KR102014825B1 - Apparatus and method for predicting reduction ratio of apnea-hypopnea index before surgery to treat sleep apnea syndrome - Google Patents

Abstract

The present invention discloses an apparatus and method for predicting a rate of reduction of apnea low apnea index before surgery for the treatment of obstructive sleep apnea. According to an embodiment of the present invention, the apnea low respiratory rate reduction rate prediction device is considered in consideration of visibility of tonsils based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula. A tonsil size grading unit for determining a tonsil size grade, a data collection unit for collecting the results of polysomnography, and the determined tonsil size grade and the collected results are used to calculate a reduction rate of apnea hypopnea index It may include a calculating unit.

Description

FIELD AND METHOD FOR PREDICTING REDUCTION RATIO OF APNEA-HYPOPNEA INDEX BEFORE SURGERY TO TREAT SLEEP APNEA SYNDROME}

The present invention relates to the technical concept of predicting the surgical outcome by using a decrease rate of apnea low apnea before surgery for the treatment of obstructive sleep apnea, specifically, based on the patient's tonsil size and sleep polysomnography data before surgery The present invention relates to an apparatus and a method for predicting a surgical outcome by calculating a reduction rate of a low breathing index.

Obstructive Sleep Apnea Syndrome (OSAS) refers to a sleep disorder characterized by narrowing or obstruction of any one or several parts of the upper airway anatomically including nasal, oral cavity, pharynx, larynx and the like.

Obstructive sleep apnea can be diagnosed when the first condition and the second condition among the following conditions are satisfied at the same time or satisfy the third condition.

The first condition may be interpreted as being one of the following symptoms.

In the first case, symptoms corresponding to the first condition may include a case in which the patient complains of drowsiness, restless sleep, fatigue or insomnia.

The second case may include a case where the patient stops breathing, gasping for breath, or wakes up with a choking.

The third case may include a case where the sleeping partner or other observer says that the patient has a habitual snoring, one or both of breathing stops during sleep.

The fourth case may also include a case where the patient has advanced hypertension, emotional disorders, cognitive impairment, coronary artery disease, stroke, heart failure, atrial fibrillation or type 2 diabetes.

The second condition is the finding of polysomnography (PSG) or portable sleep test device (OSCT), where breathing events (closed or mixed apnea, low breathing, arousal related to breathing effort) are five or more times during sleep. to be.

The third condition is the finding of a sleep polysomnography or a portable sleep test device in which a respiratory event (closed or mixed apnea, low breathing, arousal associated with breathing effort) is 15 or more times during sleep.

Treatment of obstructive sleep apnea may include weight control, posture therapy, positive airway pressure (PAP) therapy, oral appliance, surgical treatment, and the like.

For example, surgical treatment may correspond to Uvulo Palato Pharyngoplasty (UPPP).

In general, surgical treatment during obstructive sleep apnea treatment is not effective for all patients and can be effectively applied only to some patients with surgically correctable structural causes.

In addition, surgical treatment has the disadvantage of taking risks, complications and the like associated with surgery.

Therefore, it is necessary to predict the reduction rate of apnea low respiratory index before surgery to prevent risks and complications associated with surgery.

Korean Registered Patent No. 10-1596662, "A device for diagnosing the therapeutic effect of obstructive sleep apnea and its diagnostic method" Korean Laid-Open Patent Publication No. 10-2015-0123938, "Method for calculating the score of a medical proposal supporting a medical decision" Korean Patent Registration No. 10-1693808, "3D-based nasal surgery simulation method and apparatus thereof" Korean Registered Patent No. 10-1611838, "Method and system for predicting high risk group for patient safety"

An object of the present invention is to provide an apparatus and method for predicting a reduction rate of apnea low apnea index before surgery for the treatment of obstructive sleep apnea.

The present invention aims to provide help to patients from an economic, social and personal perspective by predicting surgical outcomes prior to surgery and avoiding risk complications associated with the surgery.

It is an object of the present invention to grade the tonsil size of a surgical subject, and to predict the rate of apnea hypopnea reduction based on the graded tonsil size and the result of sleep polysomnography.

The present invention aims to predict surgical outcomes based on predicted apnea reduction rate.

An object of the present invention is to improve the prediction accuracy of the apnea rate reduction rate through machine learning using the postoperative results.

According to an embodiment of the present invention, the apnea low respiratory rate reduction rate prediction device is considered in consideration of visibility of tonsils based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula. A tonsil size grading unit for determining a tonsil size grade, a data collection unit for collecting the results of polysomnography, and the determined tonsil size grade and the collected results are used to calculate a reduction rate of apnea hypopnea index It may include a calculating unit.

According to an embodiment of the present invention, the tonsil size grade determining unit determines the tonsil size grade as the first grade when the visibility of the tonsils by the tonsil pillar is not secured, and surrounds the tonsil pillar. In the case where the visibility of the tonsil is secured, the tonsil size grade may be determined as the second grade.

According to one embodiment of the present invention, the tonsil size grading unit is visible tonsil below the baseline corresponding to half the position from the tonsil pole at the distance between the tonsil pole and the palatine uvula. In this case, the tonsil size grade may be determined as the third grade, and when the visibility of the tonsil is secured above the baseline, the tonsil size grade may be determined as the fourth grade.

According to an embodiment of the present invention, the tonsil size grade determining unit may determine the tonsil size grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the first level. Determine the first grade, and if the visibility of the tonsil based on the tonsil size and area of the pharynx corresponds to the second level, determine the tonsil size grade as the second grade, and If the visibility of the tonsil based on the tonsil size and area of the pharynx corresponds to the third level, the tonsil size grade is determined as the third grade, and the tonsil size and the pharynx When the visibility of the tonsil based on the area of corresponds to the fourth level, the tonsil size grade may be determined as the fourth grade.

According to an embodiment of the present invention, the data collection unit, the first data on the apnea index calculated by adding the number of apnea and low breathing number of the surgery prediction target, the minimum oxygen saturation required for sleep of the surgery prediction target Second data for), third data for snoring rate during the sleep prediction target, and fourth data for the arousal index related to the number of awakenings of the surgery prediction target during the sleep period. .

According to an embodiment of the present invention, the calculator calculates a first value by adding the first value representing the determined one-way size grade and the second value of the first data, and the fourth value is calculated from the calculated first value. A second value is calculated by subtracting a third value of the data, and the reduction rate is calculated by adding the third value and the second value calculated by subtracting the fifth value of the third data from the fourth value of the second data. Can be.

According to an embodiment of the present invention, the calculation unit may apply a weight that is differential to the first value according to the determined one-way size grade, and reflect the first value to which the weight is applied to the first value.

According to an embodiment of the present invention, the calculator may reflect the first numerical value to the first value only when the determined one-way size grade is one of a third grade and a fourth grade.

According to an embodiment of the present invention, the calculator may calculate the reduction rate by applying a first constant and a second constant to each of the first to fifth values.

According to an embodiment of the present invention, the apnea low respiratory index reduction rate prediction apparatus may further include a machine learning unit for collecting the surgical results after the operation of the operation prediction target to update the first and second constants.

According to one embodiment of the present invention, a method for predicting apnea low respiratory rate decrease rate in the tonsil size grading unit, based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula ( tonsil) considering the visibility of the tonsil size, in the data collection unit, collecting the result of the polysomnography (polysomnography) and in the calculation unit, the determined tonsil size rating and the collected result value Calculating a rate of decrease of the apnea index.

According to an embodiment of the present invention, if the apnea low respiratory index reduction rate prediction method is not secured visibility of the tonsil (tonsil) by the tonsil, determining the tonsil size grade as a first grade, the tonsil pillar If visibility of the tonsil is ensured in the vicinity of the step, determining the tonsil size rating as a second grade, half the position from the tonsil column at a distance between the tonsil pillar and the palatine uvula Determining the tonsil size grade as the third grade when the visibility of the tonsil is secured below the baseline corresponding to the and the tonsil size when the tonsil visibility is secured above the baseline. And determining the grade as the fourth grade.

According to an embodiment of the present invention, the method for predicting apnea reduction rate of apnea may include collecting first data on an apnea index calculated by adding up the number of apneas and the number of low apneas of a surgery prediction target, and requesting sleep for the surgery prediction target. Collecting second data on the minimum oxygen saturation, collecting third data on the percentage of snoring during the sleep time of the sleep polysomnography, and the number of awakenings of the predicted subject during the sleep time. And collecting fourth data for the arousal index associated with the.

According to an embodiment of the present invention, the method for predicting apnea low respiratory rate decrease rate comprises: calculating a first value by adding a first value representing the determined one-way size scale and a second value of the first data, and calculating the first value. Subtracting a third value of the fourth data from a value to calculate a second value; and subtracting the third value and the second value calculated by subtracting a fifth value of the third data from the fourth value of the second data. Summing to calculate the reduction rate.

The present invention can provide help to patients from an economic, social and personal perspective by predicting the rate of apnea hypopnea reduction prior to surgery to avoid risk complications associated with surgery.

The present invention can grade the tonsil size of a surgical subject and predict the rate of apnea hypopnea index reduction based on the graded tonsil size and the result of sleep polysomnography.

The present invention can improve the prediction accuracy of the apnea low respiratory rate reduction rate through machine learning using the postoperative results.

FIG. 1 is a diagram illustrating a data collection environment based on a sleep polymorphism test according to an embodiment of the present invention.
2 is a view for explaining the components of the apnea low respiratory rate decrease rate prediction apparatus according to an embodiment of the present invention.
3 is a diagram illustrating machine learning about the results of surgery according to an embodiment of the present invention.
4 and 7 are flowcharts illustrating a method for predicting apnea low respiration index reduction rate according to an embodiment of the present invention.
8 is a view for explaining a comparison result for the reduction rate prediction according to an embodiment of the present invention.

Hereinafter, various embodiments of the present disclosure will be described with reference to the accompanying drawings.

The examples and terms used therein are not intended to limit the techniques described in this document to specific embodiments, but should be understood to include various modifications, equivalents, and / or alternatives to the examples.

In the following description of the various embodiments, when 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 thereof will be omitted.

In addition, terms to be described below are terms defined in consideration of functions in various embodiments, and may vary according to a user's or operator's intention or custom. Therefore, the definition should be made based on the contents throughout the specification.

In connection with the description of the drawings, similar reference numerals may be used for similar components.

Singular expressions may include plural expressions unless the context clearly indicates otherwise.

In this document, expressions such as "A or B" or "at least one of A and / or B" may include all possible combinations of items listed together.

Expressions such as "first," "second," "first," or "second," etc. may modify the components, regardless of order or importance, to distinguish one component from another. Used only and do not limit the components.

When any (eg first) component is said to be "connected (functionally or communicatively)" or "connected" to another (eg second) component, the other component is said other It may be directly connected to the component or may be connected through another component (for example, the third component).

In this specification, "configured to" is modified to have the ability to "suitable," "to," "to," depending on the context, for example, hardware or software. Can be used interchangeably with "made to", "doing", or "designed to".

In some situations, the expression “device configured to” may mean that the device “can” together with other devices or components.

For example, the phrase “processor configured (or configured to) perform A, B, and C” may be implemented by executing a dedicated processor (eg, an embedded processor) to perform its operation, or one or more software programs stored in a memory device. It may mean a general purpose processor (eg, a CPU or an application processor) capable of performing the corresponding operations.

In addition, the term 'or' means inclusive or 'inclusive or' rather than 'exclusive or'.

In other words, unless stated otherwise or unclear from the context, the expression 'x uses a or b' means any one of natural inclusive permutations.

The terms '.. unit' and '.. group' used below mean a unit for processing at least one function or operation, which may be implemented by hardware or software, or a combination of hardware and software.

FIG. 1 is a diagram illustrating a data collection environment based on a sleep polymorphism test according to an embodiment of the present invention.

Specifically, FIG. 1 shows that the apnea low respiratory index decrease rate prediction device 160 has data on the tonsil size, apnea low respiratory index, minimum oxygen saturation, snoring ratio, and arousal index of the surgical prediction target 100 in the sleep polysomnography environment. Illustrates the environment in which it is collected. Here, the environment in which the data is collected includes a test environment that is performed before or after surgery to treat obstructive sleep apnea. In addition, surgery to treat obstructive sleep apnea may include Uvulo Palato Pharyngoplasty (UPPP).

Referring to FIG. 1, a thoracic motion measuring device 110, an oxygen saturation measuring device 120, a nasal airflow measuring device 130, and a brain motion measuring device 140 are attached to a surgical prediction target 100, and a thoracic motion Measuring device 110, oxygen saturation measuring device 120, nasal airflow measuring device 130, brain motion measuring device 140 collects the first data to the fourth data collected data recording device 150 And the apnea low respiratory rate decrease rate prediction device 160 collects the stored data.

According to one embodiment of the present invention, the thoracic motion measuring apparatus 110 measures the thoracic motion of the operation predicted object 100 and calculates the apnea count and the low apnea count based on the measured thoracic motion, The apnea low respiratory index is calculated by adding the number of low breaths.

For example, the thoracic motion measuring device 110 transmits the calculated apnea low respiration index to the data recording device 150 and stores the first data.

According to an embodiment of the present invention, the oxygen saturation measuring apparatus 120 measures the oxygen saturation degree of the surgical prediction target 100 and calculates the minimum oxygen saturation degree required for sleep based on the measured oxygen saturation degree.

For example, the oxygen saturation measuring apparatus 120 transmits the calculated apnea low breathing index to the data recording apparatus 150 to store the second data.

According to one embodiment of the present invention, the nasal airflow measurement apparatus 130 calculates the snoring ratio based on the nasal airflow during the sleep time of the surgical prediction target 100.

In one example, the nasal air flow measurement device 130 stores the calculated snoring ratio as the third data.

According to an embodiment of the present invention, the brain motion measuring apparatus 140 calculates the arousal index by calculating the number of awakenings of the surgical prediction target during the sleep time of the surgical prediction target 100.

Here, the number of awakenings includes a number of times of switching from deep sleep to shallow sleep during the sleep time of the surgery prediction target 100.

In addition, the awakening index is calculated based on the number of awakenings during a specific time in the sleep time.

For example, the brain movement measuring apparatus 140 transmits the calculated wakefulness index to the data recording apparatus 150 to store the fourth data.

According to an embodiment of the present invention, the data recording apparatus 150 may collect data on the tonsil size of the surgical prediction target 100.

According to an embodiment of the present invention, the apnea low respiratory index reduction rate predicting unit 160 may collect data on the tonsil size, apnea low apnea index, minimum oxygen saturation, snoring ratio, and arousal index from the data recording device 150. Can be.

For example, the apnea low respiratory index decrease rate predictor 160 is a tonsil 101, tonsil pillar 102, palate 103, pharynx (from the surgical prediction target 100 before the operation of the surgical prediction target 100) 104) and data about palate 105.

According to an embodiment of the present invention, the apnea low respiratory rate decrease rate prediction device 160 uses data about the tonsil 101, the tonsil pillar 102, the palate 103, the pharynx 104 and the palatal 105. After determining the tonsil size grade, it can be used as variable data for calculating the rate of apnea reduction.

Hereinafter, the apnea low respiratory rate reduction rate predicting apparatus 160 and its operation method will be described in detail.

2 is a view for explaining the components of the apnea low respiratory rate decrease rate prediction apparatus according to an embodiment of the present invention.

Referring to FIG. 2, according to an embodiment of the present invention, the apnea low breathing rate reduction rate predicting apparatus 200 includes a one-way size rating determiner 210, a data collector 220, and a calculator 230.

According to an embodiment of the present invention, the tonsil size grading unit 210 considers the visibility of the tonsil based on at least one of the tonsil size, the area of the pharynx, the tonsil column and the palatine uvula. To determine the tonsil size rating.

For example, the apnea low respiratory rate decrease rate prediction apparatus 200 may level the visibility of the tonsil according to the degree to which the tonsil is covered by the tonsil pillar.

In addition, the apnea low respiratory rate decrease rate prediction apparatus 200 may level the visibility of the tonsil according to the degree to which the tonsil is identified between the tonsil column and the palate.

In addition, the apnea low respiratory rate decrease rate prediction apparatus 200 may level the visibility of the tonsil according to the degree to which the tonsil covers the pharynx. That is, the apnea low respiratory rate decrease rate prediction apparatus 200 may level the visibility of the tonsil according to the degree to which the tonsil size covers the area of the pharynx.

According to an embodiment of the present invention, the tonsil size grade determiner 210 determines the tonsil size grade as the first grade when the visibility of the tonsil by the tonsil pillar is not secured. Here, the case where the visibility of the tonsil is not secured by the tonsil pillar corresponds to a case where the tonsil is buried inside the tonsil pillar.

According to one embodiment of the present invention, the tonsil size grade determiner 210 determines the tonsil size grade as the second grade when the visibility of the tonsil is secured around the tonsil pillar.

That is, the tonsil size rating determiner 210 determines the tonsil size grade as the first grade when the tonsil is covered by the tonsil pillar, and determines the tonsil size grade as the second grade when the tonsil is identified around the tonsil pillar. do.

According to one embodiment of the present invention, the tonsil size rating determiner 210 determines the tonsil size rating when the visibility of the tonsil is secured at a distance between the tonsil pole and the palatal number at a baseline corresponding to half the position of the tonsil pole. Determined as the third grade.

For example, the tonsil size grade determiner 210 determines the tonsil size grade as the fourth grade when the visibility of the tonsil is secured above the reference line.

That is, if the tonsil size rating determiner 210 divides the distance between the tonsil column and the palate into half and does not exceed half, the one-way size grade is determined as the third grade. The grade is determined as the fourth grade.

For example, the position of the baseline corresponds to half of the distance between the tonsil column and the palate in the sphere to be predicted for surgery.

According to an embodiment of the present invention, the tonsil size grade determiner 210 may determine the tonsil size grade as the first grade when the tonsil visibility based on the tonsil size and the area of the pharynx corresponds to the first level.

For example, the tonsil size grade determiner 210 may determine the tonsil size grade as the second grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the second level.

For example, when the amygdala size occupies 25% or less of the pharynx area, the amygdala size rating determiner 210 may set the visibility of the amygdala to the first level.

For example, the tonsil size rating determiner 210 may set the visibility of the tonsil secured when the tonsil size occupies 50% or less of the area of the pharynx to the second level.

That is, the tonsil size rating determiner 210 determines the tonsil size grade as the first grade when the tonsil size occupies 25% or less of the area of the pharynx, and when the tonsil size occupies 50% or less of the area of the pharynx. It can be decided by class 2.

According to an embodiment of the present invention, the tonsil size grade determiner 210 may determine the tonsil size grade as the third grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the third level.

For example, the tonsil size grade determiner 210 may determine the tonsil size grade as the fourth grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the fourth level.

For example, the tonsil size rating determiner 210 may set the visibility of the tonsil secured when the tonsil size occupies 50% or more and occupies 75% or less of the area of the pharynx to the third level.

For example, the tonsil size grade determiner 210 may set the visibility of the tonsil secured when the tonsil size occupies 75% or more of the pharynx and 100% or less to a fourth level.

That is, when the tonsil size rating unit 210 determines that the tonsil size is 50% or more of the area of the pharynx and occupies 75% or less, the tonsil size rating is determined as the third grade, and the tonsil size is 75% or more of the area of the pharynx. If occupied by the fourth grade can be determined.

According to an embodiment of the present invention, the data collection unit 220 collects a result value of polysomnography.

For example, the data collection unit 220 collects first data on the apnea low apnea index that is calculated by adding up the number of apneas and the number of low apneas of the surgery prediction target.

According to an embodiment of the present invention, the data collection unit 220 collects second data on the minimum oxygen saturation required for the sleep of the surgical prediction target.

In one example, the data collection unit 220 collects the third data about the snoring rate during the sleep time of the surgery prediction target. Here, the snoring ratio may represent the snoring time to the total sleep time of the surgery prediction target.

According to an embodiment of the present invention, the data collection unit 220 collects fourth data on the arousal index related to the number of awakenings of the target for predicting surgery in the sleep time.

Here, the arousal index may be determined based on the number of awakenings per hour in the sleep polynomial test of the surgery prediction target.

According to an embodiment of the present invention, the calculation unit 230 may calculate a reduction rate of the apnea low respiratory index by using the tonsil size grade and the result of the sleep polymorphism test.

That is, the present invention may grade the tonsil size of the surgical subject, and predict the rate of apnea hypopnea reduction based on the graded tonsil size and the result of sleep polysomnography.

For example, the calculator 230 may calculate a reduction rate of the apnea low breathing index using Equation 1 below.

[Equation 1]

According to Equation 1, AHI reduction ratio may represent apnea low respiratory index reduction rate, TS Gr may represent a tonsil size scale, AHI may represent apnea low respiratory index, and Arousal index may represent arousal index. In addition, Minimum SaO ₂ (%) may represent the minimum oxygen saturation, Snoring (%) may represent the snoring ratio.

That is, the calculation unit 230 calculates the first value by adding the first value indicating the tonsil magnitude grade and the second value indicating the apnea low breathing index using Equation 1, and represents the arousal index at the first value. The second value is calculated by subtracting the third value.

Next, the calculation unit 230 calculates the third value by subtracting the fifth value indicating the snoring ratio from the fourth value indicating the minimum oxygen saturation degree, and adding the second value and the third value to reduce the apnea low respiratory rate index. Calculate.

In one example, the calculator 230 applies the weighted differential to the first value according to the one-way size grade, and then reflects the weighted first value to the first value.

According to an embodiment of the present invention, the calculation unit 230 reflects the first value to the first value only when the one-way size grade is any one of the third grade and the fourth grade, and the first and second grades. In either case, the first numerical value is not reflected in the first value.

For example, the calculator 230 calculates an apnea reduction rate by applying a first constant and a second constant to each of the first to fifth values, which are elements for calculating the apnea reduction rate.

According to an embodiment of the present invention, the calculation unit 230 may increase the prediction accuracy of the apnea low respiratory rate reduction rate by using the first constant and the second constant.

In one example, the first constant and the second constant may correspond to constants multiplied by the first to fifth numerical values.

According to another embodiment of the present invention, the apnea low respiratory rate decrease rate prediction apparatus 200 may further include a machine learning unit 240 and a prediction unit (not shown).

In one example, the machine learning unit 240 collects the surgical results after the operation of the surgery prediction target to update the first and second constants.

For example, the machine learning unit 240 collects the surgical results after performing the surgery to perform the machine learning, the first constant and the second constant used in Equation 1 together with the variables corresponding to the first and fifth numerical values. You can update it.

That is, the present invention can improve the prediction accuracy of the rate of apnea low apnea through machine learning using the postoperative results.

In addition, according to the present invention, the more the surgical result, the more accurate prediction can be made, and the accuracy of the surgical result prediction model can be increased based on the machine learning.

According to an embodiment of the present invention, the prediction unit (not shown) may predict the surgical result based on the rate of apnea reduction.

For example, the prediction unit (not shown) may predict a preoperative surgery result by comparing the apnea low respiratory rate decrease rate and the apnea low respiratory index of the sleep multiple examination.

According to an embodiment of the present invention, the predicting unit (not shown) may determine that the predicted surgical result is successful when the apnea low respiratory index decrease rate is greater than the threshold value of the surgical success.

For example, the predictor (not shown) determines that the rate of decrease in apnea low respiratory index, which represents the result of the comparison between the apnea low respiratory index and the apnea low respiratory index calculated from the prediction result, is greater than 50%. In this case, the surgical outcome can be predicted as a success.

That is, the predicting unit (not shown) may predict the success of the surgery when the apnea low breathing index of the sleep multiple examination decreases by more than 50% compared to the apnea low breathing index of the surgical prediction result.

In addition, the prediction unit (not shown) may predict the success of the operation when the apnea low respiration index calculated from the surgery prediction result is less than 20.

That is, the present invention can predict surgical results based on the predicted rate of apnea reduction.

In addition, the present invention can perform machine learning on the apnea lowering index reduction rate using the results of the surgery performed according to the predicted apnea lowering index reduction rate.

Accordingly, the present invention can provide help to patients from an economic, social and personal perspective by predicting surgical outcomes prior to surgery and avoiding risk complications associated with the surgery.

3 is a diagram illustrating machine learning about the results of surgery according to an embodiment of the present invention.

Specifically, Figure 3 illustrates a signal transmission network of the machine learning unit to increase the accuracy of the surgical results prediction through the machine learning using the surgical results.

Referring to FIG. 3, the machine learning unit configures a network as an input layer 301, a hidden layer 302, and an output layer 303, and collects data related to various variables x in the input layer.

For example, various variables (x) include the amygdala size rating and the results of the sleep polynomial test.

)을 산출한다.For example, the machine learning unit applies the first constant (w) and the second constant (b) to the variable (x) in the hidden layer 302, and using the equation 1 apnea low apnea index reduction rate (

) Is calculated.

)을 출력한다.According to the exemplary embodiment of the present invention, the machine learning unit may reduce the apnea low respiratory rate in the output layer 303 (

)

)의 정확도를 향상시킬 수 있다.In addition, the machine learning unit updates the first constant (w) and the second constant (b) by applying a backward propagation technique to the data of the surgical result after the operation, the apnea low respiratory index decrease rate (

) Can improve the accuracy.

4 is a diagram illustrating a flowchart related to a method for predicting apnea low respiratory rate reduction rate according to an embodiment of the present invention.

Specifically, FIG. 4 illustrates a procedure for predicting a surgical outcome by calculating an apnea lowering index reduction rate indicating how much the apnea lowering index decreases before surgery.

Referring to FIG. 4, in step 401, the apnea low respiratory index reduction rate prediction method determines a tonsil size grade.

That is, the apnea low respiration index prediction method may determine the tonsil size grade in consideration of the visibility of the tonsil based on at least one of the tonsil size, pharyngeal area, tonsil column and palatal number.

In step 402, the apnea low respiratory rate reduction rate prediction method collects the results of the sleep multiple examination.

That is, the apnea low respiration index prediction method collects at least one of the apnea low respiratory index, the minimum oxygen saturation, snoring ratio and the wakefulness index that is a result of the sleep multiple test.

In step 403, the apnea low respiratory index reduction rate prediction method calculates a decrease rate of the apnea low apnea index by using the tonsil magnitude grade and the resultant value.

That is, the apnea low respiratory index reduction rate prediction method may calculate the reduction rate of the apnea low apnea index that is reduced compared to the existing apnea low apnea index using the one-way size grade and the result value.

FIG. 5 is a flowchart illustrating a method of predicting apnea low respiration index reduction rate according to an embodiment of the present invention.

Specifically, FIG. 5 illustrates a procedure in which the apnea low respiratory index reduction rate prediction method determines the tonsil size rating in consideration of the visibility of the tonsil based on the position of the tonsil column and palate.

Referring to FIG. 5, in step 501, the apnea low respiratory rate decrease rate prediction method determines whether visibility of the tonsil by the tonsil pillar is secured.

The apnea low respiratory rate reduction rate prediction method proceeds to step 502 if the visibility of the tonsils by the tonsil column is not secured, and to step 503 if secured.

In step 502, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the first grade and ends the procedure.

In step 503, the apnea low respiratory rate reduction rate prediction method determines whether visibility of the tonsil is secured around the tonsil pillar.

The apnea low respiratory rate reduction rate prediction method proceeds to step 504 if the visibility of the tonsil is secured around the tonsil column, and to step 505 if it is not secured.

For example, the method of predicting apnea low respiration index reduction rate determines that the tonsil is further extended beyond the tonsil pillar when the tonsil visibility is not secured around the tonsil pillar.

In step 504, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the second grade.

In step 505, the method for predicting apnea reduction rate is determined whether visibility of the amygdala is secured below a baseline based on the position of the tonsil column and palate.

That is, the apnea low respiratory index reduction rate prediction method proceeds to step 506 when the tonsil is identified below the baseline corresponding to half the position from the tonsil column at the distance between the tonsil column and the palate, and confirms the tonsil above the baseline. If so, the process proceeds to step 507.

In step 506, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the third grade.

That is, the apnea low respiration index prediction method determines the tonsil size grade as the third grade when the tonsil is identified closer to the tonsil column between the tonsil column and the palate.

In step 507, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the fourth grade. That is, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the fourth grade when the tonsil is identified near the palatal number between the tonsil column and the palatal number.

FIG. 6 is a flowchart illustrating a method for predicting apnea low respiration index reduction rate according to an embodiment of the present invention.

Specifically, FIG. 6 illustrates a procedure for determining a tonsil size grade by comparing the tonsil size and the area of the pharynx in the apnea low respiratory index reduction rate prediction method.

Referring to FIG. 6, in step 601, the method of predicting apnea reduction rate compares the size of the amygdala and the area of the pharynx. In other words, the apnea low respiratory index reduction rate prediction method can compare the size of the tonsil and the area of the pharynx to determine the tonsil size grade according to the visibility of the tonsil.

In step 602, the apnea low respiratory rate reduction rate prediction method determines whether the tonsil size is 1/4 or less of the pharyngeal area, and proceeds to one of steps 603 and 604 according to the determination result.

In step 603, the apnea low respiratory rate decrease rate prediction method determines that the tonsil size is 1/4 or less of the pharyngeal area, determines the tonsil size as the first grade, and ends the procedure.

In one example, the apnea low respiratory index reduction rate prediction method determines the tonsil size grade as the first grade when the tonsil visibility based on the tonsil size and the area of the pharynx corresponds to the first level. Here, the first level includes the visibility of the tonsils when the tonsil size is found to be 1/4 or less of the area of the pharynx.

In step 604, the method for predicting apnea reduction rate is determined whether the tonsil size is 1/2 or less of the pharyngeal area, and proceeds to one of steps 605 and 606 according to the determination result.

In step 605, the apnea low respiratory rate decrease rate prediction method determines that the tonsil size is 1/2 or less of the pharyngeal area, determines the tonsil size as the second grade, and ends the procedure.

In one example, the apnea low respiratory index reduction rate prediction method determines the tonsil size grade as the second grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the second level.

In step 606, the apnea low respiratory rate decrease rate prediction method determines whether the tonsil size is 3/4 or less of the pharyngeal area, and proceeds to either step 607 or 608 according to the determination result.

In step 607, the method for predicting apnea lowering index decrease rate determines that the tonsil size is 3/4 or less of the pharyngeal area, determines the tonsil size as the third grade, and ends the procedure.

In one example, the apnea low respiratory rate reduction rate prediction method determines the tonsil size grade as the third grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the third level. Here, the third level includes the visibility of the tonsils when the tonsil size is located between 1/2 and 3/4 of the area of the pharynx.

In step 608, the apnea low respiratory rate decrease rate prediction method determines whether the tonsil size is less than or equal to the total pharyngeal area, and proceeds to step 609 or terminates the procedure according to the determination result.

In step 609, the apnea low respiratory rate decrease rate prediction method determines that the tonsil size is less than or equal to the entire pharyngeal area, determines the tonsil size as the fourth grade, and ends the procedure.

In one example, the apnea low respiratory index reduction rate prediction method determines the tonsil size grade as the fourth grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the fourth level.

Here, the fourth level includes the visibility of the tonsils when the tonsil size is located between 3/4 and 1 of the area of the pharynx.

FIG. 7 is a flowchart illustrating a method for predicting apnea low respiration index reduction rate according to an embodiment of the present invention.

Specifically, FIG. 7 illustrates a method of predicting apnea low respiratory index reduction rate according to an embodiment of the present invention by using machine learning of a surgical result to improve prediction accuracy.

Referring to FIG. 7, in step 701, the apnea low respiratory rate reduction rate prediction method according to an embodiment of the present invention calculates a reduction rate by applying a first constant and a second constant to first to fifth values.

That is, a method for predicting apnea low respiratory rate reduction rate according to an embodiment of the present invention includes a first value representing a tonsil magnitude scale, a second value representing apnea low breathing rate, a third value representing minimum oxygen saturation, and a snoring ratio. The preoperative apnea hypopnemic index reduction rate is calculated using the fourth numerical value indicated and the fifth numerical value indicating the arousal index, the first constant, and the second constant.

In step 702, the method for predicting apnea reduction rate according to an embodiment of the present invention derives a surgical result by performing a surgery.

That is, the apnea low respiratory index reduction rate prediction method performs the operation based on the surgical prediction results, and measures the apnea low apnea index from the surgical results to generate the data of the surgical results.

In step 703, the method for predicting apnea reduction rate according to an embodiment of the present invention updates the first and second constants using the surgical results.

That is, the apnea low respiratory rate decrease rate prediction method performs a machine learning based on the data of the surgical results, and updates the first constant and the second constant according to the performance result.

8 is a view for explaining a comparison result for the reduction rate prediction according to an embodiment of the present invention.

Referring to FIG. 8, the horizontal axis of the graph may represent a preoperative predicted reduction rate, and the vertical axis may represent a postoperative reduction rate. Here, the preoperative predicted reduction rate includes apnea low respiratory rate reduction rate predicted based on the method of predicting apnea low respiratory index reduction rate of the present invention before surgery, and the postoperative reduction rate is apnea based on apnea low apnea index measured after surgery Includes exponential decay rates.

The postoperative reduction rate is collected based on the results after the surgery and the graph is concentrated between 0% and 100%.

In addition, the predicted preoperative reduction rate is collected based on the preoperative calculation result, and the graph is concentrated at 0% to 100%.

That is, the apnea low respiratory index decrease rate prediction method according to an embodiment of the present invention can predict the apnea low respiratory index decrease rate with high accuracy, and can predict the preoperative surgery results based on the apnea low respiratory index decrease rate.

Methods according to the embodiments described in the claims or the specification of the present invention may be implemented in the form of hardware, software, or a combination of hardware and software.

Such software may be stored on a computer readable storage medium. The computer readable storage medium includes at least one program (software module), at least one program including instructions for causing the electronic device to perform the method of the present invention when executed by the at least one processor in the electronic device. Save it.

Such software may be in the form of volatile or non-volatile storage, such as Read Only Memory (ROM), or random access memory (RAM), memory chips, devices or In the form of memory such as integrated circuits, or in compact disc ROM (CD-ROM), digital versatile discs (DVDs), magnetic disks or magnetic tapes tape, or the like, on an optical or magnetic readable medium.

Storage device and storage media are embodiments of machine-readable storage means suitable for storing a program or programs that include instructions that, when executed, implement embodiments.

In the above-described specific embodiments, the components included in the invention are expressed in the singular or plural according to the specific embodiments presented.

However, the singular or plural expressions are selected to suit the circumstances presented for convenience of description, and the above-described embodiments are not limited to the singular or plural elements, and the singular or plural elements may be formed of the singular or However, even a component expressed in the singular may be configured in plural.

Meanwhile, in the description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the technical idea included in the various embodiments.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims below and equivalents thereof.

100: surgery prediction target 110: thoracic motion measuring device
120: oxygen saturation measuring device 130: nasal air flow measuring device
140: brain motion measuring device 150: data recording device
160: apnea low respiratory index reduction rate prediction device
200: apnea low respiratory index reduction rate prediction device
210: one-way size rating determiner 220: data collector
230: output unit 240: machine learning unit

Claims (14)

A tonsil size grade determining unit for determining a tonsil size grade in consideration of visibility of tonsil based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula;
A data collection unit collecting a result value of polysomnography; And
Comprising a calculation unit for calculating the rate of reduction of apnea low breathing index by using the determined tonsil size grade and the collected result value,
The tonsil size grade determiner determines the tonsil size grade as the first grade when the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the first level, If the visibility of the tonsil based on the size and area of the pharynx corresponds to the second level, the tonsil size grade is determined as the second grade, and the tonsil size and area of the pharynx If the visibility of the tonsil based on corresponds to the third level, determine the tonsil size grade as the third grade, and the visibility of the tonsil based on the tonsil size and the area of the pharynx If it corresponds to this fourth level, the tonsil size grade is determined as the fourth grade.
Apnea low respiratory index reduction rate prediction device.
The method of claim 1,
The tonsil size grade determining unit determines the tonsil size grade as the first grade when the visibility of the tonsil by the tonsil is not secured, and determines the tonsil of the tonsil around the tonsil pillar. When visibility is ensured, the tonsil size grade is determined as the second grade.
Apnea low respiratory index reduction rate prediction device.
The method of claim 2,
The amygdala size grading unit may determine the tonsil size when the visibility of the tonsil is secured below a reference line corresponding to half the position of the amygdala at a distance between the amygdala and the palatine uvula. A grade is determined as the third grade, and when the visibility of the tonsil is secured above the baseline, the tonsil size grade is determined as the fourth grade.
Apnea low respiratory index reduction rate prediction device.
delete The method of claim 1,
The data collection unit may include first data about an apnea low breathing index calculated by adding up the number of apneas and the number of low breaths of a surgery prediction target, and second data about a minimum oxygen saturation required for sleep of the surgery prediction target Collecting third data about a snoring ratio during the sleep prediction time of the surgery prediction target and fourth data on the wakeup index related to the number of awakenings of the surgery prediction target at the sleep time.
Apnea low respiratory index reduction rate prediction device.
A tonsil size grade determining unit for determining a tonsil size grade in consideration of visibility of tonsil based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula;
A data collection unit collecting a result value of polysomnography; And
Comprising a calculation unit for calculating the rate of reduction of apnea low breathing index by using the determined tonsil size grade and the collected result value,
The data collection unit may include first data about an apnea low breathing index calculated by adding up the number of apneas and the number of low breaths of a surgery prediction target, and second data about a minimum oxygen saturation required for sleep of the surgery prediction target Collecting third data on a snoring ratio during the sleep prediction time of the surgery prediction target and fourth data on the wakeup index related to the number of awakening of the surgery prediction target at the sleep time,
The calculator calculates a first value by adding a first value representing the determined one-way size grade and a second value of the first data, and subtracts a third value of the fourth data from the calculated first value. Calculating a reduction value by adding a second value and a third value calculated by subtracting a fifth value of the third data from a fourth value of the second data;
Apnea low respiratory index reduction rate prediction device.
The method of claim 6,
The calculator may apply a weight differentially applied to the first value according to the determined one-way size class, and reflect the first value to which the weight is applied to the first value.
Apnea low respiratory index reduction rate prediction device.
The method of claim 7, wherein
The calculator may reflect the first numerical value to the first value only when the determined one-way size grade is any one of a third grade and a fourth grade.
Apnea low respiratory index reduction rate prediction device.
The method of claim 6,
The calculator calculates the reduction rate by applying a first constant and a second constant to each of the first to fifth values.
Apnea low respiratory index reduction rate prediction device.
The method of claim 9,
Further comprising a machine learning unit for collecting the surgical results after the operation of the surgery prediction target to update the first and second constants
Apnea low respiratory index reduction rate prediction device.
In the tonsil size grading unit, determining a tonsil size grade in consideration of visibility of tonsil based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula;
Collecting a result value of polysomnography in a data collection unit; And
Calculating a reduction rate of the apnea low breathing index by using the determined tonsil size grade and the collected result value;
Determining the tonsil size grade,
If the visibility of the tonsil based on the tonsil size and area of the pharynx corresponds to the first level, the tonsil size grade is determined as the first grade, and the tonsil size and the pharynx If the visibility of the tonsil based on the area of is corresponding to the second level, the tonsil size grade is determined as the second grade, and the tonsil based on the tonsil size and the area of the pharynx When the visibility of the corresponds to the third level, the tonsil size grade is determined as the third grade, and the visibility of the tonsil based on the tonsil size and the area of the pharynx corresponds to the fourth level. If it is determined that the tonsil size grade as a fourth grade
Prediction method for apnea low respiration index reduction rate.
The method of claim 11,
Determining the tonsil size grade,
If the visibility of the tonsil by the tonsil is not secured, determining the tonsil size grade as the first grade;
When the visibility of the tonsil is secured around the tonsil column, determining the tonsil size grade as the second grade;
If the visibility of the tonsil is secured below a baseline corresponding to half the position from the tonsil column at a distance between the tonsil column and the palatine uvula, the tonsil size grade is changed to the third grade. Determining; And
If the visibility of the tonsil is secured above the baseline, determining the tonsil size grade as the fourth grade;
Prediction method for apnea low respiration index reduction rate.
The method of claim 11,
Collecting the results of the sleep polysomnic test,
Collecting first data on an apnea index calculated by adding up the number of apneas and the number of low apneas to be predicted for surgery;
Collecting second data on a minimum oxygen saturation required for sleep of the surgical target;
Collecting third data on the snoring rate during the sleep time of the polysomnography test; And
Collecting fourth data on an awakening index related to the number of awakenings of the target for predicting surgery at the sleep time;
Prediction method for apnea low respiration index reduction rate.
In the tonsil size grading unit, determining a tonsil size grade in consideration of visibility of tonsil based on at least one of tonsil size, area of pharynx, tonsil column and palatine uvula;
Collecting a result value of polysomnography in a data collection unit; And
Calculating a reduction rate of the apnea low breathing index by using the determined tonsil size grade and the collected result value;
Collecting the results of the sleep polysomnic test,
Collecting first data on an apnea index calculated by adding up the number of apneas and the number of low apneas to be predicted for surgery;
Collecting second data on a minimum oxygen saturation required for sleep of the surgical target;
Collecting third data on the snoring rate during the sleep time of the polysomnography test; And
Collecting fourth data on an awakening index related to the number of awakenings of the target for predicting surgery at the sleep time,
Calculating a rate of decrease of the apnea low breathing index,
Calculating a first value by adding a first value representing the determined one-way size grade and a second value of the first data;
Calculating a second value by subtracting a third value of the fourth data from the calculated first value; And
Calculating the reduction rate by adding the second value and the third value calculated by subtracting the fifth value of the third data from the fourth value of the second data.
Prediction method for apnea low respiration index reduction rate.

Images