WO2023096265A1

WO2023096265A1 - Device and method for deriving severity prediction score of covid-19 patient using biological signal

Info

Publication number: WO2023096265A1
Application number: PCT/KR2022/018183
Authority: WO
Inventors: 지의규; 배예슬; 이승보
Original assignee: 서울대학교병원
Priority date: 2021-11-26
Filing date: 2022-11-17
Publication date: 2023-06-01

Abstract

The present invention relates to a device and method for deriving a severity prediction score of a COVID-19 patient using a biological signal. The method for deriving a severity prediction score of a COVID-19 patient using the device for deriving a severity prediction score according to the present invention comprises the steps of: collecting personal information and a biological signal, sensed by a wearable biological signal device, of a mild COVID-19 patient; calculating a COVID-19 severity score of the patient using the personal information and biological signal of the patient; and predicting the patient's severity of COVID-19 by using the COVID-19 severity score of the patient. The personal information of the patient includes age and gender of the patient and whether or not the COVID-19 is a recurring case. The biological signal includes at least one from among a pulse rate, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and whether or not the patient has hypertension.

Description

Severity prediction scoring device and method for COVID-19 patients using biosignals

The present invention relates to an apparatus for predicting severity of COVID-19 patients using biosignals and a method thereof, and more particularly, to an apparatus for predicting severity of COVID-19 for early diagnosis of severe severity in patients with mild symptoms.

COVID-19 is a disease caused by infection with severe acute respiratory syndrome coronavirus-2, and after an incubation period of less than two weeks, it mainly causes symptoms such as cough, fever, muscle pain, headache, and shortness of breath.

As the epidemic of COVID-19 continues, it poses a serious threat to public health, with 5.5 million deaths out of 272 million reported cases of infection as of December 2021. Although most patients with COVID-19 have mild symptoms and a good prognosis, some patients rapidly deteriorate to severe conditions such as septic shock, multiple organ failure, and acute respiratory failure. Therefore, medical staff have limitations in monitoring patients in real time, and it is difficult to establish an effective treatment plan.

In addition, various studies have confirmed that Omicron, defined as a strain of COVID-19 by the World Health Organization (WHO), is less lethal but more viscous than existing strains of COVID-19. In the context of an epidemic of COVID-19 and an infectious disease caused by the spread of Omicron, identification of the status of patients with mild COVID-19 is becoming more important.

However, the conventional technology for predicting the severity of COVID-19 patients has a problem in that it predicts the severity of symptoms from severe to severe, and does not predict the severity from mild to severe.

Therefore, there is a demand for a technology for predicting that mild COVID-19 patients will become severe so that patients with mild COVID-19 can be diagnosed early and transferred to a hospital for treatment during isolation.

The background technology of the present invention is disclosed in Republic of Korea Patent Registration No. 10-2239163 (Announced on April 12, 2021).

The present invention is a COVID-19 patient using biosignals that can predict severity using personal information and biosignals of mild COVID-19 patients so that patients with mild COVID-19 can be diagnosed early and transferred to a hospital during isolation for treatment. An object is to provide a severity predictive scoring device and method.

According to an embodiment of the present invention for achieving this technical problem, in the severity prediction scoring method of a COVID-19 patient using a severity prediction scoring device, personal information of a patient with mild COVID-19 and biometrics sensed through a wearable biosignal device Collecting signals, calculating the patient's COVID-19 severity score using the patient's personal information and vital signs, and determining the patient's COVID-19 severity using the patient's COVID-19 severity score. Predicting, wherein the patient's personal information includes the patient's age, gender, and whether or not COVID-19 has recurred, and the biosignals include pulse rate, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of hypertension includes at least one of

The step of calculating the patient's COVID-19 severity score using the patient's personal information and biosignals is the step of assigning a score for each item according to a predetermined criterion for each item of the patient's personal information and biosignals, the above items A step of summing the star scores may be included.

The predetermined criterion for each item is a threshold value determined by logistic regression analysis of the personal information and biosignals of COVID-19 patients.

In the step of assigning a score for each item, assigning 0 points if the patient's age is less than 39 years old and 2 points if the patient's age is 39 years or older, 0 points if the patient's pulse rate is less than 86 beats per minute, and 0 points if the patient's pulse rate is 86 beats per minute or more. Assigning 2 points, assigning 0 points if the oxygen saturation of the patient is greater than 98% and 2 points if the patient's oxygen saturation is less than 98%, 0 points if the patient's systolic blood pressure is less than 118 mmHg, and 1 point if the patient's systolic blood pressure is greater than or equal to 118 mmHg assigning 0 points when the patient's diastolic blood pressure is less than 77 mmHg and 1 point when the patient's body temperature is greater than or equal to 77 mmHg, assigning 0 points when the patient's body temperature is less than 37 degrees, and assigning 2 points when the patient's body temperature is greater than or equal to 37 degrees, and It may include assigning 0 points if the patient does not have hypertension and 1 point if they do.

In the step of predicting the patient's COVID-19 severity, if the COVID-19 severity score is 5 or more, the patient's condition can be predicted as severe, and if the score is less than 5, it can be predicted as mild.

Severity predictive scoring device for COVID-19 patient using severity predictive scoring device, input unit for collecting personal information of a patient with mild COVID-19 and biosignal sensed through a wearable biosignal device, personal information and biosignal of the patient A calculation unit that calculates a COVID-19 severity score of the patient by using, and a prediction unit that predicts the severity of COVID-19 of the patient using the COVID-19 severity score of the patient, wherein the patient's personal information, The patient's age, gender, and whether or not COVID-19 has recurred, and the biosignals include at least one of pulse rate, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of hypertension.

As described above, according to the present invention, the present invention is a biosignal that can predict the severity of COVID-19 patients using personal information and biosignals of mild COVID-19 patients so that they can be early diagnosed and transferred to a hospital during isolation for treatment. It is possible to provide an apparatus and method for predicting severity of COVID-19 patients using

1 is a configuration diagram illustrating an apparatus for predicting severity of COVID-19 patients according to an embodiment of the present invention.

Figure 2 is a flow chart for explaining a COVID-19 patient severity prediction scoring method according to an embodiment of the present invention.

3 is a diagram for explaining a screen of a digital calendar device displaying habit achievement information by date according to an embodiment of the present invention.

4a and 4b are diagrams showing the results of comparing the COVID-19 severity prediction method according to an embodiment of the present invention with other conventional prediction methods.

Then, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that it may further include other components without excluding other components unless otherwise stated.

Then, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention.

Hereinafter, a severity prediction scoring system for COVID-19 patients using biosignals according to an embodiment of the present invention will be described with reference to FIG. 1 .

1 is a diagram for explaining an apparatus for predicting severity of COVID-19 patients according to an embodiment of the present invention.

As shown in FIG. 1, the COVID-19 patient severity prediction scoring device 100 according to the present invention includes an input unit 110, a calculation unit 120 and a prediction unit 130.

First, the input unit 110 receives personal information of a patient with mild COVID-19 from the user. At this time, the patient's personal information includes age, gender, and whether or not COVID-19 has recurred.

Also, the input unit 110 collects biosignals sensed through the wearable biosignal device. In this case, the wearable bio-signal device detects data including at least one of pulse rate per minute, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of high blood pressure, and transmits the data to the input unit 110 .

The calculation unit 120 calculates the patient's COVID-19 severity score using the patient's personal information and biosignals.

The calculation unit 120 assigns 0 points, 1 point, or 2 points according to the patient's age, pulse rate per minute, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of hypertension.

And the calculation unit 120 calculates the COVID-19 severity score by summing the scores given according to the criterion for each item.

At this time, the sum of the severity scores is the minimum score of 0 and the maximum score is 11 points.

The prediction unit 130 predicts the severity of COVID-19 mild patients through the severity score calculated by the calculation unit 120.

The prediction unit 130 predicts severe COVID-19 patients when the sum of severity scores is 5 or more, and predicts mild COVID-19 when the total score is less than 5 points.

Next, a COVID-19 patient severity prediction scoring method using the COVID-19 patient severity prediction scoring device according to an embodiment of the present invention will be described with reference to FIGS. 2 to 4B.

Figure 2 is a flow chart for explaining a COVID-19 patient severity prediction scoring method using the COVID-19 patient severity prediction scoring device according to an embodiment of the present invention, Figure 3 is a daily habit achievement according to an embodiment of the present invention A diagram for explaining a screen of a digital calendar device displaying information, and FIGS. 4a and 4b are diagrams for explaining comparison results with conventional methods for predicting the severity of COVID-19.

First, as shown in FIG. 2, the input unit 110 receives personal information of a patient with mild COVID-19 and a biosignal from the user through a wearable biosignal device (S210).

At this time, the items input to the input unit 110 are results developed and verified using data of patients with mild symptoms of COVID-19. In particular, from March 2020 to January 2021, personal information and biosignal data of patients with mild to severe COVID-19 were collected. And it was selected as valid data to confirm the validity of the data and predict the severity of patients with mild COVID-19 through the average value, standard deviation or percentage of each item.

Next, the calculation unit 120 assigns a severity score to a patient with mild COVID-19 according to the criteria for each item through the input data (S220).

As shown in FIG. 3 , the calculation unit 120 assigns 0 points when the age of the patient is less than 39 years old and 2 points when the age of the patient is 39 years or older. A score of 0 is given if the patient's pulse rate is less than 86 beats per minute, and 2 points if the pulse rate is greater than or equal to 86 beats per minute. A score of 0 is given if the patient's oxygen saturation is greater than 98%, and 2 points if it is less than 98%. A score of 0 is given if the patient's systolic blood pressure is less than 118 mmHg, and 1 point if it is greater than or equal to 118 mmHg. A score of 0 is given if the patient's diastolic blood pressure is less than 77 mmHg, and 1 point if it is greater than or equal to 77 mmHg. A score of 0 is given if the patient's body temperature is less than 37 degrees, and 2 points if the temperature is greater than or equal to 37 degrees. A score of 0 was assigned if the patient did not have hypertension, and a score of 1 was assigned if the patient did not have hypertension.

At this time, the threshold value for each item is a value determined by logistic regression analysis of the data of patients judged to be mild to severe COVID-19 through a learning model. That is, for COVID patients, the age, pulse rate per minute, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and hypertension of patients judged to be mild and patients judged to be severe were subjected to supervised learning, and 0 points were obtained through logistic regression analysis. , 1 point and 2 points are set respectively.

Next, the calculation unit 120 sums the scores given for each item from the COVID-19 patient (S230).

As shown in Figure 3, the sum of COVID-19 severity scores is a minimum of 0 and a maximum of 11 points.

Next, the prediction unit 130 compares the total score of the severity of COVID-19 patients with a reference value (S240).

That is, according to an embodiment of the present invention, the summed total score of severity is compared with a reference value of 5 points.

As a result of comparison, if the total severity score is 5 points or more, the prediction unit 130 classifies the COVID-19 patient as a severe patient (S250), and if the total severity score is less than 5 points, the COVID-19 patient is classified as a mild patient. Do (S260).

In the embodiment of the present invention, the scores for each item were classified as 0 points, 1 points, and 2 points, but these designs are changed in various ways according to the type of item and classification standard, and the 5-point standard value is also the severity criterion and item score. Depending on the total, design changes are possible.

In addition, according to an embodiment of the present invention, gender and COVID-19 recurrence can be scored and added to the score items.

Hereinafter, the prediction accuracy of the method for predicting the severity of COVID-19 patients according to an embodiment of the present invention will be described in comparison with other existing prediction methods through FIGS. 4A and 4B.

COVID-19 severity prediction method (Proposed) according to an embodiment of the present invention and other existing COVID-19 prediction methods (NEWS, REMS. qSOFA) were compared.

As shown in FIGS. 4A and 4B, the COVID-19 severity prediction method (Proposed) according to an embodiment of the present invention is applied to the existing National Early Warning Score (NEWS), Rapid Emergency Medicine Score (REMS) and quick Sepsis-QSOFA (qSOFA). When compared to the related Organ Failure Assessment) method, it was confirmed that it has significantly higher performance.

AUC (Area Under the ROC Curve) means the area under the ROC Curve (Receiver Operating Characteristics) curve of the performance evaluation model. The ROC Curve is a graph that connects (0, 0) and (1, 1) with FPR (False Positive Rate) and TPR (True Positive Rate) as x-axis and y-axis, respectively. The ROC Curve represents the changes in the x-axis and y-axis when measured while continuously changing the model judgment standard.

The closer the AUC is to 1, the higher the class classification ability, 0.5 means that there is no class classification ability, and the closer to 0, the class classification is equivalent to predicting 1 as 2 and 2 as 1. .

The AUC of the COVID-19 patient severity prediction method (Proposed) according to an embodiment of the present invention was 0.868, and the AUCs of NEWS, REMS, and qSOFA were measured as 0.646, 0.612, and 0.509, respectively.

That is, it can be seen that the COVID-19 patient severity prediction scoring method according to an embodiment of the present invention has significantly higher performance than the existing methods for predicting COVID-19 severity.

As such, according to an embodiment of the present invention, the present invention predicts the severity of mild COVID-19 patients using personal information and biosignals of patients with mild COVID-19 so that they can be early diagnosed and transferred to a hospital during isolation for treatment. It is possible to provide a scoring device and method for predicting the severity of COVID-19 patients using possible biosignals.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is only exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical scope of protection of the present invention should be determined by the technical spirit of the appended claims.

Claims

In the severity predictive scoring method of COVID-19 patients using a severity predictive scoring device,

Collecting biosignals sensed through personal information and wearable biosignal devices of patients with mild symptoms of COVID-19;

Calculating the patient's COVID-19 severity score using the patient's personal information and biosignals, and

Predicting the patient's COVID-19 severity using the patient's COVID-19 severity score,

The patient's personal information,

Including the patient's age, gender, and whether or not COVID-19 relapsed,

The biosignal is

A scoring method for predicting COVID-19 patient severity, including at least one of pulse rate, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of hypertension.
According to claim 1,

The step of calculating the patient's COVID-19 severity score using the patient's personal information and biosignals,

assigning a score for each item according to a predetermined standard for each item of the patient's personal information and biosignal;

COVID-19 patient severity prediction scoring method comprising summing the scores for each item.
According to claim 2,

The predetermined criteria for each item are,

COVID-19 patient severity prediction scoring method, which is a threshold value determined by logistic regression analysis of personal information and vital signs of COVID-19 patients.
According to claim 1 or 2,

The step of assigning scores for each item is,

Giving 0 points if the patient's age is less than 39 years old and 2 points if the patient is 39 years old or older;

Giving 0 points when the patient's pulse rate is less than 86 beats per minute and 2 points when it is 86 beats per minute or more;

Giving 0 points if the oxygen saturation of the patient is greater than 98% and 2 points if it is 98% or less;

Giving 0 points when the patient's systolic blood pressure is less than 118 mmHg and 1 point when it is 118 mmHg or more;

Giving 0 points when the patient's diastolic blood pressure is less than 77 mmHg and 1 point when it is 77 mmHg or more;

Giving 0 points if the patient's body temperature is less than 37 degrees and 2 points if it is more than 37 degrees, and

A COVID-19 patient severity prediction scoring method comprising the step of assigning 0 points if the patient does not have hypertension and 1 point if they do.
According to claim 4,

The step of predicting the patient's COVID-19 severity,

A COVID-19 patient severity prediction scoring method that predicts the patient's condition as severe if the COVID-19 severity score is 5 or more, and predicts mild if the score is less than 5.
In the severity predictive scoring device for COVID-19 patients using the severity predictive scoring device,

An input unit that collects personal information of patients with mild symptoms of COVID-19 and biosignals sensed through a wearable biosignal device;

A calculation unit that calculates the patient's COVID-19 severity score using the patient's personal information and biosignals, and

A prediction unit for predicting the patient's COVID-19 severity using the patient's COVID-19 severity score;

The patient's personal information,

Including the patient's age, gender, and whether or not COVID-19 relapsed,

The biosignal is

A scoring method for predicting COVID-19 patient severity, including at least one of pulse rate, oxygen saturation, systolic blood pressure, diastolic blood pressure, body temperature, and presence or absence of hypertension.
According to claim 6,

The calculation unit,

Scores are given for each item according to predetermined criteria for each item of the patient's personal information and biosignal;

A scoring method for predicting the severity of COVID-19 patients by summing the scores for each item above.
According to claim 7,

The predetermined criteria for each item are,

COVID-19 patient severity prediction scoring method, which is a threshold value determined by logistic regression analysis of personal information and vital signs of COVID-19 patients.
According to claim 6 or 7,

The calculation unit,

A score of 0 was given if the patient was less than 39 years of age, and 2 points if the patient was 39 years or older.

If the patient's pulse rate is less than 86 beats per minute, 0 points are given, and if it is 86 beats per minute or more, 2 points are given.

If the oxygen saturation of the patient is more than 98%, 0 points are given, and if it is less than 98%, 2 points are given.

If the patient's systolic blood pressure is less than 118 mmHg, 0 point is given, and if it is 118 mmHg or more, 1 point is given,

If the diastolic blood pressure of the patient is less than 77 mmHg, 0 point is given, and if it is 77 mmHg or more, 1 point is given,

If the patient's body temperature is less than 37 degrees, 0 points are given, and if it is 37 degrees or more, 2 points are given, and

A COVID-19 patient severity predictive scoring method that assigns a score of 0 if the patient does not have hypertension and a score of 1 if present.
According to claim 9,

The prediction unit,

A COVID-19 patient severity prediction scoring method that predicts the patient's condition as severe if the COVID-19 severity score is 5 or more, and predicts mild if the score is less than 5.