Method and device for predicting myocardial injury after noncardiac surgery

Abstract

Provided according to exemplary embodiments of the present invention is a method for predicting myocardial injury after noncardiac surgery, which is implemented by at least one processor. The method for predicting myocardial injury after noncardiac surgery may include the steps of: acquiring preoperative and intraoperative information of a subject undergoing noncardiac surgery; and determining the subject's risk of myocardial injury after noncardiac surgery on the basis of the acquired preoperative and intraoperative information of the subject undergoing noncardiac surgery.

Description

Method and device for predicting myocardial damage after noncardiac surgery

[0001] The present invention relates to a method and device for predicting myocardial damage after non-cardiac surgery. More specifically, the present invention relates to a method and device for predicting a subject's risk of myocardial damage after non-cardiac surgery based on information about the subject.

[0002] It is estimated that more than 300 million people worldwide undergo surgery each year. It has been reported that approximately 85% of major surgeries are non-cardiac surgeries.

[0003] Meanwhile, previous studies have reported that myocardial injury after noncardiac surgery (MINS) can increase patient mortality. For example, the criterion for diagnosing MINS is an increase in cardiac troponin I (cTnI) levels within 30 days after noncardiac surgery. In other words, patients with perioperative myocardial injury (PMI) who had elevated cTnI levels after noncardiac surgery had a higher risk of death than those whose cTnI levels were not elevated.

[0004] MINS is usually asymptomatic and is caused by an imbalance between myocardial oxygen supply and demand. MINS typically does not present with symptoms such as chest pain, chest tightness, or discomfort. Therefore, diagnosing MINS is challenging without routine monitoring of cTnI levels. Recent guidelines recommend measuring cTnI levels 48 hours after surgery, as MINS typically occurs within 3 days of surgery and is asymptomatic. However, testing cTnI levels after non-cardiac surgery is unrealistic due to cost and technical accessibility.

[0005] Meanwhile, tools used to predict the risk of perioperative non-cardiac surgery include the Revised Cardiac Risk Index (RCRI), the American College of Surgeons National Surgical Quality Improvement Program (NSQIP), the Myocardial Infarction and Cardiac Arrest (MICA) index, and the American College of Surgeons NSQIP Surgical Risk Calculator. However, the aforementioned tools are not specific tests for predicting MINS.

[0006] In addition, U.S. Patent Publication No. 2023-0148456 (prior document 1) discloses a method for predicting the risk of heart disease, U.S. Patent Publication No. 2021-0148931 (prior document 2) discloses a method for predicting the risk of cardiovascular disease after non-cardiac surgery, and Japanese Patent Publication No. 2022-521390 (prior document 3) discloses a method for predicting MINS.

[0007] However, considering that the incidence of MINS is approximately 20% and a significant number of cases can lead to myocardial infarction, cardiac arrest, and even death, it is necessary to develop a prediction model that can more accurately predict the incidence of MINS.

[0008] One object of the present invention is to provide a method and device for predicting myocardial injury after noncardiac surgery (MINS) in a subject who has undergone noncardiac surgery.

[0009] A method for predicting myocardial injury after noncardiac surgery (MINS), executed by at least one processor according to exemplary embodiments of the present invention, may include the steps of: acquiring information about a subject before noncardiac surgery and information about a subject during noncardiac surgery; and determining a risk of MINS of the subject based on the information about the subject before noncardiac surgery and the information about the subject during noncardiac surgery.

[0010] In one embodiment, the information of the subject prior to the non-cardiac surgery may include at least one of: basic information of the subject; diagnostic information of the subject prior to the non-cardiac surgery; and information related to the non-cardiac surgery.

[0011] In one embodiment, the basic information of the subject may include at least one of age, gender, height, and weight.

[0012] In one embodiment, the diagnostic information of the subject prior to the non-cardiac surgery may include at least one of the presence or absence of diabetes, the presence or absence of hypertension, the presence or absence of dyslipidemia, the presence or absence of coronary artery disease, the presence or absence of peripheral artery disease, the presence or absence of cerebrovascular disease, the presence or absence of kidney disease, the presence or absence of heart failure, the presence or absence of atrial fibrillation, the presence or absence of aortic valve stenosis, the presence or absence of chronic obstructive pulmonary disease, the presence or absence of dialysis, the presence or absence of anemia, the blood hemoglobin level, the blood creatinine level, and the renal glomerular filtration rate.

[0013] In one embodiment, the non-cardiac surgery related information may include at least one of the type of non-cardiac surgery, whether the non-cardiac surgery is an emergency surgery, and whether the non-cardiac surgery includes vascular surgery.

[0014] In one embodiment, the information during the non-cardiac surgery may include at least one of: the operation time of the non-cardiac surgery; information related to blood transfusion during the non-cardiac surgery; information related to hypotension during the non-cardiac surgery; and information related to heart rate during the non-cardiac surgery.

[0015] In one embodiment, the information related to blood transfusion during the non-cardiac surgery may include at least one of whether a blood transfusion is performed during the non-cardiac surgery and the amount of blood transfusion during the non-cardiac surgery.

[0016] In one embodiment, the heart rate related information during the non-cardiac surgery may include tachycardia related information during the non-cardiac surgery.

[0017] In one embodiment, the information related to hypotension during the non-cardiac surgery may include at least one of: information related to the number of occurrences of hypotension during the non-cardiac surgery; information related to the duration of hypotension occurring during the non-cardiac surgery; and information related to the severity of hypotension occurring during the non-cardiac surgery.

[0018] In one embodiment, the information related to the number of occurrences of hypotension during the non-cardiac surgery may include the total number of hypotension events that occurred during the non-cardiac surgery.

[0019] In one embodiment, the information related to the duration of hypotension occurring during the non-cardiac surgery may include at least one of: the total duration of hypotension occurring during the non-cardiac surgery; and the average duration of hypotension occurring during the non-cardiac surgery.

[0020] In one embodiment, the severity of hypotension occurring during non-cardiac surgery may include at least one of: an average of mean arterial pressure (MAP) during the total occurrence period of hypotension events during the non-cardiac surgery; an area of ​​an area under a hypotension reference value line among areas between a hypotension reference value line and a MAP curve in a MAP (y-axis)-time (x-axis) graph acquired during the non-cardiac surgery; a time-weighted average of MAP during the non-cardiac surgery; and a total number of occurrences of events in which MAP is below a severe hypotension threshold value during the non-cardiac surgery. The severe hypotension threshold value may be a value satisfying 40 mmHg or more and 60 mmHg or less.

[0021] In one embodiment, the step of determining the MINS risk may include using an artificial intelligence model learned to predict the MINS risk of the subject based on information about the subject before the non-cardiac surgery and information about the subject during the non-cardiac surgery.

[0022] In one embodiment, the artificial intelligence model may include a machine learning or artificial neural network-based classification model.

[0023] In one embodiment, the machine learning or artificial neural network-based classification model may include at least one of a logistic regression classification model and a multilayer perceptron classification model.

[0024] In one embodiment, when the MINS risk of the subject is defined as the first MINS risk, if the first MINS risk of the subject is less than a predetermined value, the method may further include: obtaining information about the subject after non-cardiac surgery; and additionally determining a second MINS risk of the subject based on the first MINS risk of the subject and the non-cardiac surgery information.

[0025] In one embodiment, the post-cardiac surgery information includes post-cardiac surgery questionnaire information for the subject, and the questionnaire information may include response data of the subject to a plurality of questionnaires.

[0026] According to exemplary embodiments of the present invention, a device for predicting myocardial damage after non-cardiac surgery may include a communication unit; a memory; and at least one processor connected to the communication unit and the memory.

[0027] The at least one processor may be controlled to obtain information about the subject before non-cardiac surgery and information about the subject during non-cardiac surgery, and determine the MINS risk of the subject based on the information about the subject before non-cardiac surgery and information about the subject during non-cardiac surgery.

[0028] According to exemplary embodiments of the present invention, an application program stored in a recording medium may be provided to execute the above-described method when operated by at least one processor.

[0029] According to exemplary embodiments of the present invention, a method and device can be provided that can more accurately predict myocardial injury after noncardiac surgery (MINS) in a subject.

[0030] According to exemplary embodiments of the present invention, a method and device can be provided for predicting a patient's MINS after non-cardiac surgery without requiring additional testing (e.g., cardiac troponin level testing). Accordingly, subjects requiring additional MINS testing can be selectively identified and guided to undergo the additional testing. This allows for efficient management of patients undergoing non-cardiac surgery, and can further prevent the occurrence of diseases and deaths associated with MINS.

[0031] FIG. 1 illustrates the structure of a system for predicting myocardial damage after non-cardiac surgery in a subject according to one embodiment of the present invention.

[0032] Figure 2 illustrates the structure of a device according to one embodiment of the present invention.

[0033] FIG. 3 illustrates a methodology for predicting myocardial damage after non-cardiac surgery according to one embodiment of the present invention.

[0034] FIG. 4 illustrates an operational flowchart for predicting myocardial damage after non-cardiac surgery according to one embodiment of the present invention.

[0035] FIG. 5 illustrates an operational flowchart for predicting myocardial damage after non-cardiac surgery according to another embodiment of the present invention.

[0036] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below, along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms.

[0037] To clearly explain embodiments of the present invention, portions irrelevant to the description may be omitted. Furthermore, when describing embodiments of the present invention, if a detailed description of a related known configuration or function is deemed to obscure the gist or description of the present invention, a detailed description thereof may be omitted.

[0038] In describing components in this specification, terms such as "first," "second," etc. may be used. The aforementioned terms are intended to distinguish one component from another for convenience of description, and unless otherwise specified, the nature, order, etc. of the components are not limited by the aforementioned terms.

[0039] In each of the steps mentioned in this specification, unless the context clearly dictates a specific order, the steps may be performed in a different order than stated. That is, the steps may be performed in the same order as stated, may be performed substantially simultaneously, or may be performed in the opposite order.

[0040] In this specification, "and / or" may mean each of the listed components and any combination of two or more of the listed components. For example, "A, B and / or C" may be used with the same meaning as "at least one of A, B, and C."

[0041] According to exemplary embodiments of the present invention, methods, devices, and services for predicting myocardial injury after noncardiac surgery (MINS) in a subject may be provided. Hereinafter, for convenience of explanation, "myocardial injury after noncardiac surgery" may be abbreviated as "MINS."

[0042] Figure 1 illustrates the structure of a MINS prediction system according to one embodiment of the present invention.

[0043] Referring to FIG. 1, the MINS prediction system may include a user device (101) and a server device (102).

[0044] The user device (101) may be a device used by the user to check and manage information about the subject (e.g., personal information, information before non-cardiac surgery, information during non-cardiac surgery, etc.). The user device (101) may connect to the server device (102) via a network and interact with the server device (102). For example, the user device (101) may provide information about the subject for MINS prediction to the server device (102) and receive the MINS prediction result.

[0045] For example, the user device (101) may be a device of various forms, such as an electronic device such as a smartphone, a laptop, a desktop computer, or a tablet PC.

[0046] The server device (102) may be a device of a service provider that provides a MINS prediction service. The server device (102) may transmit and receive information about the subject, etc., to and from the user device (101) via a network. For example, the server device (102) may store information about multiple subjects, a MINS prediction model, etc., and may provide a function for predicting the MINS of the subject in the form of a platform.

[0047] Although FIG. 1 illustrates only one user device (101), multiple devices including the user device (101) can access the server (102) and use the MINS prediction service. However, the scope of accessible subject information may vary depending on the authority granted to the user device (101).

[0048] The relationship between the server device (102) and the user device (101) as illustrated in FIG. 1 can be established when the MINS prediction service is provided in the form of a platform. According to another embodiment of the present invention, the MINS prediction service may be provided in the form of a non-network-based program rather than a platform. In this case, unlike as illustrated in FIG. 1, the MINS prediction service may be provided by a program or application installed on the user device (101) without the server device (102). Furthermore, the server device (102) may be the entity providing the program or application.

[0049] FIG. 2 illustrates the structure of a device according to one embodiment of the present invention. FIG. 2 illustrates an example of the structure of a user device (101) or a server device (102) of FIG. 1. In this specification, the user device (10) and the server device (102) may be collectively referred to as a "device."

[0050] Referring to FIG. 2, the device may include a control unit (201), a communication unit (202), and a storage unit (203).

[0051] The control unit (201) can control the overall functions and operations of the device. The control unit (201) can control components of the device, for example, the communication unit (202), the storage unit (203), and at least one other component. That is, the control unit (201) can provide information or data necessary for the operation of the components of the device, and perform operations based on information or data generated or managed by the components. For example, the control unit (201) can include at least one processor, at least one circuit, etc. For example, the at least one processor can include at least one of a central processing unit (CPU), a graphics processing unit (GPU), and a neural network processing unit (NPU). The control unit (201) can perform necessary controls so that the device operates according to various embodiments described herein. For example, the control unit (201) can control the operation of the device by executing software, programs, or commands stored in the storage unit (203).

[0052] The communication unit (202) can perform functions for transmitting and receiving signals with other devices. The communication unit (202) performs wired or wireless communication and can process signals according to the control of the control unit (201). For example, the communication unit (202) may include an RF circuit, an antenna, etc. for wireless communication, or a connection terminal, a modem, a driver module, etc. for wired communication. For example, the communication unit (202) may support at least one of various communication protocols, such as cellular communication such as LTE and 5G, short-range wireless communication such as WiFi and Bluetooth, and short-range wired communication such as Ethernet.

[0053] The storage unit (203) can store data used in the device, software for the operation of the device, programs, and commands. In addition, the storage unit (203) can provide stored data under the control of the control unit (201). In addition, the storage unit (203) can store applications, drivers, etc. to be driven by the control unit (201). For example, the storage unit (203) can include RAM such as DRAM, SRAM, etc.; ROM; EEPROM; HDD; SSD; flash storage means, etc.

[0054] Although not shown in FIG. 2, the device may further include at least one of a power supply, a display device, an input device, and an output device, depending on the type of device. For example, if the device is a user device (101), the device may further include a display device (e.g., a display), an input device (e.g., a touch panel, a button), and a power supply (e.g., a battery, a power supply).

[0055] Figure 3 illustrates a methodology for MINS prediction according to one embodiment of the present invention.

[0056] Referring to FIG. 3, MINS prediction according to one embodiment of the present invention can be performed based on acquired information about a subject (i.e., a patient). The server device (102) acquires data about the subject from various sources, including the user device (101), and analyzes information contained in the acquired data to predict MINS. For example, the server device (102) can predict the MINS risk.

[0057] The subject's information used to predict MINS risk may include information about the subject before and during non-cardiac surgery. Specifically, the server device (102) acquires information about the subject before and during non-cardiac surgery, and based on the acquired information, can predict the subject's MINS risk. By utilizing both information about the subject before and during non-cardiac surgery, the accuracy of predicting MINS risk can be significantly improved.

[0058] For example, information about a subject prior to non-cardiac surgery may refer to information about the subject that is known even before the non-cardiac surgery, and information during non-cardiac surgery may refer to information about the subject that is known only during the non-cardiac surgery (e.g., phenomena occurring to the subject during the non-cardiac surgery). Specific examples of information before non-cardiac surgery and information during non-cardiac surgery are described below.

[0059] 'MINS risk (i.e., risk of myocardial damage after noncardiac surgery)' can mean 'the risk of developing myocardial damage or a disease related to myocardial damage after noncardiac surgery.' For example, 'MINS risk' can mean the likelihood of developing myocardial damage or a disease related to myocardial damage after noncardiac surgery expressed as a score, probability, grade, or a combination thereof.

[0060] For example, the "MINS Risk" score can be used as data to determine medical interventions (e.g., additional testing, clinical treatment direction, etc.) for patients undergoing non-cardiac surgery. For example, the "MINS Risk" score can be used to determine whether to order or recommend additional testing (e.g., cardiac troponin level testing) for patients.

[0061] FIG. 4 illustrates a flowchart of operations for predicting myocardial injury (MINS) after non-cardiac surgery according to one embodiment of the present invention. FIG. 4 illustrates operations performed by a server device (102). However, as previously described, the MINS prediction service may be provided by a program or application installed on a user device (101) without a server device (102). In this case, the server device (102) described below may refer to the user device (101).

[0062] Referring to FIG. 4, the server device (102) can obtain information about a subject (i.e., a patient) before and during non-cardiac surgery (S401).

[0063] For example, information about a subject before and during non-cardiac surgery may be received by a server device (102) via a network. For example, information about a subject before and during non-cardiac surgery may be received from a user device (101), a third-party device used during the subject's non-cardiac surgery, etc.

[0064] In one embodiment, the subject's pre-cardiac surgery information may include at least one of the subject's basic information, the subject's pre-cardiac surgery diagnostic information, and the subject's non-cardiac surgery related information.

[0065] In some embodiments, the basic information of the subject may include at least one, at least two, or at least three of age, gender, height, and weight.

[0066] In some embodiments, the subject's pre-operative diagnostic information may include at least one, at least three, at least five, at least seven, or at least ten of the following: presence of diabetes, presence of hypertension, presence of dyslipidemia, presence of coronary artery disease, presence of peripheral artery disease, presence of cerebrovascular disease, presence of kidney disease (e.g., renal failure, etc.), presence of heart failure, presence of atrial fibrillation, presence of aortic stenosis, presence of chronic obstructive pulmonary disease, presence of dialysis patient, presence of anemia, blood hemoglobin level, blood creatinine level, and renal glomerular filtration rate.

[0067] Here, we briefly explain the correlation between the aforementioned preoperative diagnostic information and MINS. i) Type 2 diabetes mellitus is associated with insulin resistance, which can lead to cardiovascular disease, etc. Therefore, diabetes may be associated with MINS. ii) Increased LDL cholesterol can lead to atherosclerotic plaques and premature cardiovascular disease, etc. Therefore, dyslipidemia may be associated with MINS. iii) Decreased renal function may increase the risk of MINS. Therefore, renal disease, renal failure, glomerular filtration rate, dialysis, and serum creatinine levels may be associated with MINS. iv) Chronic obstructive pulmonary disease (COPD) may induce oxygenated hemoglobin desaturation and sympathetic hyperactivity during and / or after noncardiac surgery, leading to an imbalance in myocardial oxygen supply and demand (i.e., an increased risk of MINS). Therefore, COPD may be associated with MINS. v) Anemia can cause an imbalance in the oxygen supply and demand of the myocardium. Therefore, anemia and blood hemoglobin levels may be associated with MINS. The correlation between other preoperative diagnostic information and MINS is omitted.

[0068] In some embodiments, the non-cardiac surgery related information may include at least one or at least two of the type of the non-cardiac surgery, whether the non-cardiac surgery is an emergency surgery, and whether the non-cardiac surgery includes vascular surgery.

[0069] The correlation between the aforementioned non-cardiac surgery-related information and MINS is briefly explained. Perioperative cardiac risk may be related to the type of non-cardiac surgery, and in particular, vascular surgery has a higher risk of thrombosis and occlusion due to vascular damage than other surgeries. Therefore, the inclusion of vascular surgery may be more closely related to MINS. Furthermore, it has been confirmed that emergency surgery (e.g., surgery for acute conditions) may increase the MINS risk. According to exemplary embodiments of the present invention, by predicting MINS risk using non-cardiac surgery-related information, the accuracy of MINS risk prediction, etc. can be further improved.

[0070] In one embodiment, information regarding the subject's non-cardiac surgery may include at least one, at least two, or at least three of the following: surgery time; information regarding blood transfusion during non-cardiac surgery; information regarding hypotension during non-cardiac surgery; and information regarding heart rate during non-cardiac surgery. In this case, the accuracy of prediction of MINS risk, etc., may be further improved.

[0071] In some embodiments, information related to blood transfusion during non-cardiac surgery may include at least one of whether a blood transfusion is performed during non-cardiac surgery and the amount of blood transfusion during non-cardiac surgery.

[0072] In some embodiments, information related to hypotension during non-cardiac surgery may include at least one of: information related to the number of occurrences of hypotension during non-cardiac surgery; information related to the duration of hypotension occurring during non-cardiac surgery; and information related to the severity of hypotension occurring during non-cardiac surgery.

[0073] For example, when a subject's mean arterial pressure (MAP) falls below the hypotension threshold (H) and then rises above the hypotension threshold (P1), it can be defined as one occurrence of hypotension. For example, the hypotension threshold (H) can satisfy 60 mmHg≤H≤70 mmHg. As an example, the hypotension threshold (H) can be 65 mmHg.

[0074] In one embodiment, information regarding the number of occurrences of hypotension during non-cardiac surgery may include a total number of hypotensive events that occurred during non-cardiac surgery.

[0075] For example, a 'hypotension event' may be defined as an event in which a subject's MAP remains below the hypotension threshold (H) for a predetermined period of time (e.g., 1 minute, 30 seconds, 10 seconds, 5 seconds, 3 seconds, 1 second, etc.). As an example, a 'hypotension event' may be defined as an event in which a subject's MAP remains below 65 mmHg for more than 1 minute.

[0076] In one embodiment, information regarding the duration of hypotension occurring during non-cardiac surgery may include at least one of: total duration of hypotensive events occurring during non-cardiac surgery; and average duration of each hypotensive event occurring during non-cardiac surgery. For example, the average duration of hypotensive events may be calculated as an arithmetic mean by dividing the total duration of hypotensive events by the total number of hypotensive events.

[0077] In one embodiment, the severity of hypotension occurring during non-cardiac surgery may include (a) the mean of MAP over the total duration of the hypotensive event during the non-cardiac surgery (e.g., the mean of a function (y=f(x)) where MAP is defined as y and time as x, over the duration of the hypotensive event).

[0078] In one embodiment, the severity of hypotension occurring during non-cardiac surgery may include (b) the area between the hypotension threshold line (H) and the MAP curve, among the area under the hypotension threshold line, in the MAP (y-axis)-time (x-axis) graph obtained during the non-cardiac surgery.

[0079] In one embodiment, the severity of hypotension occurring during non-cardiac surgery may include (c) the time-weighted average MAP during non-cardiac surgery (TWA-MAP).

[0080] In one embodiment, the severity of hypotension occurring during non-cardiac surgery may include (d) the total number of events with MAP under a 'severe hypotensive threshold (SH)' during non-cardiac surgery. For example, SH may satisfy 40 mmHg≤SH<60 mmHg, preferably 42 mmHg≤SH≤58 mmHg, and more preferably 45 mmHg≤H≤55 mmHg. As an example, SH may be 50 mmHg.

[0081] The severity of hypotension occurring during non-cardiac surgery according to one embodiment of the present invention may include at least one, at least two, or at least three of the above-mentioned (a), (b), (c), and (d).

[0082] In some embodiments, information regarding the subject's heart rate during non-cardiac surgery may include information regarding tachycardia during non-cardiac surgery. For example, information regarding tachycardia during non-cardiac surgery may include at least one of: whether tachycardia occurred during non-cardiac surgery; information regarding the number of occurrences of tachycardia during non-cardiac surgery; and information regarding the duration of tachycardia occurring during non-cardiac surgery.

[0083] In one embodiment, the information related to tachycardia during non-cardiac surgery may include at least one, at least two, or at least three of: whether tachycardia occurred during non-cardiac surgery; the total number of tachycardia events that occurred during non-cardiac surgery; the total duration of tachycardia events that occurred during non-cardiac surgery; and the average duration of tachycardia events that occurred during non-cardiac surgery.

[0084] The correlation between the aforementioned non-cardiac surgical information and MINS is briefly explained. i) As the surgical duration of the subject increases, the anesthesia duration also increases, which may increase the likelihood of sympathetic hyperactivity, hypotension, tachycardia, and hypothermia. In this case, the likelihood of developing cardiac diseases in the subject may increase. Therefore, surgical duration may be associated with MINS. ii) Major bleeding may be an independent risk factor for myocardial infarction, etc. Therefore, since the amount of blood loss during surgery may be related to the amount of blood transfusion, the presence and amount of intraoperative transfusion may be associated with MINS. iii) Even brief episodes of hypotension during surgery are thought to affect myocardial injury, acute kidney injury, and mortality. In particular, the severity and duration of hypotension during non-cardiac surgery are thought to be critical factors in myocardial injury. Therefore, hypotension-related information may be associated with MINS. iv) It is believed that tachycardia can affect myocardial damage by increasing myocardial oxygen demand and decreasing diastolic coronary perfusion time. Accordingly, information related to tachycardia may be linked to MINS.

[0085] As can be seen from the examples and comparative examples described below, when predicting the MINS risk by using both information before and during a non-cardiac surgery of a subject according to exemplary embodiments of the present invention, the prediction accuracy of the MINS risk can be further improved. In particular, when predicting the MINS risk by using information related to hypotension during a non-cardiac surgery of a subject, the prediction accuracy of the MINS risk can be further improved.

[0086] The server device (102) can determine the risk of myocardial damage (MINS) after non-cardiac surgery of the subject based on the information before and during the non-cardiac surgery of the subject (S402).

[0087] For example, MINS risk can be expressed as a score, a probability, a grade, or a combination of these (e.g., a grade and the probability of that grade).

[0088] For example, the MINS risk score can be used as data for determining medical measures for a subject who has undergone non-cardiac surgery. For example, the server device (102) can transmit the MINS risk score to a user device (101), a third-party device, etc. For example, a hospital, a testing institution, etc. can determine medical measures for a subject (e.g., additional tests; cardiac troponin level tests, etc.) based on the MINS risk score.

[0089] For example, the MINS risk score may include grades A (positive) and B (negative). The MINS risk score may also include the probability of the corresponding grade. Specifically, the MINS risk score is determined by the predicted grade with the highest probability, and the probability of that grade can be displayed alongside the grade (e.g., grade A and 0.85). Hospitals, testing institutions, and other institutions can take follow-up medical measures for subjects with a MINS risk score of grade A.

[0090] In one embodiment, the server device (102) may preprocess the acquired information about the subject before and during non-cardiac surgery. For example, the preprocessing may include vectorization, standardization, normalization, etc.

[0091] In one embodiment, in step S402, an artificial intelligence model trained to predict the MINS risk of a subject may be used based on information about the subject before and during non-cardiac surgery.

[0092] That is, at step S402, an AI model trained using a learning dataset with information before and during non-cardiac surgery as input data and MINS risk as output data (i.e., label) can be utilized. For example, the labeled MINS risk may be determined by an expert or other person based on known or predetermined criteria.

[0093] In some embodiments, the AI ​​model may use undersampling or oversampling (e.g., Synthetic Minority Over-Sampling Technique; SMOTE, etc.) during training to address imbalances in the training data.

[0094] In some embodiments, the AI ​​model may include a machine learning or artificial neural network-based predictive model. For example, the artificial neural network may include an input layer, an output layer, and one or more hidden layers (a deep neural network, DNN).

[0095] In some embodiments, the artificial intelligence model may include a classification model based on machine learning or an artificial neural network. For example, the classification model may be a binary classification model or a multi-class classification model. For example, the classification model may include at least one of a decision tree classification model, a random forest (RF) classification model, a logistic regression (LR) classification model, a multilayer perceptron (MLP) classification model, a k-nearest neighbor classification model, a support vector machine (SVM) classification model, an ensemble model including XGBOOST, AdaBOOST, LightGBM, and combinations thereof.

[0096] In some embodiments, the AI ​​model may include at least one of a logistic regression (LR) classification model and an artificial neural network-based classification model (e.g., a multilayer perceptron classification model including an input layer, one or more hidden layers, and an output layer). In this case, as described below, the accuracy of MINS risk prediction based on the subject's preoperative and postoperative information may be further improved.

[0097] In one example, the AI ​​model may be a random forest classification model. In this case, the subject's preoperative information may not include at least one of the following: peripheral artery disease, dialysis status, or heart failure. Alternatively, the subject's preoperative information may not include at least one, two, or three of the following: peripheral artery disease, dialysis status, heart failure, atrial fibrillation, and aortic stenosis. In this case, the AI ​​model's MINS risk prediction accuracy can be maintained or improved, while the computational speed can be improved.

[0098] In one example, the AI ​​model may be XGBOOST. In this case, the subject's preoperative information may not include at least one of the following: peripheral artery disease, heart failure, and atrial fibrillation. Alternatively, the subject's preoperative information may not include at least one, two, or three of the following: peripheral artery disease, heart failure, atrial fibrillation, dialysis status, and aortic stenosis. In this case, the AI ​​model's MINS risk prediction accuracy can be maintained or improved, while improving computational speed.

[0099] In one example, the AI ​​model may be a logistic regression classification model. In this case, the subject's preoperative information may not include at least one of the following: presence of heart failure, dialysis status, or hypertension. Alternatively, the subject's preoperative information may not include at least one, two, or three of the following: presence of heart failure, dialysis status, hypertension, dyslipidemia, or peripheral arterial disease. In this case, the AI ​​model's accuracy in predicting MINS risk can be maintained or improved, while the computational speed can be improved.

[0100] In one example, the AI ​​model may be a multilayer perceptron classification model. In this case, the subject's preoperative information may not include height, diabetes, or anemia. Alternatively, the subject's preoperative information may not include at least one, two, or three of the following: height, diabetes, anemia, heart failure, or kidney disease. In this case, the AI ​​model's MINS risk prediction accuracy can be maintained or improved, while also improving computational speed.

[0101] FIG. 5 illustrates a flowchart of operations for predicting myocardial injury (MINS) after non-cardiac surgery according to another embodiment of the present invention. FIG. 5 may be operations performed by a server device (102). However, as described above, the MINS prediction service may be provided by a program or application installed on a user device (101) without a server device (102), in which case the server device (102) described below may refer to the user device (101). Any content overlapping with the content related to FIG. 4 described above may be omitted.

[0102] Referring to FIG. 5, the server device (102) can obtain information about the subject before and during non-cardiac surgery (S501).

[0103] The server device (102) can determine the risk of myocardial injury (MINS) after non-cardiac surgery of a subject based on information about the subject before and during the non-cardiac surgery (S502). For S501 and S502, the contents of S401 and S402 described above can be applied substantially identically, and thus a detailed description thereof will be omitted.

[0104] The server device (102) can determine whether the MINS risk level is above a predetermined value (S503). For example, assuming that the MINS risk level includes grades A and B, and that the MINS risk level decreases in alphabetical order (A→B), the server device (102) can determine whether the MINS risk level is above a predetermined level (e.g., grade A or higher).

[0105] For example, if the MINS risk is higher than a predetermined value (for example, if the MINS risk is level A or higher), the server device (102) can transmit the MINS risk to the user device (101), a third party device, etc. For example, a hospital, a testing institution, etc. can take follow-up medical measures for a subject with a MINS risk level higher than a predetermined value. For example, a hospital, a testing institution, etc. can recommend additional tests (for example, cardiac troponin level tests, etc.) to a subject with a MINS risk level of level A.

[0106] If the MINS risk is below a predetermined value (for example, if the MINS risk is level B or lower), the server device (102) can obtain the subject's post-non-cardiac surgery information (S504). For example, the subject's post-non-cardiac surgery information may be information obtained from the subject within one week, five days, three days, or 48 hours after the non-cardiac surgery. The subject's post-non-cardiac surgery information may exclude the results of tests known in the art to be tests that can directly confirm MINS (for example, cardiac troponin level tests, etc.).

[0107] For example, information about a subject following non-cardiac surgery may include information about the subject's post-surgical questionnaire. For example, the post-surgical questionnaire may include the subject's response data to multiple questionnaires. For example, the multiple questionnaires may include preset questionnaires and corresponding multiple-choice response items. For example, the questionnaire may include questions related to MINS symptoms that the subject may experience.

[0108] The server device (102) can additionally determine a second MINS risk level based on the MINS risk level (hereinafter, "first MINS risk level") of the subject determined in S502 and information after non-cardiac surgery (S505). The server device (102) can transmit the second MINS risk level to the user device (101), a third party device, etc. Hospitals, testing institutions, etc. can determine follow-up medical measures for the subject based on the second MINS risk level.

[0109] For example, the server device (102) may utilize an artificial intelligence model trained to predict a subject's first MINS risk and second MINS risk based on information about non-cardiac surgery. The artificial intelligence model may be a classification model. Since the aforementioned descriptions are substantially identical to the classification model, a detailed description thereof will be omitted.

[0110] For example, a second MINS risk level may be additionally determined for a subject whose first MINS risk level is Class B (negative), as described above. The second MINS risk level may include Class BA (positive), Class BB (negative), etc. Hospitals, testing institutions, etc. may take follow-up medical measures for a subject whose second MINS risk level is Class BA.

[0111] According to one embodiment of the present invention, for a subject whose first MINS risk score is predicted as positive, follow-up medical measures related to MINS can be taken based on the first MINS risk score, enabling prompt care for the subject. Furthermore, for a subject whose first MINS risk score is predicted as negative, a second MINS risk score can be additionally determined, thereby enhancing care for the subject.

[0112] Exemplary embodiments and comparative examples of the present invention are described. However, the embodiments described below are merely exemplary embodiments of the present invention, and the present invention is not limited to the embodiments described below.

[0113] Examples and Comparative Examples

[0114] To develop and evaluate the MINS risk prediction model (binary classification model), a training data set and a validation data set consisting of input data and output data (i.e., labels) as described in Table 1 below were used. The base model of the MINS risk prediction model is also described in Table 1 below.

[0115] Standardization was performed using StandardScaler to convert the mean of each feature of the data to 0 and the standard deviation to 1. When training the MINS risk prediction model, SMOTE was used to oversample the minority class (i.e., MINS positive data) to resolve the data imbalance.

[0116] The MINS risk prediction model was evaluated based on a confusion matrix, including accuracy, precision, recall, and F1-score. Since the formulas for accuracy and other factors are already well-known, a detailed description will be omitted. The evaluation results are presented in Table 2.

[0117] Distinction Input data Output data (result data) Model Example 1 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-2, A-2-3, A-2-4, A-2-6, A-2-7, A-2-11, A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2 [Information during non-cardiac surgery] B-1, B-2, B-3-1, B-3-2, B-3-3, B-3-4, B-3-5, B-3-6, B-3-7, B-4CXGBOOST Example 2 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-2, A-2-3, A-2-4, A-2-6, A-2-7, A-2-11, A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2 [Information during non-cardiac surgery] B-1, B-2, B-3-1, B-3-2, B-3-3, B-3-4, B-3-5, B-3-6, B-3-7, B-4 CRF Example 3 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-4, A-2-6, A-2-7, A-2-9, A-2-10, A-2-11, A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2 [Information during non-cardiac surgery] B-1, B-2, B-3-1, B-3-2, B-3-3, B-3-4, B-3-5, B-3-6, B-3-7, B-4CLR Example 4 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-4, A-2-2, A-2-3, A-2-4, A-2-5, A-2-6, A-2-9, A-2-10, A-2-11, A-2-12, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2 [Information during non-cardiac surgery] B-1, B-2, B-3-1, B-3-2, B-3-3, B-3-4, B-3-5, B-3-6, B-3-7, B-4CMLP Comparative Example 1 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-2, A-2-3, A-2-4, A-2-6, A-2-7, A-2-11,A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2CXGBOOSTComparative Example 2 [Information before non-cardiac surgery]A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-2, A-2-3, A-2-4, A-2-6, A-2-7, A-2-11, A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2CRFComparative Example 3 [Information before non-cardiac surgery]A-1-1, A-1-2, A-1-3, A-1-4, A-2-1, A-2-4, A-2-6, A-2-7, A-2-9, A-2-10, A-2-11, A-2-13, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2CLRComparative Example 4 [Information before non-cardiac surgery] A-1-1, A-1-2, A-1-4, A-2-2, A-2-3, A-2-4, A-2-5, A-2-6, A-2-9, A-2-10, A-2-11, A-2-12, A-2-14, A-2-15, A-2-16, A-3-1, A-3-2CMLP,

[0118] Input data (input variables)

[0119] A. Information before non-cardiac surgery

[0120] A-1. Basic Information

[0121] A-1-1: Age

[0122] A-1-2: Gender

[0123] A-1-3: Key

[0124] A-1-4: Weight

[0125] A-2. Diagnostic Information Before Noncardiac Surgery

[0126] A-2-1: Diabetes

[0127] A-2-2: Presence of high blood pressure

[0128] A-2-3: Presence of dyslipidemia

[0129] A-2-4: Presence of coronary artery disease

[0130] A-2-5: Presence of peripheral arterial disease

[0131] A-2-6: Presence of cerebrovascular disease

[0132] A-2-7: Presence of kidney disease

[0133] A-2-8: Presence of heart failure

[0134] A-2-9: Presence of atrial fibrillation

[0135] A-2-10: Presence of aortic stenosis

[0136] A-2-11: Presence of chronic obstructive pulmonary disease

[0137] A-2-12: Whether the patient is on dialysis

[0138] A-2-13: Presence of anemia

[0139] A-2-14: Blood hemoglobin level

[0140] A-2-15: Blood creatinine level

[0141] A-2-16: Renal glomerular filtration rate

[0142] A-3. Information on non-cardiac surgery

[0143] A-3-1: Whether the above non-cardiac surgery is considered an emergency surgery

[0144] A-3-2: Whether the above non-cardiac surgery includes vascular surgery

[0145] B. Information during non-cardiac surgery

[0146] B-1. Surgery time

[0147] B-2. Blood transfusion during non-cardiac surgery

[0148] B-3. ​​Information on hypotension during non-cardiac surgery (a "hypotension event" is defined as a mean arterial pressure (MAP) below 65 mmHg that persists for more than 1 minute)

[0149] B-3-1: Total number of hypotensive events during noncardiac surgery

[0150] B-3-2: Total duration of hypotensive events occurring during noncardiac surgery

[0151] B-3-3: Mean duration of hypotensive events occurring during noncardiac surgery

[0152] B-3-4: Mean MAP value during hypotensive events during noncardiac surgery

[0153] B-3-5: In the MAP (y-axis, mmHg)-time (x-axis, minute) graph obtained during non-cardiac surgery, the area between the MAP=65 mmHg line and the MAP curve among the areas below the MAP=65 mmHg line

[0154] B-3-6: TWA-MAP during noncardiac surgery

[0155] B-3-7: Total number of events with MAP less than 50 mmHg during noncardiac surgery

[0156] B-4. Total duration of tachycardia occurring during noncardiac surgery

[0157] Output data

[0158] C. MINS positive or negative

[0159] ClassificationAccuracyPrecisionRecallF1-scoreExample 10.850.330.130.22Example 20.85---Example 30.890.540.880.67Example 40.890.570.50.53Comparative Example 10.850.40.250.31Comparative Example 20.87---Comparative Example 30.790.330.630.43Comparative Example 40.850.330.130.18

[0160] References to "one embodiment" of the principles of the present invention and various variations of this expression in this specification mean that a particular feature, structure, characteristic, etc., is included in at least one embodiment of the principles of the present invention in connection with that embodiment. Accordingly, the expression "in one embodiment" and any other variations disclosed throughout this specification are not necessarily all referring to the same embodiment.

[0161] The methods according to the various embodiments of the present invention described above may be implemented as a computer program or mobile application to be executed by combining a computer as hardware and stored on a medium. Alternatively, the steps of the methods or algorithms described in connection with the embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in a random access memory (RAM), a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a hard disk, a removable disk, a CD-ROM, or any other form of computer-readable recording medium well known in the art to which the present invention pertains. In addition, the algorithm may be produced in the form of an installation file and provided in the form of an online download, and for this purpose, may be stored on a server accessible through an online software market.

[0162] All embodiments and conditional examples disclosed in this specification are intended to help those skilled in the art understand the principles and concepts of the present invention. Those skilled in the art will appreciate that the present invention can be implemented in modified forms without departing from the essential characteristics thereof. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is set forth in the claims, not the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

Claims

1. A method for predicting myocardial injury (MINS) after non-cardiac surgery, the method comprising: A step of obtaining information about the subject before and during non-cardiac surgery; and A method for predicting MINS, comprising: a step of determining the MINS risk of the subject based on information before the non-cardiac surgery and information during the non-cardiac surgery of the subject.

2. In claim 1, A MINS prediction method, wherein the information of the subject before the non-cardiac surgery includes at least one of the following: basic information of the subject; diagnostic information of the subject before the non-cardiac surgery; and information related to the non-cardiac surgery.

3. In claim 2, The basic information of the above subject includes at least one of age, gender, height, and weight. The above-mentioned diagnosis information of the subject prior to the above-mentioned non-cardiac surgery includes at least one of the following: presence of diabetes, presence of hypertension, presence of dyslipidemia, presence of coronary artery disease, presence of peripheral artery disease, presence of cerebrovascular disease, presence of kidney disease, presence of heart failure, presence of atrial fibrillation, presence of aortic valve stenosis, presence of chronic obstructive pulmonary disease, presence of dialysis patient, presence of anemia, blood hemoglobin level, blood creatinine level, and renal glomerular filtration rate. A MINS prediction method, wherein the non-cardiac surgery-related information includes at least one of the type of the non-cardiac surgery, whether the non-cardiac surgery is an emergency surgery, and whether the non-cardiac surgery includes a vascular surgery.

4. In claim 1, A MINS prediction method, wherein the information during the non-cardiac surgery includes at least one of: the operation time of the non-cardiac surgery; information related to blood transfusion during the non-cardiac surgery; information related to hypotension during the non-cardiac surgery; and information related to heart rate during the non-cardiac surgery.

5. In claim 4, The information related to blood transfusion during the above non-cardiac surgery includes at least one of whether blood transfusion is performed during the above non-cardiac surgery and the amount of blood transfusion during the above non-cardiac surgery. A MINS prediction method, wherein the information related to heart rate during the above non-cardiac surgery includes information related to tachycardia during the above non-cardiac surgery.

6. In claim 4, A MINS prediction method, wherein the information related to hypotension during the non-cardiac surgery comprises at least one of: information related to the number of occurrences of hypotension during the non-cardiac surgery; information related to the duration of hypotension that occurred during the non-cardiac surgery; and information related to the severity of hypotension that occurred during the non-cardiac surgery.

7. In claim 6, Information on the occurrence of hypotension during the above non-cardiac surgery includes the total number of hypotension events that occurred during the above non-cardiac surgery, A MINS prediction method, wherein information related to the duration of hypotension that occurred during the non-cardiac surgery includes at least one of: the total duration of hypotension that occurred during the non-cardiac surgery; and the average duration of hypotension that occurred during the non-cardiac surgery.

8. In claim 6, The severity of hypotension occurring during non-cardiac surgery includes at least one of: an average of mean arterial pressure (MAP) during the total occurrence period of hypotension events during said non-cardiac surgery; an area of ​​an area between a hypotension reference value line and a MAP curve in a MAP (y-axis)-time (x-axis) graph acquired during said non-cardiac surgery that exists under a hypotension reference value line; a time-weighted average of MAP during said non-cardiac surgery; and a total number of occurrences of events in which MAP is below a severe hypotension threshold value during said non-cardiac surgery. A MINS prediction method, wherein the above severe hypotension threshold value is a value satisfying 40 mmHg or more and less than 60 mmHg.

9. In claim 1, A method for predicting MINS, wherein the step of determining the MINS risk comprises using an artificial intelligence model learned to predict the MINS risk of the subject based on information before the non-cardiac surgery of the subject and information during the non-cardiac surgery of the subject.

10. In claim 9, The above artificial intelligence model is a MINS prediction method including a classification model based on machine learning or an artificial neural network.

11. In claim 10, A MINS prediction method, wherein the machine learning or artificial neural network-based classification model includes at least one of a logistic regression classification model and a multilayer perceptron classification model.

12. In claim 1, When the MINS risk of the above subject is defined as the first MINS risk, if the first MINS risk of the above subject is less than a predetermined value, A step of obtaining information after the non-cardiac surgery of the subject; and A method for predicting MINS, further comprising the step of additionally determining a second MINS risk of the subject based on the first MINS risk of the subject and the post-cardiac surgery information.

13. In claim 12, The above post-cardiac surgery information includes post-cardiac surgery questionnaire information for the subject, A MINS prediction method, wherein the above questionnaire information includes the subject's response data to multiple questionnaires.

14. In a device for predicting myocardial damage (MINS) after non-cardiac surgery, Department of Communications; memory; and comprising at least one processor connected to the communication unit and the memory; A MINS prediction device, wherein said at least one processor obtains information about the subject before non-cardiac surgery and information about the subject during non-cardiac surgery, and determines the MINS risk of the subject based on the information about the subject before non-cardiac surgery and information about the subject during non-cardiac surgery.

15. An application program stored on a recording medium that executes the method according to claim 1 when operated by at least one processor.

Images