected as arguments.

When displaying the possibility of a disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency with probability, for example, 100 is added to the value obtained by the above mathematical formulas of FIGS. 8 to 12. May be displayed by multiplying, or when displaying by rank, threshold values are set for each of a plurality of ranks, and the values obtained by the above mathematical formulas of FIGS. 8 to 12 are compared with the threshold values. The rank may be determined by doing so.

Each of a to r in the mathematical formulas of FIGS. 8 to 12 is a weighting coefficient for each parameter. Further, the weighting coefficients a to r in each mathematical expression may be the same value or different values in each mathematical expression. As an argument, "1" may be set when the condition is satisfied (selected), "0" may be set when the condition is not satisfied (not selected), or any value may be set. If there are options such as "unknown" and "unconfirmed" instead of two options as input information, the value indicating that option may be reflected. For example, it can be set as "1" when it is satisfied, "0" when it is not satisfied, "Unknown", and "0.5" when it is "unconfirmed".

The prediction processing unit 21 executes prediction processing by multivariate analysis as shown in FIGS. 8 to 12 described above, and also uses a part or all of the input information received by the input information reception processing unit 20 as an input value to be a machine. Predictive processing may be performed using learning and deep learning, and the output values may be the possibility of disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency.

As described above, the output processing unit 22 sends the possibility of a disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency in the prediction processing unit 21 to the emergency team terminal 3, and the emergency team terminal 3 sends the possibility. The possibility of disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency are displayed (S120). An example of the screen at this time is shown in FIG.

On the screen shown in FIG. 13, when the possibility is a certain value or more and the rank is a predetermined rank or more, the display may emphasize the color, thickness, size, font, and the like.

In the above-described first embodiment, when the emergency team member inputs each input information on the emergency team terminal 3 and performs a predetermined operation, the input information is sent from the emergency team terminal 3 to the management computer 2, and the management computer 2 receives the input information. Although the case where the input information reception processing unit 20 accepts the input information has been described, the input information input by the emergency team member on the emergency team terminal 3 may be sequentially sent from the emergency team terminal 3 to the management computer 2.

In this case, the input information reception processing unit 20 sequentially passes the received input information to the prediction processing unit 21, and the prediction processing unit 21 sequentially sequentially determines the possibility of a predetermined disease such as stroke, myocardial infarction, or sepsis. Perform the process of predicting the possibility of hospitalization and the degree of urgency. When predictive processing is performed based on sequentially input information, processing cannot be executed as it is because there is no corresponding input information for items that have not been input. Therefore, for example, in the mathematical formulas of FIGS. 8 to 12, uninput items are excluded from the calculation formula as parameters and calculated, or a predetermined value such as the average value of the input items is calculated as an argument. do. Alternatively, the calculation process may be performed when the input information of the number of items equal to or larger than the predetermined number of input items is input.

If there is a possibility that the disease is above a predetermined threshold value, the items that have not been entered for the disease and the number of the items are specified. As for the items that have not been input and the number of the items, the items that have not been input among the parameters of the mathematical formulas for the prediction processing unit 21 to calculate the possibility of each disease, for example, the mathematical formulas of FIGS. 8 to 10 have not been input. It can be specified as an input item and the number of the items.

The output processing unit 22 sends the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency calculated by the prediction processing unit 21 to the emergency services terminal 3 each time, and the possibility of a disease predetermined by the emergency services terminal 3. ， Display the possibility of hospitalization, part or all of the urgency. When the prediction processing unit 21 identifies the items that have not been input and the number of the items, the output processing unit 22 also sends them to the emergency services terminal 3 each time, and displays the display on the emergency services terminal 3. Let me do it. Then, the ambulance crew terminal 3 displays the diseases that may exceed a predetermined threshold value, the possibility thereof, and the number of uninput items, and highlights the input items so as to prompt the input for the uninput items. You may. In this case, various highlights such as changing the name of the input item, the color, font, and size of the input field, and blinking can be performed. FIG. 14 shows an example of a screen in which the prediction processing unit 21 calculates that there is a possibility of sepsis, and the body temperature and the number of uninput items are one as non-input items. Further, FIG. 15 shows an example of a screen in which the prediction processing unit 21 calculates that there is a possibility of stroke and the number of non-input items such as pupil and light reflex is four.

Further, the output processing unit 22 highlights the input items for the uninput items as described above, and displays the uninput items by changing to the normal screen display or superimposing it on the normal screen display. You may be prompted to enter. For example, FIG. 16 shows an example of a screen in which the prediction processing unit 21 calculates that there is a possibility of stroke, and the number of non-input items is four, such as upper limb paralysis and facial paralysis. Is shown. Then, a screen prompting the input of the uninput item is displayed superimposed on the screen displayed on the ambulance crew terminal 3, and the input of the uninput item is urged. FIG. 17 shows a case where after inputting the paralysis of the upper limbs of FIG. 16, a screen prompting the input of the next uninput item, facial paralysis, is displayed. The input of paralysis of the upper limbs, which was an uninput item in FIG. 16, changed the possibility of stroke (diagnosis probability) calculated by the prediction processing unit 21 in FIG. 17, and the number of uninput items was also reduced to 3. It's changing.

In addition, although the case where the uninput items are input one by one is shown in FIGS. 16 and 17, as shown in FIG. 18, the uninput items may be input so that they can be input collectively. In addition, instead of superimposing the display, the position of the normal screen display may be changed to display the uninput item at a position where it is easy to input, thereby prompting the input.

Further, the ambulance crew terminal 3 may be able to input uninput items by sequentially switching the display such as scrolling the screen in order to the display position of the uninput items. This is schematically shown in FIG. In FIG. 19, the suspicion of stroke is 68%, and the non-input items are the pupil and the light reflex, and the presence or absence of diabetes as a medical history. When the input items of the pupil and the light reflex are input, the input items are input. Is schematically shown to detect and scroll to the screen for inputting the presence or absence of diabetes as a medical history.

In addition, in the ambulance crew terminal 3, an input screen for general input (first input screen) and an input screen for necessary items when a predetermined disease is equal to or higher than a predetermined threshold value (second input screen). Screen) and may be provided. In this case, the ambulance crew inputs the input items from the general input screen (first input screen) on the ambulance terminal 3 and outputs the possibility of the disease calculated by the prediction processing unit 21 in the same manner as described above. It is received from the processing unit 22 and transmitted to the management computer 2. Then, when the output processing unit 22 accepts from the output processing unit 22 that the possibility of the disease calculated by the prediction processing unit 21 of the management computer 2 is equal to or higher than a predetermined threshold value, a predetermined operation, for example, a display indicating that there is a possibility of a disease. By selecting, or automatically, the display may be switched to the input screen (second input screen) of the necessary items related to the disease, and the input may be performed. In this case, the input screen (first input screen) for general input is a screen for inputting vital signs, symptoms / interview items / observation items, history, treatment details, etc., and is a necessary item for each disease. The input screen (second input screen) of is preferably a collective input screen including input items corresponding to parameters indicating whether or not there is a possibility of a predetermined disease. When switching from the first input screen to the second input screen, the information of the input items input on the first input screen is directly reflected in the corresponding input items on the second input screen.

This is schematically shown in FIG. FIG. 20 is a general input screen (first input screen), in which a predetermined possibility of a disease and uninput items are displayed. Then, by pressing the button of the disease, the prediction processing unit 21 identifies the input items corresponding to the parameters for calculating the possibility of the disease, and as shown in FIG. 21, the necessary items related to the disease Display the input screen (second input screen) for displaying the input items (specified input items). On the second input screen, the input items already input on the first input screen are reflected as they are, and the input items that have not been input are highlighted. Then, by selecting the "Other detailed items" button in FIG. 21, the screen can be switched back to the general input screen (first input screen) of FIG. 20.

As described above, by executing the process as in the second embodiment, when the possibility of a predetermined disease is equal to or higher than a predetermined threshold value, information on necessary input items for the predictive process for the disease is urged. It is possible to give priority to the possibility of the disease. Therefore, the ambulance crew can obtain highly accurate information on the possibility of the disease of the patient to be transported, promptly select the appropriate medical institution to be transported, and promptly inform the medical institution of the possibility of the disease. Can be communicated.

In the first and second embodiments, the output processing unit 22 outputs the result of the prediction processing unit 21 to the emergency services terminal 3, but for example, information on a medical institution with which the emergency services may contact. Is input from the ambulance crew terminal 3, and the output processing unit 22 of the management computer 2 that has received the information is in the prediction processing unit 21 on both or / or the computer of the medical institution and / or the ambulance crew terminal 3. The result may be output and displayed. As a result, the medical institution can quickly confirm the result in the prediction processing unit 21.

In this case, as a medical institution with which the ambulance crew may contact, for example, the location information of the ambulance crew or the ambulance crew terminal 3 can be acquired by a GPS device or the like, and based on the location information, it is possible to respond to emergency transportation in the vicinity. A medical institution may be specified, and the results of the prediction processing unit 21 may be output and displayed on the computers of the specified medical institution, and a request for acceptance of emergency transportation may be notified. As a result, instead of the ambulance crew making an acceptance request over the phone, etc., the acceptance request is sent online to multiple medical institutions along with information such as the patient's objective illness possibility, hospitalization possibility, and urgency. On the other hand, it can be done at the same time. In this case, at the medical institution that accepts the acceptance request, by pressing a predetermined operation, for example, the acceptable button on the computer of the medical institution, the computer of the medical institution notifies the management computer 2 that the acceptance request can be accepted. Sent. Then, information on acceptable medical institutions, such as a name, is sent and displayed to the emergency services terminal 3 via the management computer 2. When the management computer 2 receives the acceptance notification from the computer of any one medical institution, the computer of the other medical institution that made the above-mentioned acceptance request is notified of the information indicating the completion of the acceptance request. As a result, it is possible to prevent duplicate acceptance notifications. In the computer of the medical institution that has received the information indicating the completion of the acceptance request, it is preferable that the button or the like cannot be pressed so that the acceptance notification cannot be sent to the management computer 2.

As another embodiment of the above-mentioned Examples 1 to 3, even if the emergency transport support system 1 has a weather information processing unit 23 that further acquires weather information and passes it to the predictive information processing unit 21. good. FIG. 22 shows an example of the overall configuration of the emergency transport support system 1 in this case.

The meteorological information processing unit 23 acquires from a predetermined meteorological information providing service in addition to the date and time and area (a certain range such as prefectures and municipalities). As a weather information providing service, for example, data (weather information) indicating the weather provided from a server of a weather forecasting organization such as the Japan Meteorological Agency or a weather forecasting company is acquired. Examples of weather information include identification information indicating the weather (weather such as sunny, cloudy, rain, and snow), temperature, humidity, wind speed, identification information indicating the wind direction (wind direction), pressure, and precipitation. The meteorological information processing unit 23 may acquire a part or all of these as meteorological information, or may acquire data related to other meteorological conditions.

As the weather information, data indicating arbitrary weather can be acquired and used. For example, among the acquired weather information, the minimum, average, and maximum temperature on the day of onset (the day when the emergency transport was performed), the daily range in the area where the emergency transport was performed, the difference from the previous day, and the minimum and minimum temperature on the day of the onset. Average / maximum pressure, barometric pressure difference, and difference from the previous day, minimum / average / maximum relative humidity on the day of onset, humidity difference, and difference from the previous day, temperature difference at dawn (two time points at dawn (for example, 1:00 am) (5 am) temperature difference), the difference between the average temperature in the predetermined period before the onset date (for example, 2, 3, 4, 5, 6 days before the onset date) and the average temperature on the onset date. In addition to the day of onset, the average with the day before the onset, the average during a predetermined period, for example, the weekly average may be used. Further, it is possible to use only a part of the acquired meteorological information without using all of the acquired meteorological information, or a value calculated by a predetermined calculation formula using a part or all of the acquired meteorological information. ..

The prediction processing unit 21 in this embodiment uses the input information such as the symptoms of the patient who received the input by the input information reception processing unit 20 and the weather information acquired by the weather information processing unit 23, and may have a predetermined disease. ， Perform the process of predicting the possibility of hospitalization and the degree of urgency. Various methods can be used for this prediction process.

For example, there is a method in which input information such as information of a patient who has received input is used as a variable (parameter), and multivariate analysis is executed by a predetermined mathematical formula based on the parameter.

Further, the prediction processing unit 21 uses machine learning, deep learning, and the like based on the input information such as the symptoms of the patient who received the input and the weather information acquired by the weather information processing unit 23. The process may be executed. In this case, for a learning model in which the weighting coefficient between neurons in each layer of the neural network consisting of a large number of layers is optimized, input information such as the patient's symptom and the meteorological information processing unit 23 that received the above input. The weather information acquired in step 3 may be input, and based on the output value, the possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency may be used as output values. As a learning model, a part or all of input information such as a patient's symptom, a part or all of weather information, and a data obtained by giving the patient's disease, hospitalization, and urgency as correct answer data can be used. ..

When machine learning and deep learning are used, the prediction processing unit sends a part or all of the input information and a part or all of the weather information to a predetermined server that executes the machine learning and deep learning processing. The possibility of a predetermined disease, the possibility of hospitalization, and the degree of urgency may be received as output values from the server.

The prediction processing unit 21 may use the input information received by the input information reception processing unit 20 and the weather information acquired by the weather information processing unit 23 as they are, but in order to maintain the accuracy, the prediction processing unit 21 executes preprocessing. You may.

For example, regarding the input information, among the items to be input on the input screens of FIGS. 4 to 7, 50% or more of the items are not input, or the variance of the variable value of each item is a predetermined value (for example, 5% or less). For example, the item of is excluded.

Further, the weather information may be subject to prediction processing by the prediction processing unit 21 by calculating, for example, a difference from the previous day, a difference from the average value, a difference between predetermined times in one day, and the like.

In this embodiment, as the prediction processing in the prediction processing unit 21, a case where the processing is executed by using gradient boosting, which is one of the machine learning algorithms, particularly XGBost will be described.

Next, an example of the processing process in this embodiment will be described with reference to the flowchart of FIG.

In the prediction processing unit 21, the input information reception processing unit 20 has input information in which 50% or more of the necessary input items have not been input in the past, and the value of each variable. The input information was input to the prediction processing unit 21 in advance for learning, excluding the input information having a variance of 5% or less.

In addition, the meteorological information processing unit 23 acquired the meteorological information of a predetermined day in the area where the emergency transportation was carried out. Of the acquired weather information, the minimum, average, and maximum temperature on the day of onset (the day when the emergency transport was performed), the daily range in the area where the emergency transport was performed, and the difference from the previous day, the minimum, average, and average on the day of the onset. Maximum pressure, pressure difference, and difference from the previous day, minimum / average / maximum relative humidity on the day of onset, humidity difference, and difference from the previous day, temperature difference at dawn (two time points at dawn (for example, 1:00 am and 5 am) (Time) temperature difference), the difference between the average temperature of the specified period before the onset date (for example, 2, 3, 4, 5, 6 days before the onset date) and the average temperature on the onset date, totaling 21 variables (parameters) did. ^{Then, after performing a χ 2} (chi-square) test as a single variable regression on the above-mentioned 21 variables for the objective variable, one of the pairs having a high correlation coefficient of p <0.05 was excluded. As a result, five variables were selected: the minimum temperature on the onset date, the maximum temperature, the diurnal air temperature, the maximum humidity, and the difference in the average temperature for the two days before the onset date. Then, the information input by the emergency personnel (information excluding the information excluded above among the input information received by the input information reception processing unit 20) and the model including each variable from these five variables one by one. Was created, and a model based on the lowest temperature on the day of onset was finally selected as a variable for which AUC was improved.

These input information and meteorological information (in this embodiment, the lowest temperature on the onset date) are read into the prediction processing unit 21 to generate a prediction model, and the prediction processing in the prediction processing unit 21 is executed.

Similar to Examples 1 to 3, when an ambulance crew is dispatched by calling 119 or the like to carry out emergency transportation of a patient, the ambulance crew performs a predetermined operation to provide information about the patient from the ambulance crew terminal 3 ( Input information) is input via the input screens shown in FIGS. 4 to 7.

The input information is not limited to the above, but is not limited to the above, but is patient attribute information (name, age, gender, blood type, telephone number, address, family medical institution, chronic illness, patient's own information such as the medicine being taken). And the patient's relevant information (such as family information) may be accepted. In addition, the season (spring, summer, autumn, winter, etc.), the current date, the day of the week, the time, the time zone, etc. may be entered as input information. The input information can be appropriately changed according to the information used in the prediction processing in the prediction processing unit 21.

Then, when the ambulance crew inputs each of the above input information on the ambulance crew terminal 3 and performs a predetermined operation, the input information is sent from the ambulance crew terminal 3 to the management computer 2, and the input information reception processing unit in the management computer 2 The input information is accepted at 20 (S200).

In addition, the meteorological information processing unit 23 acquires meteorological information (in this embodiment, the lowest temperature on the onset date) from the meteorological information providing service (S210). In addition, the necessary weather information may be acquired from the predetermined server of the weather information providing service each time, or the necessary weather information may be extracted from the acquired and stored weather information in advance. good.

The input information reception processing unit 20 passes a part or all of the input information received from the emergency team terminal 3 to the prediction processing unit 21, and the weather information processing unit 23 receives a part or all of the acquired weather information (book). In the example, the lowest temperature on the day of onset) is passed to the prediction processing unit 21. Then, the prediction processing unit 21 uses a part or all of the input information and a part or all of the weather information to determine a predetermined disease such as stroke, myocardial infarction, sepsis possibility, hospitalization possibility, and urgency. (S220).

The prediction processing unit 21 uses a part or all of the input information received by the input information reception processing unit 20 and a part or all of the weather information acquired by the weather information processing unit 23 as input values, and is a machine based on a prediction model generated in advance. Predictive processing is executed by learning and deep learning (S220), and the possibility of diseases (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency are output as output values.

As described above, the output processing unit 22 sends the possibility of a disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency in the prediction processing unit 21 to the emergency team terminal 3, and the emergency team terminal 3 sends the possibility. Display one or more of the possibility of disease (for example, stroke, myocardial infarction, sepsis), the possibility of hospitalization, and the degree of urgency (S230).

For example, as in the fourth embodiment, the prediction processing of the possibility of stroke was executed using the input information input by the emergency personnel and the weather information (the lowest temperature on the onset day in this embodiment), which is shown in FIG. 24. As described above, the training data of AUC (Area Under the Curve) was 0.992 and the test data was 0.979, which were extremely high values and were found to be extremely accurate. FIG. 24A shows the accuracy rate (Accuracy), sensitivity (sensitivity), specificity (specificity), F value (F1-score), and AUC values of the training data and the test data in a tabular format. Further, FIG. 24B shows a ROC curve (receiver operating characteristic curve) for the training data and the test data.

As described above, as a modification of Examples 1 to 4, by using weather information in addition to input information such as symptoms input by paramedics as information used for prediction processing in the prediction processing unit 21. It is possible to execute more accurate prediction processing.

As in the third embodiment, the output processing unit 22 may send the predicted result to the computer of the medical institution and display it on the computer.

Further, the prediction processing unit 21 executes the prediction processing based on the weather information acquired by the weather information processing unit 23 without using a part or all of the input information received from the input information reception processing unit 20. May be good.

Further, the timing at which the weather information processing unit 23 acquires the weather information may be stored in advance, or the weather information may be acquired after the input information reception processing unit 20 receives the input information.

As another modification of Example 4, the disease can be further subdivided and predicted as the prediction processing in the prediction processing unit 21. For example, for stroke, subarachnoid hemorrhage, cerebral infarction (other than main artery occlusion), cerebral hemorrhage, and main artery occlusion may also be predicted. Regarding myocardial infarction (acute heart disease), myocardial infarction (ST-segment elevation), myocardial infarction (non-ST-segment elevation), unstable arrhythmia, coronary spasmodic arrhythmia, congestive heart failure, and takotsubo cardiomyopathy , Tachyarrhythmia, slow arrhythmia, and other subcategories of cardiovascular disease may also be predicted.

In this case, as the learning model, as in Example 4, part or all of the input information such as the patient's symptoms, part or all of the weather information, the disease of the patient and the subcategory of the disease, the presence or absence of hospitalization, It is possible to use the data in which the urgency is given as the correct answer data.

The prediction processing unit 21 uses a part or all of the input information received by the input information reception processing unit 20 and a part or all of the weather information acquired by the weather information processing unit 23 as input values, and is a machine based on a prediction model generated in advance. Predictive processing is executed by learning and deep learning, and the possibility of diseases (for example, stroke, myocardial infarction, sepsis and their subcategories), the possibility of hospitalization, and the degree of urgency are output as output values.

As described above, the output processing unit 22 sends the disease (for example, stroke, myocardial infarction, sepsis) in the predictive processing unit 21 and the possibility of its subcategory, the possibility of hospitalization, and the degree of urgency to the emergency team terminal 3, and the emergency team terminal. In 3, the disease (for example, stroke, myocardial infarction, sepsis) and its subcategory possibility, hospitalization possibility, and urgency are displayed (S230).

For example, as in the fifth embodiment, when the prediction processing of the possibility of the stroke subcategory was executed using the input information and the weather information input by the emergency personnel, as shown in FIG. 25, AUC (Area Under the Curve) was executed. : Area under the time curve) is 0.970 for training data of submucosal bleeding, 0.945 for test data, 0.920 for learning data of cerebral infarction (other than main artery occlusion), 0.889 for test data, cerebral bleeding The training data was 0.920, the test data was 0.888, the training data for main artery occlusion was 0.907, and the test data was 0.888. did. FIG. 25 (a) shows the AUC of the training data and the test data for each subcategory of stroke in a tabular format. Further, FIG. 25 (b) shows a ROC curve (receiver operating characteristic curve) for the training data and the test data of each subcategory.

As described above, as a modification of Examples 1 to 4, by using weather information in addition to input information such as symptoms input by paramedics as information used for prediction processing in the prediction processing unit 21. Predictive processing for disease subcategories can also be performed.

By using the emergency transport support system 1 of the present invention, objective information can be transmitted to a medical institution. Therefore, even in medical institutions, it is possible to obtain objective information on the diseases of patients to be transported, the possibility of hospitalization, the degree of urgency, and the like. This will lead to the optimization of emergency medical care.

1: Emergency transport support system 2: Management computer 3: Emergency team terminal 20: Input information reception processing unit 21: Prediction processing unit 22: Output processing unit 23: Meteorological information processing unit 70: Arithmetic logic unit 71: Storage device 72: Display device 73: Input device 74: Communication device

Claims (24)

It is an emergency transportation support system used for emergency transportation.
An input information reception processing unit that accepts input information including patient symptoms from the emergency services terminal used by the emergency services, and
A predictive processing unit that calculates the possibility of hospitalization and / or the urgency of emergency transportation of the patient using part or all of the input information.
An output processing unit that outputs the possibility of hospitalization of the patient and / or the urgency of emergency transportation calculated by the prediction processing unit.
An emergency transport support system characterized by having. The prediction processing unit
The hospitalization possibility and / or urgency of the patient is calculated by using the information of the patient's symptoms, vital signs, and medical history in the input information.
The emergency transport support system according to claim 1. The prediction processing unit
Calculate the possibility of a predetermined disease based on a part or all of the input information.
The emergency transport support system according to claim 1 or 2. The predetermined diseases include any one or more of stroke, myocardial infarction, and sepsis.
The emergency transport support system according to claim 3. The output processing unit
The calculation result in the prediction processing unit is sent to the emergency services terminal and / or the computer of the medical institution.
The emergency transport support system according to any one of claims 1 to 4. The emergency transport support system is
Based on the location information of the ambulance crew or the ambulance crew terminal, identify a medical institution that can handle emergency transport in the vicinity.
The request for acceptance of emergency transportation and the result of calculation by the output processing unit in the prediction processing unit are sent to the computers of the specified one or more medical institutions.
The emergency transport support system according to any one of claims 1 to 5. The emergency transport support system is
When the acceptance notification is received from the computer of the medical institution that sent the acceptance request, information indicating the completion of the acceptance request is sent to the computers of the other medical institutions that sent the acceptance request, and the acceptance notification is sent. Do not send,
The emergency transport support system according to claim 6. The output processing unit
In the emergency services terminal or the computer of the medical institution, if the calculation result in the prediction processing unit is equal to or higher than a predetermined value or a predetermined rank or higher, highlighting is performed.
The emergency transport support system according to any one of claims 1 to 7. The ambulance crew terminal
When the calculation result of the possibility of the disease is equal to or higher than the predetermined threshold value, a display prompting the input of the input items corresponding to the calculation of the possibility of the disease is performed.
The emergency transport support system according to claim 3 or 4. The ambulance crew terminal
As a display prompting to input the non-input item, highlight the non-input item, add the non-input item as an input item and display it, superimpose the non-input item, and display up to the display position of the non-input item. To switch, to do by one or more of
The emergency transport support system according to claim 9. The ambulance crew terminal
When the calculation result of the possibility of the disease is equal to or higher than the predetermined threshold value, the first input screen is switched to the second input screen including the input items used for calculating the possibility of the disease. It is possible,
The emergency transport support system according to claim 9 or 10. The ambulance crew terminal
When the calculation result of the possibility of the disease is equal to or higher than the predetermined threshold value, the first input screen automatically switches to the second input screen including the input items used to calculate the possibility of the disease. Switch the display
The emergency transport support system according to any one of claims 9 to 11. The unfilled items are
It is specified by using the input items corresponding to the parameters that have not been input among the parameters used to calculate the possibility of the disease.
The emergency transport support system according to any one of claims 9 to 12. The emergency transport support system is
It has a meteorological information processing department that acquires meteorological information.
The prediction processing unit
Use some or all of the weather information to calculate the likelihood of hospitalization and / or the urgency of emergency transport for the patient.
The emergency transport support system according to any one of claims 1 to 13. It is an emergency transportation support system used for emergency transportation.
An input information reception processing unit that accepts input information including patient symptoms from the emergency services terminal used by the emergency services, and
With a prediction processing unit that calculates any one or more of the patient's disease possibility, hospitalization possibility, and emergency transportation urgency using part or all of the input information and part or all of the weather information. ，
An output processing unit that outputs the results calculated by the prediction processing unit, and
An emergency transport support system characterized by having. The prediction processing unit
Calculate the likelihood for the patient's disease subcategory,
The emergency transport support system according to claim 15. It is an ambulance crew terminal used by ambulance crews for ambulance transportation.
Accepts input of input information including patient's symptoms
Displays the possibility of hospitalization and / or the urgency of emergency transport of the patient calculated based on some or all of the input information.
An ambulance crew terminal that features that. It is an ambulance crew terminal used by ambulance crews for ambulance transportation.
Accepts input of input information including patient's symptoms
When the possibility of a predetermined disease calculated based on a part or all of the input information is equal to or higher than a predetermined threshold value, input the non-input items among the input items corresponding to the calculation of the possibility of the disease. Display to prompt
An ambulance crew terminal that features that. It is an ambulance crew terminal used by ambulance crews for ambulance transportation.
Accepts input of input information including patient's symptoms
Display one or more of the patient's illness potential, hospitalization probability, or emergency transport urgency calculated using part or all of the input information and part or all of the weather information. ，
An ambulance crew terminal that features that. Computer,
Input information reception processing unit that accepts input information including patient symptoms from the emergency services terminal used by the emergency services,
A predictive processing unit that calculates the possibility of hospitalization of the patient and / or the urgency of emergency transportation based on a part or all of the input information.
An output processing unit that outputs the possibility of hospitalization of the patient and / or the urgency of emergency transportation calculated by the prediction processing unit.
An emergency transport support program characterized by functioning as. Computer,
Input information reception processing unit that accepts input information including patient symptoms from the emergency services terminal used by the emergency services,
A predictive processing unit that calculates one or more of the patient's illness possibility, hospitalization possibility, or emergency transportation urgency using part or all of the input information and part or all of the weather information.
Output processing unit that outputs the result calculated by the prediction processing unit,
An emergency transport support program characterized by functioning as. The ambulance crew terminal used by the ambulance crew for emergency transport,
Processing unit that accepts input of input information including patient's symptoms,
A processing unit that displays the possibility of hospitalization of the patient and / or the urgency of emergency transportation calculated based on a part or all of the input information.
A computer program characterized by functioning as. The ambulance crew terminal used by the ambulance crew for emergency transport,
Processing unit that accepts input of input information including patient's symptoms,
When the possibility of a predetermined disease calculated based on a part or all of the input information is equal to or higher than a predetermined threshold value, input the non-input items among the input items corresponding to the calculation of the possibility of the disease. Processing unit that prompts
A computer program characterized by functioning as. The ambulance crew terminal used by the ambulance crew for emergency transport,
Processing unit that accepts input of input information including patient's symptoms,
Display one or more of the patient's illness potential, hospitalization probability, or emergency transport urgency calculated using part or all of the input information and part or all of the weather information. Processing unit,
A computer program characterized by functioning as.

Images