WO2023282600A1

WO2023282600A1 - Apparatus and method for determining prognosis during or after surgery

Info

Publication number: WO2023282600A1
Application number: PCT/KR2022/009712
Authority: WO
Inventors: 이승미; 정영미; 이형철; 김도균; 이가람
Original assignee: 서울대학교병원; 더 트러스티스 오브 더 유니버시티 오브 펜실베니아
Priority date: 2021-07-09
Filing date: 2022-07-06
Publication date: 2023-01-12

Abstract

An apparatus for determining a prognosis during or after surgery, according to one embodiment, comprises: a data acquisition unit for acquiring oxygen saturation measured during surgery on a patient, an electrocardiogram measured during the surgery, and hematocrit measured during the surgery; and a processor for determining a prognosis during or after the surgery on the patient on the basis of the acquired oxygen saturation, the acquired electrocardiogram, and the acquired hematocrit.

Description

Device and method for determining prognosis during or after surgery

It relates to an apparatus and method for determining an intraoperative or postoperative prognosis based on a patient's intraoperative vital signs.

When complications occur after surgery, enormous medical costs are incurred, socio-economic costs increase, and patients' quality of life deteriorates.

On the other hand, as the importance of data is recently emphasized, there is an increasing demand to secure a model for determining the prognosis during or after surgery using medical big data. Research into the development of predictable technologies is ongoing.

An object of the present invention is to provide an apparatus and method for determining a prognosis during or after surgery based on vital signs of a patient during surgery.

An apparatus for determining prognosis during or after surgery according to an aspect includes: a data acquisition unit for acquiring oxygen saturation measured during surgery of a patient, electrocardiogram measured during surgery, and hematocrit measured during surgery; and a processor determining a prognosis during or after surgery of the patient based on the obtained oxygen saturation level, the obtained electrocardiogram, and the obtained hematocrit; can include

The processor extracts an ST segment from the acquired electrocardiogram, and determines a prognosis during or after surgery of the patient by using a prognostic judgment model based on the extracted ST segment, the acquired oxygen saturation level, and the acquired hematocrit. can judge

The data obtaining unit further acquires at least one of blood test data before surgery, surgery information, clinical information, intra-arterial blood pressure measured during the surgery, volumetric pulse wave measured during the surgery, and heart rate measured during the surgery. and the processor determines a prognosis during or after surgery of the patient further based on at least one of the pre-surgery blood test data, the surgery information, the clinical information, the intra-arterial blood pressure, the volumetric pulse wave, and the heart rate can do.

The preoperative blood test data includes prothrombin time, albumin level, hemoglobin level, platelet level, activated partial thromboplastin time, aspartate transaminase (AST) ) level, ALT (alanine transaminase) level, sodium (Na) level, potassium (K) level, glucose level, blood urea nitrogen (BUN) level, creatinine (Cr) level, estimated glomerular filtration rate (eGFR) and hematocrit can include

The surgery information may include the type of surgery, emergency surgery, and the type of anesthesia used during surgery.

The clinical information may include age, height, sex, and disease.

The processor extracts, as a first feature, at least one of an area under an intra-arterial blood pressure waveform for each heartbeat, a mean blood pressure value, a systolic blood pressure value, and a diastolic blood pressure value from the intra-arterial blood pressure, and a systolic peak value from the volume pulse wave , at least one of a diastolic peak value, an average peak value, a peak velocity, an area under a waveform, a first order differential waveform, and a second order differential waveform is extracted as a second feature, and the extracted first feature and the extracted second A prognosis of the patient during or after surgery may be determined further based on at least one of the characteristics, the blood test data before surgery, the surgery information, the clinical information, and the heart rate.

The intraoperative or postoperative prognosis may include the possibility of death during surgery, the possibility of death within a predetermined period after surgery, the possibility of massive blood transfusion, the possibility of hospitalization in an intensive care unit, the possibility of myocardial damage, and the possibility of acute renal injury.

A method for determining prognosis during or after surgery according to another aspect includes acquiring oxygen saturation measured during surgery of a patient, electrocardiogram measured during surgery, and hematocrit measured during surgery; and determining an intraoperative or postoperative prognosis of the patient based on the acquired oxygen saturation level, the acquired electrocardiogram, and the acquired hematocrit. can include

The determining may include extracting an ST segment from the obtained electrocardiogram; and determining an intraoperative or postoperative prognosis of the patient using a prognostic judgment model based on the extracted ST segment, the acquired oxygen saturation level, and the acquired hematocrit; can include

The acquiring may further include at least one of blood test data before surgery, surgery information, clinical information, intra-arterial blood pressure measured during the surgery, volumetric pulse wave measured during the surgery, and heart rate measured during the surgery. Obtaining and the determining may further include at least one of the pre-surgery blood test data, the surgery information, the clinical information, the intra-arterial blood pressure, the volumetric pulse wave, and the heart rate during or after surgery of the patient. prognosis can be determined.

The clinical information may include age, height, sex, and disease.

The determining may include extracting, as a first feature, at least one of an area under an intra-arterial blood pressure waveform for each heartbeat, a mean blood pressure value, a systolic blood pressure value, and a diastolic blood pressure value from the intra-arterial blood pressure; extracting at least one of a systolic peak value, a diastolic peak value, an average peak value, a peak velocity, an area under a waveform, a first order differential waveform, and a second order differential waveform from the volume pulse wave as a second feature; and the extracted first feature, the extracted second feature, the pre-surgery blood test data, the surgery information, the clinical information, and the heart rate further based on at least one of the patient's intraoperative or postoperative prognosis judging; can include

By using intraoperative vital sign data and machine learning or deep learning technology, the patient's prognosis during or after surgery can be accurately determined.

In addition, by using the vital sign data measured in real time during surgery, it is possible to determine the intraoperative or postoperative prognosis in real time during surgery, and through this, it is possible to perform intraoperative hemodynamic parameter correction and preventive measures.

1 is a diagram illustrating a prognosis determination system during or after surgery according to an exemplary embodiment.

2 is a diagram illustrating an apparatus for generating a prognostic judgment model according to an exemplary embodiment.

3 is a diagram illustrating a prognosis determination device according to an exemplary embodiment.

4 is a diagram illustrating a prognosis determination device according to an exemplary embodiment.

5 is a diagram illustrating a method for generating a prognostic judgment model according to an exemplary embodiment.

6 is a diagram illustrating a method for determining a prognosis according to an exemplary embodiment.

7 is a diagram illustrating the performance of a mass transfusion determination model according to an exemplary embodiment.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In adding reference numerals to components of each drawing, it should be noted that the same components have the same numerals as much as possible even if they are displayed on different drawings. In addition, in describing the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description will be omitted.

Meanwhile, in each step, each step may occur in a different order from the specified order unless a specific order is clearly described in context. That is, each step may be performed in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

Terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of a user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. Terms are only used to distinguish one component from another. Singular expressions include plural expressions unless the context clearly dictates otherwise, and terms such as 'include' or 'have' refer to features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. It is intended to specify that something exists, but it should be understood that it does not preclude the possibility of the existence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

In addition, the division of components in the present specification is merely a classification for each main function in charge of each component. That is, two or more components may be combined into one component, or one component may be divided into two or more for each more subdivided function. In addition, each component may additionally perform some or all of the functions of other components in addition to its main function, and some of the main functions of each component are dedicated to other components. may be performed. Each component may be implemented in hardware (eg, a processor or memory, etc.) or software, or a combination of hardware and software.

Referring to FIG. 1 , an intraoperative or postoperative prognosis determination system 10 (hereinafter, a prognosis determination system) according to an exemplary embodiment is an intraoperative or postoperative prognosis determination model generating device 100 (hereinafter, a prognosis determination model) generation device) and an intraoperative or postoperative prognosis determination device 200 (hereinafter referred to as a prognosis determination device).

The prognostic judgment model generation apparatus 100 uses at least one of vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information for a plurality of patients to generate machine information. A model (hereinafter referred to as a prognosis estimation model) capable of determining the prognosis of a patient undergoing surgery during or after surgery may be generated through running or deep learning.

Here, vital sign data during surgery is a vital signal of a patient measured during surgery and may include intra-arterial blood pressure, heart rate, oxygen saturation, plethysmograph, and electrocardiogram. Here, the volume pulse wave may be a photoplethysmograph, but is not limited thereto. Preoperative blood test data is blood test data tested before surgery, including prothrombin time, albumin level, hemoglobin level, platelet level, activated partial thromboplastin time. thromboplastin time), AST (aspartate transaminase) level, ALT (alanine transaminase) level, sodium (Na) level, potassium (K) level, glucose level, blood urea nitrogen (BUN) level, creatinine (Cr) level, estimated glomerular filtration rate (eGFR) and hematocrit; Surgical information includes the type of surgery, emergency surgery, and type of anesthesia used during surgery, and clinical information includes age, height, gender, and disease (e.g., liver disease, vascular disease, heart disease, and nervous system disease). , hypertension, diabetes, tuberculosis, asthma, thyroid disease, kidney disease, blood disease, chronic obstructive pulmonary disease, other diseases, etc.).

In addition, the intraoperative or postoperative prognosis is, for example, the possibility of death during surgery, the possibility of death within a certain period after surgery, the possibility of acute renal injury, the possibility of massive blood transfusion, the possibility of hospitalization in an intensive care unit, the possibility of myocardial damage, the possibility of infection after surgery, Possible postoperative pain, possible time to reach adequate postoperative pain, possible nausea and vomiting, possible postoperative delirium, possible postoperative complications, possible respiratory failure, possible acute anemia, possible thrombocytopenia, possible heart failure, possible coagulopathy, possible acidosis , likelihood of malnutrition, likelihood of sepsis, likelihood of an acute coronary event, likelihood of shock, and likelihood of a longer than average hospital stay after surgery for a similar procedure in a similar hospital setting.

The prognostic judgment device 200 uses the prognostic estimation model generated by the prognostic judgment model generating device 100 to provide vital sign data during surgery, hematocrit data measured during surgery, preoperative blood test data, surgical information, and clinical data of the target patient. A prognosis during or after surgery of a target patient may be determined from at least one of the pieces of information.

Hereinafter, the prognosis judgment model generating device 100 and the prognosis judgment device 200 will be described in detail with reference to FIGS. 2 and 3 .

Referring to FIG. 2 , an apparatus 100 for generating a prognosis judgment model according to an exemplary embodiment may include a data collection unit 110 and a processor 120 .

The data collection unit 110 collects vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgery information, and clinical information of a plurality of patients, and the condition during or after surgery of the patient. can be collected Here, the condition during or after surgery is, for example, death during surgery, death within a predetermined period after surgery, acute renal injury, massive blood transfusion, intensive care unit admission, myocardial damage, postoperative infection, surgery Postoperative pain, time to reach adequate postoperative pain, nausea and vomiting, postoperative delirium, postoperative complications, respiratory failure, acute anemia, thrombocytopenia, heart failure, coagulopathy, acidosis, These may include malnutrition, sepsis, acute coronary events, shock, and longer-than-average length of stay in hospital for similar procedures performed in similar hospital settings.

For example, the data collection unit 110 collects vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgery information, and clinical data for a plurality of patients from a medical database using wired/wireless communication technology. information can be collected. At this time, the wireless communication technology includes Bluetooth communication, Bluetooth Low Energy (BLE) communication, Near Field Communication (NFC), WLAN communication, Zigbee communication, Infrared Data Association (IrDA) communication, It may include Wi-Fi Direct (WFD) communication, ultra-wideband (UWB) communication, Ant+ communication, Wi-Fi communication, Radio Frequency Identification (RFID) communication, 3G communication, 4G communication, and 5G communication, but this is an example. only, and is not limited thereto.

The processor 120 may control overall operations of the apparatus 100 for generating a prognosis judgment model.

The processor 120 may extract features to be used in generating a prognostic judgment model from the collected intraoperative vital sign data. For example, if the collected vital sign data during surgery is intra-arterial blood pressure data, the processor 120 characterizes the area under the intra-arterial blood pressure waveform for each heartbeat, the systolic blood pressure value, the diastolic blood pressure value, the average blood pressure value, and the like. can be calculated or extracted. Also, if the collected intraoperative vital sign data is an electrocardiogram, the processor 120 may extract the ST segment as a feature. In addition, when the collected intraoperative vital sign data is a volumetric pulse wave, the processor 120 calculates a systolic peak value, a diastolic peak value, and an average peak value ((systolic peak value + diastolic peak value * 2 )/3), the rate of the peak identified by the interval between the waveforms, the area under the curve, the first differential waveform (velocity plethysmography), and the second differential waveform (acceleration plethysmography). can be calculated or extracted.

The processor 120 learns at least one of intraoperative vital sign data or features extracted therefrom, hematocrit data measured during surgery, preoperative blood test data, surgical information, and clinical information, and a corresponding intraoperative or postoperative state. Using the data as data, a prognostic judgment model may be generated through machine learning or deep learning. According to an exemplary embodiment, the processor 120 determines during surgery based on at least one of vital sign data during surgery or features extracted therefrom, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information. Alternatively, a machine learning or deep learning model can be trained to predict a patient's condition after surgery.

Referring to FIG. 3 , an apparatus 200 for determining prognosis according to an exemplary embodiment may include a data acquirer 210 and a processor 220 .

The data acquisition unit 210 may acquire vital sign data during surgery and hematocrit data measured during surgery of the patient.

For example, the data acquisition unit 210 may be disposed in an operating room to measure vital signs of a patient undergoing surgery, such as a blood pressure measuring device, a heart rate measuring device, an oxygen saturation measuring device, a pulse wave measuring device, and an electrocardiogram measuring device. It includes devices and the like, and it is possible to obtain vital sign data of a patient during surgery using these devices. In addition, the data acquisition unit 210 includes a hematocrit measurement device capable of measuring hematocrit, and the hematocrit of a patient undergoing surgery may be obtained by using the hematocrit device.

In another example, the data acquisition unit 210 receives vital sign data during surgery and hematocrit measured during surgery from an external device that measures and/or stores vital signs and/or hematocrit of the patient during surgery, thereby providing a patient of intraoperative vital sign data and intraoperatively measured hematocrit can be obtained. At this time, the data acquisition unit 210 may use a wired or wireless communication technology.

The data acquisition unit 210 may acquire preoperative blood test data of the patient, surgical information, and clinical information of the patient.

For example, the data acquisition unit 210 includes a predetermined input means, and receives pre-surgery blood test data, surgical information, clinical information, etc. of a patient from a user through the predetermined input means, thereby pre-surgery blood test data. , surgery information and clinical information can be acquired.

For another example, the data acquisition unit 210 receives pre-surgery blood test data, surgical information, patient clinical information, and the like from an external device that stores pre-surgery blood test data, surgical information, and clinical information. Preoperative blood test data of the patient, surgical information, and clinical information of the patient may be obtained.

The processor 220 may control overall operations of the prognosis determination device 200 .

The processor 220 may determine a prognosis during or after surgery of the patient based on at least one of vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information of the patient. .

The processor 220 may extract features from the patient's intraoperative vital sign data. For example, if the acquired intraoperative vital sign data is intra-arterial blood pressure data, the processor 220 characterizes the area under the intra-arterial blood pressure waveform for each heartbeat, the systolic blood pressure value, the diastolic blood pressure value, and the average blood pressure value. or can be extracted. In addition, if the acquired intraoperative vital sign data is an electrocardiogram, the processor 220 may extract the ST segment as a feature. In addition, when the collected intraoperative vital sign data is a volumetric pulse wave, the processor 220 calculates a systolic peak value, a diastolic peak value, an average peak value ((systolic peak value+diastolic peak value*2)/3), and It can be calculated or extracted as features such as the velocity of the peak identified at intervals, the area under the waveform, the first differential waveform, and the second differential waveform.

The processor 220 uses the prognosis judgment model to determine whether or not during surgery is performed based on at least one of vital sign data during surgery or features extracted therefrom, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information. Prognosis can be assessed after surgery.

For example, the processor 220 may determine an intraoperative or postoperative prognosis based on oxygen saturation measured during surgery, ST segment, and hematocrit measured during surgery.

For another example, the processor 220 may calculate or extract features (e.g., below the intra-arterial blood pressure waveform for each heartbeat) from the intra-arterial blood pressure data, in addition to oxygen saturation, ST-segment, and hematocrit measured during the surgery. area, systolic blood pressure value, diastolic blood pressure value, and mean blood pressure value), features calculated or extracted from the volume pulse wave (e.g., systolic peak value, diastolic peak value, average peak value, velocity of the peak identified by the interval between waveforms, below the waveform An intraoperative or postoperative prognosis may be determined by further using at least one of area, first differential waveform, and second differential waveform), heart rate, blood test data before surgery, surgical information, and clinical information.

The processor 220 may update the prognosis determination model based on the user's feedback if there is a user's feedback on symptoms during or after the patient's surgery. For example, if the prognosis determined based on the prognosis judgment model is different from the intraoperative or postoperative symptom input as user's feedback, the processor 220 may use the patient's vital sign data during surgery or features extracted therefrom, during surgery A prognostic judgment model may be additionally trained by using at least one of measured hematocrit data, preoperative blood test data, surgical information, and clinical information, and intraoperative or postoperative symptoms input as user feedback as new learning data. Accordingly, it is possible to increase the diagnostic accuracy of the prognosis judgment model.

According to an embodiment, the function of updating the prognostic judgment model may be performed by the apparatus 100 for generating a prognostic judgment model.

4 is a diagram illustrating a prognosis determination device according to an exemplary embodiment. The prognosis determination device 300 according to the exemplary embodiment of FIG. 4 may be another embodiment of the prognosis determination device 200 of FIG. 1 .

Referring to FIG. 4 , the prognosis determination apparatus 400 according to an exemplary embodiment includes a data acquisition unit 210, a processor 220, an input unit 410, a storage unit 420, a communication unit 430, and an output unit ( 440) may be included. Here, since the data acquisition unit 210 and the processor 220 are the same as those described above with reference to FIG. 3, detailed descriptions thereof will be omitted.

The input unit 410 may receive various manipulation signals and information from the user. According to an embodiment, the input unit 410 includes a key pad, a dome switch, a touch pad, a jog wheel, a jog switch, and hardware/software buttons. etc. may be included. In particular, when a touch pad forms a mutual layer structure with a display, it may be referred to as a touch screen.

The storage unit 420 may store programs or commands for operation of the prognosis determination apparatus 400 , and may store data input or collected in the prognosis determination apparatus 400 and processed data. Also, the storage unit 420 may store a prognosis judgment model and the like.

The storage unit 420 is a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg, SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), PROM (Programmable Read Only Memory), magnetic memory, magnetic disk, optical disk At least one type of storage medium may be included. In addition, the prognosis determination device 400 may operate an external storage medium such as a web storage that performs the storage function of the storage unit 420 on the Internet.

The communication unit 430 may communicate with an external device. For example, the communication unit 430 transmits data input to the prognosis determination device 400, stored data, processed data, etc. to an external device, or receives various data for determining a prognosis during or after surgery from the external device. can

In this case, the external device may be a medical device using data input to the prognosis determination device 400, stored data, processed data, or the like, or a printing or display device for outputting results. In addition, the external device may be a digital TV, desktop computer, mobile phone, smart phone, tablet, laptop, PDA (Personal Digital Assistants), PMP (Portable Multimedia Player), navigation device, MP3 player, digital camera, wearable device, etc., but is limited thereto It doesn't work.

The communication unit 430 may communicate with an external device using wired or wireless communication technology. At this time, the wireless communication technology includes Bluetooth communication, BLE (Bluetooth Low Energy) communication, Near Field Communication (NFC), WLAN communication, Zigbee communication, Infrared Data Association (IrDA) communication, and WFD. (Wi-Fi Direct) communication, UWB (ultra-wideband) communication, Ant+ communication, WIFI communication, RFID (Radio Frequency Identification) communication, 3G communication, 4G communication, and 5G communication, etc., but this is only an example. , but is not limited thereto.

The output unit 440 may output data input to the prognosis determination device 400, stored data, and processed data. According to an embodiment, the output unit 440 may output the prognosis determination result and the like using at least one of an auditory method, a visual method, and a tactile method. To this end, the output unit 440 may include a display, a speaker, a vibrator, and the like.

The method for generating the prognostic judgment model of FIG. 5 may be performed by the apparatus 100 for generating the prognostic judgment model of FIG. 2 .

Referring to FIG. 5 , the apparatus for generating a prognostic judgment model provides vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgical information and clinical information for a plurality of patients, and during surgery of the corresponding patient. Alternatively, postoperative conditions may be collected (510).

For example, the prognostic judgment model generating device uses wired/wireless communication technology to collect vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgery information, and clinical information for a plurality of patients from a medical database using wired/wireless communication technology. etc. can be collected.

The device for generating a prognostic judgment model is a prognostic judgment model using at least one of intraoperative vital sign data, intraoperative hematocrit data, preoperative blood test data, surgical information, and clinical information, and a patient's intraoperative or postoperative condition. can be generated (520).

According to an exemplary embodiment, the apparatus for generating a prognostic judgment model may extract features to be used in generating a prognostic judgment model from collected intraoperative vital sign data. For example, if the vital sign data collected during surgery is intra-arterial blood pressure data, the prognostic judgment model generating device features the area under the intra-arterial blood pressure waveform of each heartbeat, systolic blood pressure value, diastolic blood pressure value, average blood pressure value, etc. can be extracted with In addition, when the collected intraoperative vital sign data is an electrocardiogram, the apparatus for generating a prognostic judgment model may extract the ST segment as a feature. In addition, when the collected intraoperative vital sign data is a volumetric pulse wave, the prognostic judgment model generating device is a systolic peak value, a diastolic peak value, an average peak value ((systolic peak value + diastolic peak value * 2) / 3), between waveforms It can be calculated or extracted based on the speed of the peak identified at the interval of , the area under the waveform, the first differential waveform, and the second differential waveform.

In addition, the prognostic judgment model generation device includes at least one of vital sign data during surgery or features extracted therefrom, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information, and a corresponding intraoperative or postoperative state A prognostic judgment model may be generated through machine learning or deep learning by using as learning data. For example, the device for generating a prognostic judgment model is based on at least one of vital sign data during surgery or features extracted therefrom, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information, during surgery or surgery. Then, machine learning or deep learning models can be trained to predict the patient's condition.

The prognosis determination method of FIG. 6 may be performed by the

prognosis determination apparatus

200 or 400 of FIG. 3 or 4 .

Referring to FIG. 6 , the apparatus for determining prognosis may acquire vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information of the patient (610).

For example, the prognosis determination device may include various devices disposed in an operating room to measure vital signs of a patient undergoing surgery, such as a blood pressure measuring device, a heart rate measuring device, an oxygen saturation measuring device, a pulse wave measuring device, and an electrocardiogram measuring device. It can be used to obtain vital sign data of a patient during surgery. In addition, the prognosis determination device may obtain the hematocrit of a patient undergoing surgery using a hematocrit device.

The prognosis determination device may determine the patient's prognosis during or after surgery based on at least one of vital sign data during surgery, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information of the patient ( 620).

According to an exemplary embodiment, the prognosis determination device may extract features from the patient's intraoperative vital sign data. For example, if the acquired vital sign data during surgery is intra-arterial blood pressure data, the prognosis determination device may extract features of the area under the intra-arterial blood pressure waveform for each heartbeat, the systolic blood pressure value, the diastolic blood pressure value, and the average blood pressure value. can In addition, when the obtained intraoperative vital sign data is an electrocardiogram, the prognosis determination device may extract the ST segment as a feature. In addition, when the collected intraoperative vital sign data is a volumetric pulse wave, the prognostic judgment device is a systolic peak value, a diastolic peak value, an average peak value ((systolic peak value + diastolic peak value * 2) / 3), and the interval between waveforms. The speed of the peak identified by , the area under the waveform, the first differential waveform, and the second differential waveform can be extracted as features.

The prognostic judgment device uses a prognostic judgment model, based on at least one of vital sign data during surgery or features extracted therefrom, hematocrit data measured during surgery, blood test data before surgery, surgical information, and clinical information during surgery or surgery. prognosis can be assessed.

For example, the device for determining prognosis may determine an intraoperative or postoperative prognosis based on oxygen saturation measured during surgery, ST segment, and hematocrit measured during surgery.

For another example, the prognostic device may include features extracted from intra-arterial blood pressure data (e.g., area under the intra-arterial blood pressure waveform for each heartbeat, systolic blood pressure value, diastolic blood pressure value, and mean blood pressure value), features extracted from the volume pulse wave (e.g., systolic peak value, diastolic peak value, average peak value, peak velocity determined by the interval between waveforms, area under the waveform, first derivative A prognosis during or after surgery may be determined by further using at least one of a waveform, a second differential waveform), a heart rate, pre-surgery blood test data, surgical information, and clinical information.

According to an exemplary embodiment, if there is a user's feedback about symptoms during or after surgery of the patient, the prognosis determination device may update the prognosis determination model based on the user's feedback. For example, the prognostic determination device may be used to measure intraoperative or postoperative symptoms input as user feedback and a prognosis determined based on a prognostic judgment model, the patient's intraoperative vital sign data or features extracted therefrom, and intraoperative measurement. The prognosis judgment model may be additionally trained by using at least one of the received hematocrit data, preoperative blood test data, surgical information, and clinical information, and intraoperative or postoperative symptoms input as user feedback as new learning data.

[Example]

A total of 17,986 patient data were collected from the Seoul National University Hospital database. A total of 17,986 patient data was randomly divided, and 12,535 patient data were used as a training set to generate a mass transfusion judgment model, and 5,451 patient data were used as a verification set to verify the performance of the mass transfusion judgment model.

Massive transfusion was defined as transfusion of 3 or more red blood cells over 1 hour.

Intra-arterial blood pressure waveform for each heartbeat extracted from oxygen saturation measured during surgery, heart rate measured during surgery, hematocrit measured during surgery, blood test data before surgery, surgical information and clinical information, and intra-arterial blood pressure data measured during surgery A mass transfusion judgment model was created using the lower area, systolic blood pressure value, diastolic blood pressure value, average blood pressure value, and the ST segment extracted from electrocardiogram data measured during surgery, and its performance was judged. As a result, FIG. 7 was obtained.

Referring to FIG. 7 , it was confirmed that the AUROC (area under the receiver characteristic curve) of the mass transfusion determination model generated using the above variables was about 0.974, which is a very high level.

The above-described embodiments may be implemented as computer readable codes on a computer readable recording medium. A computer-readable recording medium may include all types of recording devices storing data that can be read by a computer system. Examples of computer-readable recording media may include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical disk, and the like. In addition, the computer-readable recording medium may be distributed among computer systems connected through a network, and may be written and executed as computer-readable codes in a distributed manner.

So far, the present invention has been looked at with respect to its preferred embodiments. Those skilled in the art to which the present invention pertains will be able to understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the scope of the present invention should be construed to include various embodiments within the scope equivalent to those described in the claims without being limited to the above-described embodiments.

Claims

a data acquisition unit that acquires oxygen saturation measured during surgery of the patient, an electrocardiogram measured during the surgery, and hematocrit measured during the surgery; and

a processor determining a prognosis during or after surgery of the patient based on the acquired oxygen saturation level, the acquired electrocardiogram, and the acquired hematocrit; including,

Intraoperative or postoperative prognostic device.
According to claim 1,

the processor,

Extracting an ST segment from the obtained electrocardiogram, and determining a prognosis during or after surgery of the patient using a prognostic judgment model based on the extracted ST segment, the obtained oxygen saturation, and the obtained hematocrit;

Intraoperative or postoperative prognostic device.
According to claim 1,

The data obtaining unit further acquires at least one of blood test data before surgery, surgery information, clinical information, intra-arterial blood pressure measured during the surgery, volumetric pulse wave measured during the surgery, and heart rate measured during the surgery. do,

The processor further determines a prognosis during or after surgery of the patient based on at least one of the pre-surgery blood test data, the surgery information, the clinical information, the intra-arterial blood pressure, the volumetric pulse wave, and the heart rate,

Intraoperative or postoperative prognostic device.
According to claim 3,

The preoperative blood test data includes prothrombin time, albumin level, hemoglobin level, platelet level, activated partial thromboplastin time, aspartate transaminase (AST) ) level, ALT (alanine transaminase) level, sodium (Na) level, potassium (K) level, glucose level, blood urea nitrogen (BUN) level, creatinine (Cr) level, estimated glomerular filtration rate (eGFR) and hematocrit including,

Intraoperative or postoperative prognostic device.
According to claim 3,

The surgical information includes the type of anesthesia used during surgery, emergency surgery, and

Intraoperative or postoperative prognostic device.
According to claim 3,

The clinical information includes age, height, sex and disease,

Intraoperative or postoperative prognostic device.
According to claim 3,

the processor,

Extracting at least one of an area under an intra-arterial blood pressure waveform for each heartbeat, a mean blood pressure value, a systolic blood pressure value, and a diastolic blood pressure value from the intra-arterial blood pressure as a first feature, and a systolic peak value and a diastolic peak value from the volume pulse wave , At least one of a mean peak value, a velocity of a peak, an area under a waveform, a first order differential waveform, and a second order differential waveform is extracted as a second feature, the extracted first feature, the extracted second feature, and the operation Determining a prognosis during or after surgery of the patient further based on at least one of the whole blood test data, the surgical information, the clinical information, and the heart rate,

Intraoperative or postoperative prognostic device.
According to claim 1,

The intraoperative or postoperative prognosis includes the possibility of death during surgery, the possibility of death within a predetermined period after surgery, the possibility of massive blood transfusion, the possibility of hospitalization in an intensive care unit, the possibility of myocardial damage and the possibility of acute renal injury,

Intraoperative or postoperative prognostic device.
obtaining an oxygen saturation level measured during surgery of the patient, an electrocardiogram measured during the surgery, and a hematocrit measured during the surgery; and

determining an intraoperative or postoperative prognosis of the patient based on the acquired oxygen saturation level, the acquired electrocardiogram, and the acquired hematocrit; including,

How to determine the prognosis during or after surgery.
According to claim 9,

The step of judging is

extracting an ST segment from the obtained electrocardiogram; and

determining an intraoperative or postoperative prognosis of the patient using a prognostic judgment model based on the extracted ST segment, the obtained oxygen saturation level, and the obtained hematocrit; including,

How to determine the prognosis during or after surgery.
According to claim 9,

The acquiring may further include at least one of blood test data before surgery, surgery information, clinical information, intra-arterial blood pressure measured during the surgery, volumetric pulse wave measured during the surgery, and heart rate measured during the surgery. acquire,

The determining step further determines a prognosis during or after surgery of the patient based on at least one of the pre-surgery blood test data, the surgery information, the clinical information, the intra-arterial blood pressure, the volumetric pulse wave, and the heart rate. doing,

How to determine the prognosis during or after surgery.
According to claim 11,

The preoperative blood test data includes prothrombin time, albumin level, hemoglobin level, platelet level, activated partial thromboplastin time, aspartate transaminase (AST) ) level, ALT (alanine transaminase) level, sodium (Na) level, potassium (K) level, glucose level, blood urea nitrogen (BUN) level, creatinine (Cr) level, estimated glomerular filtration rate (eGFR) and hematocrit including,

How to determine the prognosis during or after surgery.
According to claim 11,

The surgical information includes the type of anesthesia used during surgery, emergency surgery, and

How to determine the prognosis during or after surgery.
According to claim 11,

The clinical information includes age, height, sex and disease,

How to determine the prognosis during or after surgery.
According to claim 11,

The step of judging is

extracting at least one of an area under an intra-arterial blood pressure waveform for each heartbeat, a mean blood pressure value, a systolic blood pressure value, and a diastolic blood pressure value from the intra-arterial blood pressure as a first feature;

extracting at least one of a systolic peak value, a diastolic peak value, an average peak value, a peak velocity, an area under a waveform, a first order differential waveform, and a second order differential waveform from the volume pulse wave as a second feature; and

Determining a prognosis of the patient during or after surgery based on at least one of the extracted first feature, the extracted second feature, the pre-surgery blood test data, the surgery information, the clinical information, and the heart rate doing; including,

How to determine the prognosis during or after surgery.
According to claim 9,

The intraoperative or postoperative prognosis includes the possibility of death during surgery, the possibility of death within a predetermined period after surgery, the possibility of massive blood transfusion, the possibility of hospitalization in an intensive care unit, the possibility of myocardial damage and the possibility of acute renal injury,

How to determine the prognosis during or after surgery.