Method, device, and computer program for predicting cardiac arrest result in pre-hospital stage

Abstract

The present invention relates to a method for predicting a cardiac arrest result in a pre-hospital stage. The method comprises the steps of: receiving, as training data, at least one among demographic information, field environment information, and pre-hospital cardiovascular life support information; training a prediction model with the received training data and according to whether the return of spontaneous circulation is implemented by cardiovascular life support in a pre-hospital stage; and inputting demographic information, field environment information, and cardiovascular life support information about an emergency patient to the trained prediction model to predict whether the return of spontaneous circulation is in the pre-hospital stage.

Description

Method, device, and computer program for predicting the outcome of cardiac arrest in the prehospital stage

[0001] The present invention relates to a method, a device, and a computer program for predicting the outcome of cardiac arrest in the pre-hospital stage, and more particularly, to a method, a device, and a computer program for predicting the outcome of cardiac arrest in the pre-hospital stage to provide information helpful in making a decision on whether to perform cardiopulmonary resuscitation in the pre-hospital stage or advanced cardiopulmonary resuscitation in the hospital stage.

[0002] Prehospital care plays a vital role in the management of out-of-hospital cardiac arrest (OHCA). Previous studies have shown that several components of prehospital care, such as chest compression duration, defibrillation, and transport, are closely associated with OHCA patient outcomes. However, clinical decision-making in the OHCA setting remains challenging due to the difficulty in identifying the cause and predicting the patient's resuscitation prognosis.

[0003] Because it is difficult to predict whether return of spontaneous circulation (ROSC) will occur in the field without performing advanced cardiovascular life support (ACLS), it is unclear when a "scoop and run" to the hospital for ACLS is warranted. Because CPR in the moving ambulance can be degraded, "stay and play" resuscitation is prudent if ROSC is likely at the scene. However, if the medical cause is unknown and ROSC is difficult, transport to the hospital will allow for greater resource allocation and optimal care. The prehospital phase is time-sensitive and influenced by the field environment, but with less information and resources available than the in-hospital phase, making the decision between a "scoop and run" and a "stay and play" approach can be challenging.

[0004] While several papers have proposed machine learning-based OHCA patient outcome prediction models, relatively few have been built to take the prehospital stage into account. Predictive models for the prehospital stage are rare, and predictions are made only at the prehospital resuscitation stage. Field conditions can constantly change, and OHCA outcomes, particularly in the prehospital stage, can vary over time.

[0005] Therefore, there is a growing need for a real-time prediction model useful for pre-hospital resuscitation by emergency medical services (EMS).

[0006] The present invention aims to provide a method, device, and computer program for predicting the outcome of cardiac arrest at a predetermined time interval in a pre-hospital stage, which predicts the return of spontaneous circulation and further prognosis in real time during emergency resuscitation attempts.

[0007] In order to achieve the above object, the present invention is characterized in that it comprises a step of inputting at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data, a step of training a prediction model using the input learning data and whether or not ROSC is achieved through cardiopulmonary resuscitation in the pre-hospital stage, and a step of inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of an emergency patient into the trained prediction model to predict whether or not ROSC is achieved in the pre-hospital stage.

[0008] Preferably, the step of inputting the learning data may input at least one of city, gender, race, and age as demographic information.

[0009] Preferably, the step of inputting the learning data includes inputting at least one of location type and time as field environment information, and the location type can be classified into at least one of home, medical facility, public building, nursing home, street, industrial area, transportation center, recreation area, and ambulance.

[0010] Preferably, the step of inputting the learning data may input at least one of bystander cardiac arrest information, bystander CPR information, use of an automated external defibrillator (AED) by a first responder, a bystander, or an emergency medical service (EMS), the patient's initial rhythm, a method of securing an airway before the hospital, and whether or not epinephrine was administered before the hospital, as pre-hospital CPR information.

[0011] Preferably, the step of training the prediction model may include dividing the cardiopulmonary resuscitation time in the pre-hospital stage into a certain time period, dividing the input learning data into a certain time period, and training the prediction model with the learning data divided into a certain time period.

[0012] Preferably, the cardiopulmonary resuscitation time in the pre-hospital stage may be defined as the time from the ambulance's arrival at the scene to the arrival at the emergency room.

[0013] Preferably, the step of training the prediction model additionally trains the prediction model using survival rate until discharge from the emergency room, 30-day survival rate, and CPC (Cerebral Performance Category) per minute, and the step of predicting whether or not ROSC occurs in the pre-hospital stage can additionally predict survival rate until discharge from the emergency room, 30-day survival rate, and CPC (Cerebral Performance Category) per minute.

[0014] Preferably, the step of training the prediction model may use a random forest, a light gradient boosting machine (LightGBM), or extreme gradient boosting (XGBoost) as the prediction model.

[0015] In addition, the present invention is a device for predicting the outcome of cardiac arrest in a pre-hospital stage, comprising: a processor including one or more cores; and a memory; wherein the processor receives at least one or more of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data, trains a prediction model using the received learning data and whether or not ROSC is achieved by cardiopulmonary resuscitation in the pre-hospital stage, and inputs demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of an emergency patient into the learned prediction model to predict whether or not ROSC is achieved in a predetermined time unit in the pre-hospital stage.

[0016] In addition, the present invention is characterized in that it is a computer program including commands stored in a computer-readable storage medium and causing a computer to perform the following operations, wherein the operations include: an operation of receiving at least one or more of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data; an operation of training a prediction model using the input learning data and whether or not ROSC is achieved through cardiopulmonary resuscitation in the pre-hospital stage; and an operation of inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of an emergency patient into the trained prediction model to predict whether or not ROSC is achieved in the pre-hospital stage.

[0017] The present invention has the advantage of being able to predict whether or not ROSC will occur in the pre-hospital stage through a predictive model, and additionally predict the survival rate until discharge from the emergency room, the 30-day survival rate, and the CPC (Cerebral Performance Category) per minute, thereby helping in decision-making on whether to perform cardiopulmonary resuscitation in the pre-hospital stage or in the hospital stage.

[0018] In addition, the present invention has the advantage of being able to predict whether or not spontaneous circulation has been restored at a pre-hospital stage that changes over time in certain time units.

[0019] Figure 1 illustrates a flowchart of a method for predicting the outcome of cardiac arrest in a pre-hospital stage according to an embodiment of the present invention.

[0020] FIG. 2 is a diagram showing data required for prediction and a target to be predicted in a method for predicting the outcome of cardiac arrest in a pre-hospital stage according to an embodiment of the present invention.

[0021] Figure 3 shows a graph comparing the prediction results of a prediction model according to an embodiment of the present invention with the prediction results of a model that does not use a time-adaptive cohort.

[0022] Figures 4 to 7 are graphs showing the characteristic importance of ROSC in the field in the prediction model predictions at 0, 5, and 10 minutes after EMS cardiopulmonary resuscitation.

[0023] Figure 8 shows a configuration diagram of a device for predicting the outcome of cardiac arrest in a pre-hospital stage according to an embodiment of the present invention.

[0024] Figure 9 illustrates a schematic diagram of a computing environment according to an embodiment of the present invention.

[0025] In a method for predicting the outcome of cardiac arrest in the pre-hospital stage,

[0026] A step of inputting at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data;

[0027] A step of training a prediction model using the input learning data and whether or not spontaneous circulation was restored by cardiopulmonary resuscitation in the pre-hospital stage; and

[0028] A step of predicting whether or not ROSC will occur in the pre-hospital stage by inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of emergency patients into the learned prediction model;

[0029] A method comprising:

[0030] Hereinafter, the present invention will be described in detail with reference to the contents described in the attached drawings. However, the present invention is not limited or restricted by the exemplary embodiments. The same reference numerals in each drawing indicate components that perform substantially the same functions.

[0031] The purpose and effects of the present invention can be naturally understood or made clearer by the following description, and the purpose and effects of the present invention are not limited solely by the following description. Furthermore, in describing the present invention, if a detailed description of known technologies related to the present invention is deemed to unnecessarily obscure the gist of the present invention, such detailed description will be omitted.

[0032] The terminology used herein is merely used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the description of the invention, but do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0033] While terms like "first" and "second" may be used to describe various components, these components should not be limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, a first component could be referred to as a "second component," and similarly, a second component could also be referred to as a "first component."

[0034] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0035] When interpreting components, even if there is no explicit description, it is interpreted as including the margin of error. When describing temporal relationships, for example, when temporal continuity is described with phrases such as "after," "following," "next to," or "before," this also includes cases where the relationship is not continuous, unless "immediately" or "directly" is used.

[0036] Hereinafter, the technical configuration of the present invention will be described in detail with reference to the attached drawings.

[0037] Figure 1 illustrates a flowchart of a method for predicting the outcome of cardiac arrest in a pre-hospital setting according to an embodiment of the present invention. Referring to Figure 1, the method for predicting the outcome of cardiac arrest in a pre-hospital setting may include a step of receiving training data (S100), a step of training a prediction model (S200), and a step of predicting whether or not ROSC will occur in the pre-hospital setting (S300).

[0038] Providing CPR in an ambulance to a patient with out-of-hospital cardiac arrest can degrade the quality of CPR. Therefore, "stay and play" CPR is recommended to save time during transport and provide high-quality CPR. If the survival rate from pre-hospital CPR is determined to be low, rapid transport to a hospital is recommended. Determining whether pre-hospital or in-hospital CPR is appropriate is challenging. Therefore, the method for predicting pre-hospital cardiac arrest outcomes according to an embodiment of the present invention proposes a field ROSC prediction model using pre-hospital input variables along with a time-adaptive cohort.

[0039] FIG. 2 is a diagram illustrating data required for prediction and a target to be predicted in a method for predicting the outcome of cardiac arrest in the pre-hospital stage according to an embodiment of the present invention. Referring to FIG. 2, the step of receiving learning data (S100) may receive as learning data at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of out-of-hospital cardiac arrest patients transported via emergency medical services (EMS). The step of receiving learning data (S100) may additionally receive survival rate until emergency room discharge, 30-day survival rate, and CPC (Cerebral Performance Category) per minute of out-of-hospital cardiac arrest patients transported via emergency medical services (EMS).

[0040] An example of the training data used to develop the present invention is the Pan-Asia Resuscitation Outcomes Study (PAROS) database, which includes patients transported from out-of-hospital cardiac arrest by emergency medical services (EMS) in Asia from 2009 to 2018. This database covers seven Pacific Asian countries. Patients under 18 years of age were excluded from this database, as were patients transported privately or by public transportation rather than by EMS. Patients who were not resuscitated by the EMS team or who experienced cardiac arrest after arrival at the hospital were also excluded from this database. Finally, patients missing one or more of the essential time information, such as the date and time of cardiac arrest and the time of report to the dispatch center, were also excluded from this database.

[0041] The step of receiving learning data (S100) may receive demographic information, such as at least one of city, gender, race, and age.

[0042] The step of receiving learning data (S100) is to receive at least one of location type and time as field environment information, and the location type can be classified into at least one of home, medical facility, public building, nursing home, street, industrial area, transportation center, recreation area, and ambulance.

[0043] The step of receiving learning data (S100) may receive at least one of the following information: pre-hospital CPR information, bystander cardiac arrest information, bystander CPR information, use of an automated external defibrillator (AED) by a first responder, bystander, or emergency medical service (EMS), the patient's initial rhythm, pre-hospital airway securing method, and pre-hospital epinephrine administration.

[0044] The step of training the prediction model (S200) can train the prediction model using the input training data and whether or not spontaneous circulation was restored by cardiopulmonary resuscitation in the pre-hospital stage.

[0045] The step (S200) of training the prediction model involves dividing the pre-hospital CPR time into time periods, dividing the input training data into time periods, and training the prediction model using the time-separated training data. This is for real-time prediction.

[0046] Learning data divided by a certain period of time can be called a "time-adaptive cohort." The above-mentioned period of time may be in minutes, preferably 1 minute, but this is only an example and may be determined as an appropriate time for real-time prediction. The step (S100) of receiving learning data according to an embodiment of the present invention can create a time-adaptive cohort by dividing the pre-hospital CPR time into 1-minute periods. The pre-hospital CPR time can be defined as the time from the ambulance's arrival at the scene to the arrival at the emergency room. In other words, every minute during which CPR continues from the time the ambulance arrives at the scene can be defined as an adaptive cohort.

[0047] The step of training the prediction model (S200) can train the prediction model using a time-adaptive cohort, which means training the prediction model with each data set every minute, and training with multiple data sets over time.

[0048] The step of training the prediction model (S200) does not use the CPR time from the scene to the hospital as training data, but can only be used as a criterion for distinguishing time-adaptive cohorts.

[0049] The step (S200) of training the prediction model can additionally utilize the survival rate until discharge from the emergency room, the 30-day survival rate, and the CPC (Cerebral Performance Category) per minute to train the prediction model. Here, the survival rate until discharge from the emergency room indicates whether the patient was hospitalized, transferred, or discharged for further treatment while still alive. The 30-day survival rate indicates whether the patient survived 30 days after cardiac arrest. The CPC per minute can indicate the predicted result of whether the patient's final CPC is 1 or 2.

[0050] The step of training a prediction model (S200) can use a tree-based ensemble model that shows the best performance in classifying structured data. As a specific example, the step of training a prediction model (S200) can use a random forest, a light gradient boosting machine (LightGBM), or an extreme gradient boosting (XGBoost) as the prediction model. As described below, the algorithm that shows the best performance when comparing the area under the receiver operating characteristic curve (AUROC) is the light gradient boosting machine (LightGBM), so it is preferable to use the gradient boosting machine (LightGBM) in the step of training a prediction model (S200).

[0051] The step (S300) of predicting whether or not ROSC is achieved in the pre-hospital stage can predict whether or not ROSC is achieved in the pre-hospital stage by inputting the emergency patient's demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information into the learned prediction model.

[0052] The step (S300) of predicting ROSC in the pre-hospital stage can additionally predict survival rate until discharge from the emergency room, 30-day survival rate, and CPC (Cerebral Performance Category) per minute.

[0053]

[0054] Below, the simulation results of the present invention are described.

[0055] The demographic and demographic information used in the simulation was from 2009 to 2018. The PAROS database contained 160,263 OHCA patients from 13 Asian countries. Patients under 18 years of age (n=2,495) and patients without adequate time data (n=114) were excluded from the time-adaptive cohort construction, and 157,654 patients were ultimately enrolled in the study.

[0056] The basic characteristics of the variables considered in model building are shown in Table 1 below, and it can be seen that on-site ROSC was achieved in 11,995 out of a total of 160,263 patients.

[0057] ROSCP values ​​at site / waypointNo (N=145,656)Yes (N=11,995)Location type <.001 Home Residential35,687 (24.5%)2,160 (18%) Healthcare685 (0.5%)99 (0.8%) Public / Commercial Buildings3,112 (2.1%)473 (3.9%) Nursing Homes3,519 (2.4%)210 (1.8%) Streets / Highways3,375 (2.3%)290 (2.4%) Industrial757 (0.5%)66 (0.6%) Transportation Centers203 (0.1%)28 (0.2%) Recreational Sites654 (0.4%)107 (0.9%) EMS / Private Ambulance93 (0.1%)15 (0.1%) Other2,936 (2%)255 (2.1%) Unknown None 94,635 (65%) 8,292 (69.1%) Gender <.001 Female 60,354 (41.4%) 4,120 (34.3%) Male 85,302 (58.6%) 7,875 (65.7%) Age <.001 < 60 31,086 (21.3%) 3,064 (25.5%) 60-69 22,371 (15.4%) 2,340 (19.5%) 70-79 34,349 (23.6%) 2,932 (24.4%) >= 80 57,850 (39.7%) 3,659 (30.5%) Reason for arrest .001 Trauma 593 (0.4%) 23 (0.2%) Non-traumatic 4,752 (3.3%) 388 (3.2%) Unknown140,311 (96.3%)11,584 (96.6%) Arrest Witness <.001 Not Witnessed88,633 (60.9%)3,134 (26.1%) EMS / Private Ambulance823 (0.6%)134 (1.1%) Bystander-Healthcare Provider52,813 (36.3%)8,590 (71.6%) Unknown3,387 (2.3%)137 (1.1%) Witness CPR <.001 Yes62,501 (42.9%)6,302 (52.5%) No82,935 (56.9%)5,682 (47.4%) Unknown220 (0.2%)11 (0.1%) Initiate Initial CPR <.001 CPR not initiated9,499 (6.5%)637 (5.3%) Emergency medical technician571 (0.4%)108 (0.9%) Ambulance crew26,803 (18.4%)1,513 (12.6%) Bystander11,398 (7.8%)1,794 (15%) Unknown97,385 (66.9%)7,943 (66.2%) Bystander applied AED <.001 Yes1,513 (1%)338 (2.8%) No123,423 (84.7%)9,881 (82.4%) Unknown20,720 (14.2%)1,776 (14.8%) First arrested rhythm <.001 VF8,532 (5.9%)3,226 (26.9%) VT313 (0.2%)106 (0.9%) PEA23,419 (16.1%)3,917 (32.7%) Asystole94,177 (64.7%)2,627 (21.9%) Unknown19,215 (13.2%)2,119 (17.7%) Out-of-hospital defibrillation <.001 Yes18,412 (12.6%)4,588 (38.2%) No127,239 (87.4%)7,407 (61.8%) Unknown5 (0%)0 (0%) Defibrillation performed by first responders <.001 Yes573 (0.4%)239 (2%) No103,729 (71.2%)8,083 (67.4%) Unknown41,354 (28.4%)3,673 (30.6%) Performed by ambulance crew Defibrillation <.001 Yes11,713 (8%)2,920 (24.3%) No116,652 (80.1%)7,184 (59.9%) Unknown17,291 (11.9%)1,891 (15.8%) Defibrillation administered by bystander-healthcare worker <.001 Yes81 (0.1%)27 (0.2%) No12,145 (8.3%)858 (7.2%) Unknown133,430 (91.6%)11,110 (92.6%) Defibrillation administered by bystander to lying person <.001 Yes142 (0.1%)50 (0.4%) No104,161 (71.5%)8,285 (69.1%) Unknown41,353 (28.4%)3,660 (30.5%) Defibrillation by bystander family members <.001 Yes108 (0.1%)17 (0.1%) No12,145 (8.3%)858 (7.2%) Unknown133,403 (91.6%)11,120 (92.7%) Mechanical CPR device used in EMS / civilian ambulance <.001 Yes6,798 (4.7%)490 (4.1%) No69,018 (47.4%)5,124 (42.7%) Unknown69,840 (47.9%)6,381 (53.2%) Prehospital airway <.001 Yes59,473 (40.8%)5,410 (45.1%) No85,096 (58.4%)6,546 (54.6%) Unknown No 1,087 (0.7%) 39 (0.3%) Advanced airway method type <.001 No airway progress 85,096 (58.4%) 6,546 (54.6%) Oral / nasal ET 10,059 (6.9%) 1,449 (12.1%) Combitube 69 (0%) 7 (0.1%) LMA 19,593 (13.5%) 1,322 (11%) King airway 21,587 (14.8%) 2,084 (17.4%) Other 8,163 (5.6%) 548 (4.6%) Unknown 1,089 (0.7%) 39 (0.3%) Prehospital epinephrine administration <.001 Yes 15,794 (10.8%) 4,420 (36.8%) No 128,482 (88.2%)7,513 (62.6%)Unknown1,380 (0.9%)62 (0.5%) .

[0058] A time-adaptive cohort was created for each minute of CPR performed in the field, and each time-adaptive cohort was randomly divided into a training set and a test set. A predictive model was trained based on this cohort, and the simulation results are shown in [Table 2] below.

[0059] On-site ROSC Emergency room survival 30-day survival CPC 1 or 20 minutes 0.8640.9180.8520.9171 minutes 0.8640.9190.8510.9152 minutes 0.8640.9180.8480.9153 minutes 0.8630.9180.8460.9144 minutes 0.8620.9170.8420.9115 minutes 0.8620.9160.8400.9086 minutes 0.8610.9160.8350.9027 minutes 0.8600.9170.8330.9028 minutes 0.8620.9180.8290.8999 minutes 0.8620.9180.8240.89210 minutes 0.8660.9200.8 210.88711 minutes 0.8670.9210.8220.88212 minutes 0.8680.9230.8240.87113 minutes 0.8700.9240.8210.87114 minutes 0.8670.9240.8170.86915 minutes 0.8720.9260.8190.86416 minutes 0.8700.9290.8110.86617 minutes 0.8680.9290.8070.87018 minutes 0.8660.9300.8020.86419 minutes 0.8680.9310.8050.83520 minutes 0.8670.9320.8030.840

[0060] Table 2 shows the output of the prediction model for each minute from the EMS team's arrival on scene until 20 minutes later. For predicting ROSC at the scene, the prediction model achieved an area under the curve (AUROC) score of 0.85 to 0.87. For predicting survival to emergency department discharge, the prediction model achieved an AUROC score of 0.91 to 0.93, and for predicting 30-day survival, the prediction model achieved an AUROC score of 0.80 to 0.86. For predicting CPC, the prediction model had an AUROC score of 0.84 to 0.91.

[0061] Figure 3 illustrates a graph comparing the prediction results of a prediction model according to an embodiment of the present invention with the prediction results of a model that does not use a time-adaptive cohort. Referring to Figure 3, it can be seen that the prediction model according to an embodiment of the present invention exhibits a lower predicted survival rate than the model that does not use a time-adaptive cohort.

[0062] Figures 4 to 7 are graphs showing the characteristic importance of ROSC in the field in the prediction model predictions at 0, 5, and 10 minutes after EMS cardiopulmonary resuscitation.

[0063] Referring to Figures 4 through 7, age, ambulance arrival time, and initial rhythm were the most influential factors in each time cohort. However, the importance of features varied over time. For example, in the initial 0-minute prediction model, epinephrine administration was the fourth most important feature. In the 5-minute prediction model, the fourth most important aspect was prehospital airway management. In the 10-minute prediction model, the person observing the cardiac arrest was the fourth most important feature. As the prediction model according to the embodiment of the present invention is built minute by minute, the importance values ​​of specific features also varied over time.

[0064] In conclusion, the prediction model according to the present invention was able to predict ROSC in the field with an area under the curve (AUROC) score of 0.8 or higher based on a time-adaptive cohort. The likelihood of ROSC in the field is crucial for patient prognosis, and the prediction model according to the present invention can provide the information necessary for decisions such as when to initiate transport and where to transport patients by predicting potential ROSC in the field and long-term outcomes.

[0065] The predictive model according to the present invention provides clear predictions every minute, with new predictions generated every minute. These real-time predictions can assist emergency response teams in making real-time decisions at the scene of an arrest. Since the predictive model according to the present invention is trained using machine learning, it predicts outcomes similar to real-world scenarios, and when compared to previous studies, it was confirmed to produce similar results. Notably, as shown in Figure 3, the survival rate in the predictive model decreases sharply over time, providing further evidence that it reflects the real world.

[0066] FIG. 8 illustrates a configuration diagram of a device (100) for predicting the outcome of a pre-hospital cardiac arrest, which is another embodiment of the present invention. Referring to FIG. 8, the configuration of the device (100) for predicting the outcome of a pre-hospital cardiac arrest is merely a simplified example. In one embodiment of the present invention, the device (100) for predicting the outcome of a pre-hospital cardiac arrest may include other components for implementing the computing environment of the device (100), and only some of the disclosed components may constitute the device (100).

[0067] A device (100) for predicting the outcome of cardiac arrest in the pre-hospital stage may include a processor (110) including one or more cores, a memory (120), and a network (130).

[0068] The processor (110) may be configured with one or more cores, and may include a processor for data analysis and deep learning, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU) of a computing device. The processor (110) may read a computer program stored in the memory (120) and perform data processing for machine learning according to an embodiment of the present disclosure. According to an embodiment of the present disclosure, the processor (110) may perform operations for learning a neural network. The processor (110) may perform calculations for learning a neural network, such as processing input data for learning in deep learning (DL), extracting features from input data, calculating errors, and updating weights of a neural network using backpropagation. At least one of the CPU, GPGPU, and TPU of the processor (110) may process learning of a network function. For example, a CPU and a GPGPU can jointly process network function learning and data classification using network functions. Furthermore, in one embodiment of the present disclosure, processors of multiple computing devices can be jointly used to process network function learning and data classification using network functions. Furthermore, a computer program executed on a computing device according to one embodiment of the present disclosure may be a CPU, GPGPU, or TPU executable program.

[0069] The processor (110) may receive at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data. The processor (100) may perform the step (S100) of receiving the aforementioned learning data.

[0070] The processor (110) can train a prediction model using the input learning data and whether spontaneous circulation was restored through cardiopulmonary resuscitation in the pre-hospital stage. The processor (110) can perform the step (S200) of training the aforementioned prediction model.

[0071] The processor (110) can predict whether ROSC will occur in the pre-hospital stage by inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of the emergency patient into the learned prediction model. The processor (100) can perform a step (S300) of predicting whether ROSC will occur in the pre-hospital stage.

[0072] The memory (120) can store any form of information generated or determined by the processor (110) and any form of information received by the network (130).

[0073] The memory (120) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. The computing device (100) may also operate in relation to web storage that performs the storage function of the memory (120) on the internet. The description of the above-described memory is merely an example, and the present disclosure is not limited thereto.

[0074] The network (130) may use any known wired or wireless communication system. The network (130) may receive endoscope images and the like from related devices or systems.

[0075] The network (130) can transmit and receive information, user interfaces, etc. processed by the processor (110) through communication with other terminals. For example, the network (130) can provide a user interface generated by the processor (100) to a client (e.g., a user terminal). In addition, the network (130) can receive external input from a user authorized as a client and transmit it to the processor (110). At this time, the processor (110) can process operations such as outputting, modifying, changing, and adding information provided through the user interface based on the external input of the user received from the network (130).

[0076] Meanwhile, a device (100) for predicting the outcome of cardiac arrest in a pre-hospital setting according to one embodiment of the present disclosure may include a server as a computing system that transmits and receives information through communication with a client. In this case, the client may be any type of terminal capable of accessing the server.

[0077] In an additional embodiment, a device (100) for predicting cardiac arrest outcomes in the pre-hospital stage may include any type of terminal that receives data resources generated from any server and performs additional information processing.

[0078] Another embodiment of the present invention, a computer program for predicting the outcome of cardiac arrest in the pre-hospital stage, may include an operation for receiving training data, an operation for training a prediction model, and an operation for predicting whether or not ROSC has occurred in the pre-hospital stage.

[0079] The actions input as learning data may include at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information. The actions input as learning data refer to the actions performed in the aforementioned step (S100) of inputting learning data.

[0080] The operation of training the predictive model can be performed using the input training data and the presence or absence of ROSC following CPR in the pre-hospital setting. The operation of training the predictive model refers to the operation performed in the aforementioned step (S200) of training the predictive model.

[0081] The operation of predicting ROSC in the pre-hospital stage can predict ROSC in the pre-hospital stage by inputting the emergency patient's demographic information, field environment information, and pre-hospital CPR information into a learned prediction model. The operation of predicting ROSC in the pre-hospital stage refers to the operation performed in the step (S300) of predicting ROSC in the pre-hospital stage described above.

[0082] Figure 9 illustrates a schematic diagram of a computing environment according to an embodiment of the present invention.

[0083] Although the present disclosure has been described above as being generally implemented by a computing device, those skilled in the art will appreciate that the present disclosure may be implemented in combination with computer-executable instructions and / or other program modules that may be executed on one or more computers and / or as a combination of hardware and software.

[0084] Generally, program modules include routines, programs, components, data structures, and the like that perform particular tasks or implement particular abstract data types. Furthermore, those skilled in the art will appreciate that the methods of the present disclosure can be implemented with other computer system configurations, including single-processor or multiprocessor computer systems, minicomputers, mainframe computers, as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, and the like, each of which may be operatively connected to one or more associated devices.

[0085] The described embodiments of the present disclosure can also be practiced in distributed computing environments, where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0086] Computers typically include a variety of computer-readable media. Computer-readable media can be any media that can be accessed by a computer, and includes both volatile and nonvolatile media, transitory and non-transitory media, removable and non-removable media. By way of example, and not limitation, computer-readable media can include computer-readable storage media and computer-readable transmission media. Computer-readable storage media includes both volatile and nonvolatile media, transitory and non-transitory media, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be accessed by a computer and used to store the desired information.

[0087] Computer-readable transmission media typically includes any information delivery media that embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave or other transport mechanism. The term modulated data signal means a signal that has one or more of its characteristics set or changed so as to encode information in the signal. By way of example, and not limitation, computer-readable transmission media includes wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, or other wireless media. Combinations of any of the above are also intended to be included within the scope of computer-readable transmission media.

[0088] An exemplary environment for implementing various aspects of the present disclosure is illustrated, including a computer (1000), which includes a processing unit (1020), a system memory (1030), and a system bus (1010). The system bus (1010) connects system components, including but not limited to the system memory (1030), to the processing unit (1020). The processing unit (1020) may be any of a variety of commercially available processors. Dual processors and other multiprocessor architectures may also be utilized as the processing unit (1020).

[0089] The system bus (1010) may be any of several types of bus structures that may be additionally interconnected to a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. The system memory (1030) includes read-only memory (ROM) (1034) and random access memory (RAM) (1032). A basic input / output system (BIOS) is stored in non-volatile memory (1034), such as ROM, EPROM, or EEPROM, and includes basic routines that help transfer information between components within the computer (1000), such as during start-up. The RAM (1032) may also include high-speed RAM, such as static RAM, for caching data.

[0090] The computer (1000) also includes an internal hard disk drive (HDD) (1050) (e.g., EIDE, SATA) - which may also be configured for external use within a suitable chassis (not shown), a magnetic floppy disk drive (FDD) (1060) (e.g., for reading from or writing to removable diskettes), and an optical disk drive (1070) (e.g., for reading from or writing to CD-ROM disks or other high-capacity optical media such as DVDs). The hard disk drive (1050), the magnetic disk drive (1060), and the optical disk drive (1070) may be connected to the system bus (1010) by a hard disk drive interface, a magnetic disk drive interface, and an optical drive interface, respectively. Interfaces for implementing external drives include at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

[0091] These drives and their associated computer-readable media provide non-volatile storage of data, data structures, computer-executable instructions, and the like. In the case of the computer (1000), the drives and media correspond to storing any data in a suitable digital format. While the description of computer-readable media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art will appreciate that other types of computer-readable media, such as zip drives, magnetic cassettes, flash memory cards, cartridges, and the like, may also be used in the exemplary operating environment, and that any such media may contain computer-executable instructions for performing the methods of the present disclosure.

[0092] A number of program modules, including an operating system (1092), one or more application programs (1094), other program modules (1096), and a database (1098), may be stored in the drive and RAM (1032). All or portions of the operating system, applications, modules, and / or data may also be cached in RAM (1032). It will be appreciated that the present disclosure may be implemented in various commercially available operating systems or combinations of operating systems.

[0093] A user may enter commands and information into the computer (1000) via one or more wired / wireless input devices (1042), such as a keyboard and a pointing device such as a mouse. Other input devices (not shown) may include a microphone, an IR remote control, a joystick, a game pad, a stylus pen, a touch screen, and the like. These and other input devices are often connected to the processing unit (1020) via an input / output interface (1040) that is connected to the system bus (1010), but may be connected by other interfaces such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR interface, and the like.

[0094] A monitor or other type of display device is also connected to the system bus (1010) via an interface such as a video adapter. In addition to the monitor, the computer typically includes other peripheral output devices (not shown) such as speakers, a printer, and so on.

[0095] The computer (1000) may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) (1082), via wired and / or wireless communications. The remote computer(s) (1082) may be a workstation, a computing device computer, a router, a personal computer, a portable computer, a microprocessor-based entertainment device, a peer device, or other conventional network node, and may generally include many or all of the components described for the computer (1000). The logical connections include wired / wireless connections to a local area network (LAN) and / or a larger network, such as a wide area network (WAN). Such LAN and WAN networking environments are common in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which may be connected to a worldwide computer network, such as the Internet.

[0096] When used in a LAN networking environment, the computer (1000) is connected to a local network (not shown) via a wired and / or wireless communication network interface or adapter (not shown). The adapter (not shown) may facilitate wired or wireless communication to the LAN (not shown), which may also include a wireless access point installed therein for communicating with the wireless adapter (not shown). When used in a WAN networking environment, the computer (1000) may include a modem (not shown), be connected to a communication computing device on the WAN (not shown), or have other means for establishing communications over the WAN (not shown), such as via the Internet. The modem (not shown), which may be internal or external and wired or wireless, is connected to the system bus (1010) via a serial port interface (not shown). In a networked environment, program modules described for the computer (1000), or portions thereof, may be stored in a remote memory / storage device (not shown). It will be appreciated that the network connections shown are exemplary and that other means of establishing a communications link between computers may be used.

[0097] The computer (1000) operates to communicate with any wireless device or object that is arranged and operates via wireless communication, such as a printer, a scanner, a desktop and / or portable computer, a portable data assistant (PDA), a communication satellite, any equipment or location associated with a radio-detectable tag, and a telephone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, the communication may be a predefined structure, as in a conventional network, or simply an ad hoc communication between at least two devices.

[0098] Wi-Fi (Wireless Fidelity) enables connections to the Internet and other devices without wires. Wi-Fi is a wireless technology that allows devices, such as computers, to send and receive data anywhere within the coverage area of ​​a base station, both indoors and outdoors, similar to cell phones. Wi-Fi networks use wireless technologies called IEEE 802.11 (a, b, g, etc.) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks can operate in the unlicensed 2.4 and 5 GHz radio bands, at data rates of, for example, 11 Mbps (802.11a) or 54 Mbps (802.11b), or in products that include both bands (dual-band).

[0099] Those skilled in the art will appreciate that information and signals may be represented using any of a variety of different technologies and techniques. For example, the data, instructions, commands, information, signals, bits, symbols, and chips referenced in the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0100] Those skilled in the art will appreciate that the various illustrative logical blocks, modules, processors, means, circuits, and model steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, various forms of programs or design code (referred to herein, for convenience, as software), or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

[0101] The various embodiments presented herein can be implemented as a method, apparatus, or article of manufacture using standard programming and / or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical disks (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Furthermore, various storage media presented herein include one or more devices and / or other machine-readable media for storing information.

[0102] It should be understood that the specific order or hierarchy of steps in the presented processes is merely an example of exemplary approaches. It should be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present disclosure based on design priorities. The appended method claims provide elements of various steps in a sample order, but are not intended to be limited to the specific order or hierarchy presented.

[0103] The description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the embodiments disclosed herein, but is to be construed in the broadest scope consistent with the principles and novel features disclosed herein.

[0104] The embodiments of the present invention described above are not implemented solely through devices and methods. They may also be implemented through programs that implement functions corresponding to the configurations of the embodiments of the present invention, or through recording media containing such programs. Such recording media may be executed not only on servers but also on user terminals.

[0105] Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

[0106] The present invention aims to provide a method, device, and computer program for predicting the outcome of cardiac arrest at a predetermined time interval in a pre-hospital stage, which predicts the return of spontaneous circulation and further prognosis in real time during emergency resuscitation attempts.

Claims

1. In a method for predicting the outcome of cardiac arrest in the pre-hospital stage, A step of inputting at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data; A step of training a prediction model using the input learning data and whether or not spontaneous circulation was restored by cardiopulmonary resuscitation in the pre-hospital stage; and A step of predicting whether or not ROSC will occur in the pre-hospital stage by inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of emergency patients into the learned prediction model; A method comprising:

2. In paragraph 1, The step of inputting the above learning data is: A method for receiving demographic information, including at least one of city, gender, race, and age.

3. In paragraph 1, The step of inputting the above learning data is: As field environment information, at least one of location type and time is entered, A method wherein the above location type is classified as at least one of a home, a medical facility, a public building, a nursing home, a street, an industrial area, a transportation center, a recreational location, and an ambulance.

4. In paragraph 1, The step of inputting the above learning data is: A method for receiving at least one of bystander cardiac arrest information, bystander CPR information, use of an automated external defibrillator (AED) by a first responder, bystander, or Emergency Medical Service (EMS), the patient's initial rhythm, prehospital airway establishment method, and prehospital epinephrine administration as prehospital cardiopulmonary resuscitation information.

5. In paragraph 1, The step of training the above prediction model is: A method for dividing cardiopulmonary resuscitation time in the pre-hospital stage into a certain number of hours, dividing input learning data into certain number of hours, and training the prediction model with the learning data divided into certain number of hours.

6. In paragraph 6, The CPR time in the above pre-hospital stage is: A method defined as the time from the time an ambulance arrives at the scene until the time it arrives at the emergency room.

7. In paragraph 1, The step of training the above prediction model is: The above prediction model was trained by additionally using survival rate until discharge from the emergency room, 30-day survival rate, and CPC (Cerebral Performance Category) per minute. The step of predicting whether or not there will be a return of spontaneous circulation in the above pre-hospital stage is as follows: A method for additionally predicting survival to emergency room discharge, 30-day survival, and Cerebral Performance Category (CPC) per minute.

8. In paragraph 1, The step of training the above prediction model is: A method using a random forest, a light gradient boosting machine (LightGBM), or an extreme gradient boosting (XGBoost) as the above prediction model.

9. As a device for predicting the outcome of cardiac arrest in the pre-hospital stage, a processor comprising one or more cores; and memory; Including, The above processor, At least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information is input as learning data. A prediction model is trained using the input learning data and whether or not spontaneous circulation was restored by cardiopulmonary resuscitation in the pre-hospital stage, and A device that predicts whether or not ROSC will occur in the pre-hospital stage by inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of emergency patients into a learned predictive model.

10. A computer program stored in a computer-readable storage medium and including commands that cause a computer to perform the following operations, wherein the operations are: An action of inputting at least one of demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information as learning data; An operation to train a prediction model using the input learning data and whether or not spontaneous circulation was restored by cardiopulmonary resuscitation in the pre-hospital stage; and A computer program stored in a computer-readable storage medium, comprising: an operation for predicting whether or not ROSC is achieved in the pre-hospital stage by inputting demographic information, field environment information, and pre-hospital cardiopulmonary resuscitation information of an emergency patient into a learned predictive model;

Images