KR20170123924A - Method and apparatus of predicting probability of occuring oral cavity cancer - Google Patents

Abstract

The present invention relates to a method and apparatus for predicting the probability of an oral cavity cancer. The method for predicting the probability of an oral cavity cancer according to an embodiment of the present invention includes the steps of: receiving a clinical parameter of a patient and a parameter of each of proteins related to the oral cavity cancer of the patient; calculating a predictive score of the incidence of the oral cavity cancer based on the parameters; and calculating the probability of the oral cavity cancer within a predetermined period based on the predictive score of the incidence of the oral cavity cancer. Accordingly, the present invention can provide the method and apparatus for predicting the probability of an oral cavity cancer, capable of predicting the incidence of the oral cavity cancer with high probability.

Description

[0001] METHOD AND APPARATUS FOR PREDICTING PROBABILITY OF OCCURING ORAL CAVITY CANCER [0002]

The present invention relates to a method and apparatus for predicting the incidence of oral cancer, and more particularly, to a method and apparatus for predicting the incidence of oral cancer that can predict the incidence of oral cancer with high accuracy.

Oral cancer is one of the 10 most common cancer in the world, and the incidence varies widely depending on race, region, and socioeconomic conditions. For example, in developed countries, oral cancer accounts for less than 5% of all cancers, but it is reported to be the third most common cancer in developing countries after cervical cancer and stomach cancer. In Europe, in particular, there were 50,000 new cases of oral cancer in 1995 and 20,000 people died of oral cancer. In addition, 60% of males and 40% of women with oral cancer in the United States reported that males were more likely to have oral cancer than females. In Korea, about 5% of the total cancers of oral cancer occur, and the mortality rate is 5th in the case of men, followed by gastric cancer, lung cancer, liver cancer and colon cancer, and 7th in women. . In Korea, 1.7 men and 0.7 women per 100,000 men die annually from oral squamous cell carcinoma. In the past, oral cancer was predominantly present in older people, but in recent years it has been reported that the number of oral cancer patients is increasing due to the increase in smoking and drinking.

These oral malignancies are highly malignant and can be cured 80 to 90% initially, but more than half of the patients are found after more than three thirds of oral cancer, so the cure rate is low and the prognosis is bad. Furthermore, surgical resection of oral cancer may result in widespread facial damage and malfunctioning, for example, vocalization, esophageal damage, and may also cause aesthetic and psychological problems, so early detection and prevention of oral cancer is very difficult It can be said that it is important. Treatment is performed using surgical, chemotherapy, or radiation alone or in combination. Recently, immunotherapy, thermotherapy, photodynamic therapy, and gene therapy have been introduced experimentally and clinically, but radiation therapy is still considered to be an end-stage, recurrence, and systematic condition that are not suitable for surgery. .

Oral cancer progresses from the normal tissue to the cancer through the precancerous lesion. Therefore, screening of patients who are likely to progress to oral cancer at the stage of oral precancerous lesion is very important to prevent oral cancer. In Korea, vitiligo and pancreatic lesions are known as precancerous lesions that cause cancer. In Sri Lanka, submucosal fibrosis is known as a precancerous lesion. Thus, it is very important to identify patients who are likely to progress to oral cancer at the stage of precancerous lesions.

Thus, there is a growing need to provide a method for predicting the incidence of oral cancer by observing differences in the expression of various proteins in oral precancerous lesions that can progress to cancer and oral precancerous lesions that will not progress to cancer.

[Related Technical Literature]

Anti-IGF-1R monoclonal antibodies and compositions for treating cancer comprising the same (Korean Patent Laid-Open No. 10-2014-0138533)

The precancerous lesions that cause oral cancer are very diverse. Studies on various precancerous lesions that cause oral cancer have been actively conducted, but only the influence of a single precancerous lesion on the development of oral cancer has been studied in the past. For example, there were biomarkers associated with Ki-67 expression and biomarkers associated with p53.

However, since precancerous lesions that generate oral cancer are variable, it is difficult to predict whether specific precancerous lesions develop into actual oral cancer in each patient. Therefore, when one precancerous lesion predicts oral cancer, it may be very unreliable. Accordingly, the inventors of the present invention have invented a method capable of predicting the incidence of oral cancer at a high probability by considering various precancerous lesions capable of generating oral cancer. Furthermore, the inventors of the present invention have been able to increase the probability of predicting the onset of oral cancer in consideration of clinical variables in precancerous lesions.

Accordingly, a problem to be solved by the present invention is to provide a method and apparatus for predicting the incidence of oral cancer that can predict the onset of oral cancer with a high probability.

Another problem to be solved by the present invention is to provide a method and apparatus for predicting the onset of oral cancer, which can calculate the probability of taking oral cancer within a predetermined period by combining a plurality of proteins that cause oral cancer and clinical variables.

The problems of the present invention are not limited to the above-mentioned problems, and other problems not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method of predicting the incidence of oral cancer, the method comprising: receiving clinical parameters of a patient and parameters of a plurality of proteins associated with oral cancer of a patient; And calculating a probability of developing oral cancer within a predetermined period based on the predicted score of oral cancer development.

According to another aspect of the present invention, the step of receiving the parameter may include outputting an input nomogram that allows the parameter to be input.

According to still another aspect of the present invention, the step of calculating the predicted oral cancer incidence score may include the step of outputting a risk graph indicating the risk of developing oral cancer based on the predicted oral cancer incidence predictive score.

According to another aspect of the present invention, the method may further include outputting a probability nomogram that is configured to calculate a probability of developing oral cancer within at least a predetermined period of time.

According to another aspect of the present invention, the clinical parameters include the age, gender, gum condition, and tongue status of the patient, and the protein associated with oral cancer may include at least two kinds of proteins.

According to another aspect of the invention, the method may further comprise determining an optimal clinical parameter and an optimal protein using a survival function prior to receiving the parameter.

According to another aspect of the invention, the clinical variables may include the age of the patient.

According to another aspect of the present invention, the plurality of proteins may comprise at least p53 and CA9.

In order to solve the above-mentioned problems, an apparatus for predicting the incidence of oral cancer according to another embodiment of the present invention includes a receiver for receiving clinical parameters of a patient and parameters of a plurality of proteins related to oral cancer of a patient, And a probability calculating unit for calculating a probability of developing oral cancer within a predetermined period based on the prediction score of the onset of oral cancer.

According to another aspect of the present invention, an output unit connected to at least one of the receiving unit, the score calculating unit, and the probability calculating unit may be further included.

According to another aspect of the present invention, the output unit may output an input nomogram to which parameters can be input.

According to another aspect of the present invention, the output unit may output a risk graph indicating the risk of developing oral cancer based on the estimated predicted oral cancer incidence score.

According to another aspect of the present invention, the output unit may output a probability nomogram that is configured to calculate a probability of developing oral cancer within at least a predetermined period.

According to another aspect of the present invention, the clinical parameters include the age, gender, gum condition, and tongue status of the patient, and the protein associated with oral cancer may include at least two kinds of proteins.

According to still another aspect of the present invention, before receiving the parameter, the optimizing unit may further comprise an optimizing unit for determining an optimum clinical parameter and an optimal protein using a survival function.

According to another aspect of the invention, the clinical variables may include the age of the patient.

According to another aspect of the present invention, the plurality of proteins may comprise at least p53 and CA9.

According to an aspect of the present invention, there is provided a computer-readable recording medium storing instructions for providing a method for predicting the incidence of oral cancer according to an embodiment of the present invention, And calculates a predicted incidence of oral cancer based on the parameters and calculates the incidence probability of oral cancer within a predetermined period based on the predicted incidence of oral cancer.

The details of other embodiments are included in the detailed description and drawings.

The present invention has an effect of providing a method and apparatus for predicting the onset of oral cancer that can predict the onset of oral cancer with a high probability.

The present invention provides a method and apparatus for predicting the incidence of oral cancer that can calculate the probability of taking oral cancer within a predetermined period by combining clinical proteins and a plurality of proteins that cause oral cancer.

The effects according to the present invention are not limited by the contents exemplified above, and more various effects are included in the specification.

FIG. 1 shows a schematic configuration of a device for predicting the onset of oral cancer according to an embodiment of the present invention and a relationship with a patient / examinee.
FIG. 2 shows a procedure for calculating the probability of developing oral cancer according to the method for predicting the incidence of oral cancer according to an embodiment of the present invention.
3 illustrates an exemplary input / output nomogram according to an embodiment of the present invention.
Figure 4 illustrates an exemplary accuracy distribution graph for each clinical parameter and protein combination in accordance with one embodiment of the present invention.
5 illustrates an exemplary model nomogram according to another embodiment of the present invention.
FIGS. 6A and 6B illustrate an exemplary optimal input / output nomogram showing the probability of developing oral cancer according to another embodiment of the present invention. FIG.
7 illustrates an exemplary risk graph depicting the risk of developing oral cancer according to another embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention and the manner of achieving them will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

The shapes, sizes, ratios, angles, numbers, and the like disclosed in the drawings for describing the embodiments of the present invention are illustrative, and thus the present invention is not limited thereto. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. Where the terms "comprises", "having", "done", and the like are used in this specification, other portions may be added unless "only" is used. Unless the context clearly dictates otherwise, including the plural unless the context clearly dictates otherwise.

In interpreting the constituent elements, it is construed to include the error range even if there is no separate description.

Although the first, second, etc. are used to describe various components, these components are not limited by these terms. These terms are used only to distinguish one component from another. Therefore, the first component mentioned below may be the second component within the technical spirit of the present invention.

Like reference numerals refer to like elements throughout the specification unless otherwise specified.

It is to be understood that each of the features of the various embodiments of the present invention may be combined or combined with each other partially or entirely and technically various interlocking and driving is possible as will be appreciated by those skilled in the art, It may be possible to cooperate with each other in association.

Hereinafter, terms used in this specification will be defined.

Clinical variables herein are meant to serve as variables in clinical trials on oral cancer. That is, the clinical parameters may be clinical parameters that may affect the onset of oral cancer such as the patient's age, sex, occupation, gum status, tongue status, drinking status and smoking status.

In the present specification, the optimal clinical variable means a clinical variable determined to have the greatest influence on the development of oral cancer among various clinical parameters that may affect the development of oral cancer.

The term " optimal protein " as used herein means a protein that is determined to have the greatest influence on the development of oral cancer among various proteins that may affect onset of oral cancer.

FIG. 1 shows a schematic configuration of a device for predicting the onset of oral cancer according to an embodiment of the present invention and a relationship with a patient / examinee. FIG. 2 shows a procedure for calculating the probability of developing oral cancer according to the method for predicting the incidence of oral cancer according to an embodiment of the present invention. Will be described later with reference to Figs. 1 and 2 for convenience of explanation.

Referring to FIG. 1, the apparatus 100 for predicting the onset of oral cancer includes a receiving unit 110, a score calculating unit 120, and a probability calculating unit 130. The apparatus 100 for predicting the onset of oral cancer may be a device for driving web pages, applications, and the like, for example, a generic computing device, a mobile device, a server, and the like. The apparatus for predicting the incidence of oral cancer 100 may include components of a processor, a storage unit, a memory, an input unit and an output unit. The receiving unit 110, the score calculating unit 120 and the probability calculating unit 130 Lt; / RTI > elements. Server, the apparatus 100 for predicting the onset of oral cancer may be driven to transmit the calculated values to another device having an output unit. The oral cancer outbreak probability prediction apparatus 100 can receive clinical parameters of the patient 10 and parameters for each of a plurality of proteins associated with oral cancer to predict the probability of developing oral cancer for the patient 10. [

More specifically, in the apparatus 100 for predicting the onset of oral cancer, the receiving unit 110 receives the clinical parameters of the patient 10 and the parameters of the plurality of proteins related to the oral cancer of the patient 10 (S210).

1 and 2, the apparatus for predicting the incidence of oral cancer 100 receives parameters for predicting the incidence of oral cancer in the patient 10 through the receiving unit 110. [ That is, the receiving unit 110 receives the clinical parameters of the patient 10 and parameters for each of a plurality of proteins associated with oral cancer of the patient 10. [ For example, the receiving unit 110 may receive clinical parameters such as the sex, age, gingival state, and tongue status of the patient 10, and may include parameters for proteins related to oral cancer such as p53, CA9 and P16 .

Furthermore, an output unit (not shown) connected to the receiving unit 110 can visually output an input nomogram for inputting parameters. Here, although not shown, the output unit may be included in the apparatus 100 for predicting the onset of oral cancer. In addition, the input nomogram is a nomogram that can input parameters required for predicting the development of oral cancer. For a specific method in which the receiver 110 receives various parameters through the input nomogram, see FIGS. 3 and 5 Will be described later.

Next, in the apparatus 100 for predicting the onset of oral cancer, the score calculating unit 120 calculates a predicted oral cancer incidence score for the patient based on the parameters (S220).

The score calculating unit 120 receives a parameter from the receiving unit 110 and calculates a predicted oral cancer incidence score for the patient based on the parameter. Specifically, the score calculating unit 120 may match a predetermined value for each parameter for each of a plurality of proteins related to clinical parameters and oral cancer. For example, if the age of the clinical variables is less than 54 years old, the score is 35, and if the value of p53 in the oral cancer-related protein is low, 100 points can be matched to the p53 parameter. Although not limited, parameters associated with a protein can be obtained by measuring the amount of expression thereof through immunohistochemical staining.

Accordingly, the score calculating unit 120 may calculate the predicted score of oral cancer incidence by summing the scores matched to the parameters for calculating the probability of developing oral cancer. The predictive score of oral cancer incidence can be used to calculate the probability of developing oral cancer. In particular, the predicted score of oral cancer development may improve the accuracy of the probability of developing oral cancer by adjusting some of the scores matched to each of the parameters.

Furthermore, the output unit (not shown) connected to the score calculating unit 120 may output a risk graph indicating the risk of oral cancer based on the calculated predicted score of the onset of oral cancer. Here, the risk graph is a graph showing the onset probability corresponding to the predicted oral cancer incidence score according to the change of time. The graph of the risk outputted according to the predicted oral cancer incidence score calculated by the score calculating unit 120 is shown in FIG. 6 Will be described later.

Next, in the apparatus 100 for predicting the onset of oral cancer, the probability calculation unit 130 calculates the onset probability of oral cancer within a predetermined period based on the estimated predicted oral cancer incidence score (S230).

The probability calculating unit 130 receives the predicted score of the onset of oral cancer calculated by the score calculating unit 120 and calculates the probability of developing oral cancer for the patient based on the received score. Specifically, the probability calculator 130 may match the probability of developing oral cancer corresponding to the predicted score of oral cancer incidence through the monogram. In addition, the probability calculation unit 130 may match the probability that oral cancer will occur after a predetermined period of time through a monogram. Here, the predetermined period may be, for example, 5 years, 10 years and 15 years, and may be more finely determined as the time for predicting the onset of oral cancer, or may be determined as a longer period.

Thus, the monogram that matches the predictive score of oral cancer incidence and the incidence of oral cancer incidence can be determined by the formula for calculating the incidence of oral cancer. For example, the expression for calculating the probability of developing oral cancer can be expressed as [Equation 1], [Equation 2] and [Equation 3] for the predetermined period of 5 years, 10 years and 15 years, respectively.

Here, 'total' means the predicted score of onset of oral cancer, and the constants described in [Equation 1], [Equation 2] and [Equation 3] are determined by a nonlinear least square method in a predetermined period It is a value obtained by calculating a predictive probability value for each of them, and may be changed by various variables. Further, the output unit (not shown) connected to the probability calculation unit 130 may further output a probability nomogram that is configured to calculate the probability of developing oral cancer within a predetermined period. A probabilistic nomogram constructed to calculate the probability of developing oral cancer by matching the prediction score of oral cancer incidence with the probability of developing oral cancer will be described later with reference to FIG.

Accordingly, when a patient or examinee inputs a clinical parameter and a protein parameter for predicting the incidence of oral cancer in the apparatus 100 for predicting the onset of oral cancer, the clinical parameters and the parameters for the protein are quantified to predict the onset of oral cancer And the probability of developing oral cancer can be calculated. Based on the prediction score of the onset of oral cancer and the probability of onset, the patient / examinee closely monitors the onset of oral cancer, and if the probability of oral cancer is high, the oral cancer can be diagnosed through more precise examination, You can provide an opportunity.

The respective components of the apparatus 100 for predicting the onset of oral cancer are shown as individual components for the sake of convenience of explanation, and they may be implemented in one configuration or separated into two or more configurations depending on the implementation method.

3 illustrates an exemplary input / output nomogram according to an embodiment of the present invention.

Referring to FIG. 3, the input / output nomogram 300 includes an input nomogram 310 and an output nomogram 320. The input nomogram 310 includes a clinical parameter input unit 310a and a protein parameter input unit 310b and the output nomogram 320 includes a mouth cancer incidence prediction score output unit 321 and a plurality of oral cancer incidence probability output units 322, 323, and 324, respectively. In FIG. 3, the input / output nomogram 300 is not limited to the illustrated input and output units, and may further include an input unit for inputting additional clinical parameter and protein parameters, The probability of onset can be further displayed.

The clinical parameter input unit 310a includes a sex input unit 311 and an age input unit 312. [ The gender input unit 311 can receive the gender of the patient, and the age input unit 312 can be configured to select 54 years old or older and 54 years old or less based on 54 years old. For example, if the patient is female, the gender input unit 311 may receive 'Female', and the age input unit 312 may receive '<54' when the patient is under 54 years of age. Thus, the patient can enter his or her gender and age through a very simple criterion.

In addition, the clinical variables may include the age, gender, gum condition, and tongue status of the patient, and only the age and gender may be input through the input / output nomogram 300. That is, the patient / examinee can input only the more important parameters for predicting the incidence of oral cancer through the input / output nomogram 300. Further, the optimum clinical variables can be determined using the survival function before receiving the clinical parameter through the input / output nomogram 300, and the patient / examinee inputs only the parameters for the optimal clinical parameter through the input / output nomogram 300 You may. Accordingly, the patient / examinee can predict a more accurate probability of developing oral cancer by inputting only one parameter for the optimal clinical parameter. A detailed description of the basis for determining the optimal clinical parameters will be described later with reference to FIG.

The protein parameter input unit 310b includes a p53 input unit 313, a CA9 input unit 314, and a P16 input unit 315. [ The protein parameter input unit 310b may receive the detection levels of p53, CA9, and P16 obtained in the test for predicting the onset of oral cancer. For example, when a large number of p53, CA9, and P16 are detected from a patient through a test for predicting the development of oral cancer, the p53 input unit 313 receives 'high' and the CA9 input unit 314 receives 'high' The P16 input unit 315 receives 'high'. Accordingly, the patient / examinee can easily input the degree of detection of the protein related to oral cancer without inputting complicated numerical values of the degree of detection of the proteins obtained in the test for predicting the oral cancer incidence. However, the present invention is not limited thereto. In various embodiments, the protein parameter input unit 310b may be configured to receive information by numerical value.

3, the proteins associated with oral cancer are p53, CA9, and P16. However, the proteins related to oral cancer may include at least two kinds of proteins, and two types of proteins can be identified through the input / output mammogram 300 Only parameters for the parameters may be input. For example, only parameters for p53 and CA9 may be input through the input / output nomogram 300. [ That is, the optimum protein may be determined using the survival function before receiving the protein parameter related to oral cancer through the input / output nomogram 300, and only the parameter for the optimal protein may be input through the input / output nomogram 300 . Thus, only the parameters for the optimal protein are input, and a more accurate probability of developing oral cancer can be predicted. A detailed explanation on the basis for determining the optimal protein will be described later with reference to Fig.

The output nomogram 320 includes an oral cancer incidence prediction score output unit 321, a 5-year oral cancer incidence probability output unit 322, a 10-year oral cancer incidence probability output unit 323, and a 15-year oral cancer incidence probability output unit 324).

Specifically, the oral cancer outcome prediction score output unit 321 outputs the estimated oral cancer outcome score based on the input parameters. For example, the sex of the patient is 'Female', the age is 'under 54', the detection level of p53 in the patient's mouth is 'high', the detection level of CA9 is 'high' Quot; high &quot;, the mouth cancer onset prediction score output unit 321 can output 50 points. The method of calculating the predicted oral cancer incidence score will be described later with reference to FIG.

5 years later, the oral cancer incidence probability output unit 322 outputs the calculated result by substituting the output predicted score of oral cancer incidence into [Equation 1], and the oral cancer incidence probability output unit 323 after 10 years outputs the output of the oral cancer incidence The predicted score is substituted into the expression (2), and the calculated result is output. After 15 years, the oral cancer incidence probability output unit 324 substitutes the output predicted score of the oral cancer incidence into the expression (3) do. For example, when the prediction score of oral cancer development is 50 points, the probability of the oral cancer development output 322 is 0.7433933984551221 five years later, the probability of the oral cancer development output 323 is 0.5776311966231944 10 years later, 324) can output 0.4849962443395142.

Accordingly, when the patient / examinee inputs only the clinical parameter and the oral cancer-related protein parameter through the input / output nomogram 300, the predicted score of oral cancer development and the probability of oral cancer development can be easily confirmed. Furthermore, a procedure for determining the optimal clinical parameters and optimal protein to be able to predict the probability of developing oral cancer by the convenience of the patient / examinee and using only the minimum parameters will be described below with reference to FIG.

Figure 4 illustrates an exemplary accuracy distribution graph for each clinical parameter and protein combination in accordance with one embodiment of the present invention.

Referring to FIG. 4, the accuracy distribution graph 400 includes a prediction probability accuracy 410 for the first combination, a prediction probability accuracy 420 for the second combination, a prediction probability accuracy 430 for the third combination, 4 combination, a prediction probability accuracy 450 for the fifth combination, and a prediction probability accuracy 460 for the sixth combination.

The prediction probability accuracy 410 for the first combination takes into account the protein associated with one type of oral cancer and indicates the accuracy of the results of predicting the probability of developing oral cancer without considering a separate clinical parameter. The prediction probability accuracy 420 for the second combination takes into account the protein associated with one kind of oral cancer and indicates the accuracy of the prediction result of the probability of developing oral cancer when the age of the clinical variables is considered. The predictive probability accuracy for the third combination (430) represents the accuracy of the predicted outcome of oral cancer in the case of considering two types of oral cancer-related proteins and without consideration of separate clinical variables. The predicted probability accuracy (440) for the fourth combination represents the accuracy of the predicted outcome of oral cancer in consideration of the age of the clinical parameters, considering the proteins associated with two types of oral cancer. Prediction probability accuracy (450) for the 5th combination indicates the accuracy of the predicted outcome of oral cancer in consideration of three types of oral cancer-related proteins and no consideration of separate clinical variables. The predicted probability accuracy (460) for the sixth combination represents the accuracy of the results of predicting the probability of developing oral cancer when considering three types of proteins related to oral cancer and considering the age among clinical variables.

The predicted probability accuracy for each combination can represent a maximum value, a minimum value, a 75% value, a 25% value, and a median (445). Specifically, the accuracy maximum value 441 and the minimum accuracy value 442 are determined by the prediction probability accuracy 440 for the fourth combination on the basis of the prediction probability accuracy 440 for the fourth combination, Is displayed. The accuracy 75% value 443 and the accuracy 25% value 444 are represented by rectangles and the accuracy 75% value 443 is displayed to be associated with the accuracy maximum value 441 and the accuracy 25% And is displayed to be associated with the minimum value 442. The accuracy median 445 may be displayed inside a rectangle made by an accuracy 75% value 443 and an accuracy 25% value 444. [

Comparing the prediction probability accuracy for each combination with reference to FIG. 4, the prediction probability accuracy increases gradually from the prediction probability accuracy 410 for the first combination to the prediction probability accuracy 460 for the sixth combination. For example, the prediction probability accuracy 430 for the third combination based on information on two kinds of proteins is higher than the prediction probability accuracy 410 for the first combination based on information on only one kind of protein The predicted probability accuracy 450 for the fifth combination based on information about the three types of proteins is generally higher than the predicted probability accuracy 410 for the first combination and the predicted probability accuracy 430 for the third combination Generally, it appears high. That is, as the type of protein associated with oral cancer is increased, the accuracy of predicting the onset of oral cancer also increases.

In addition, when based on information on the same type of protein, the prediction probability accuracy is higher in combinations with added clinical variables. For example, comparing the prediction probability accuracy 430 for the third combination and the prediction probability accuracy 440 for the fourth combination based on the information of the two types of proteins, The prediction probability accuracy 440 for the fourth combination computed is generally higher.

That is, as the number of proteins associated with oral cancer is increased and the clinical variables are considered together, the accuracy of predicting the onset of oral cancer may increase.

4, there is no significant difference between the accuracy of the prediction probability based on the information on the two kinds of proteins and the accuracy of the prediction probability based on the information on the three kinds of proteins, It may be effective to predict the incidence of oral cancer by using only information on two types of proteins since the patient may be burdened or unnecessary. In addition, the prediction probability accuracy (440) for the fourth combination, in which the clinical variables are additionally considered in the information on the two kinds of proteins, is obtained by using the prediction probability accuracy (460), it may be effective to calculate the probability of developing oral cancer based on information on two kinds of proteins and clinical variables.

Further, in order to determine the optimal clinical parameter among the two kinds of proteins and the clinical variables among the plurality of proteins specifically related to oral cancer, the optimal clinical parameter and optimal protein can be determined using the survival function before receiving the parameters. Specifically, in order to determine the optimal protein, the same number of proteins among 10 types of proteins related to oral cancer are repeatedly extracted to determine two kinds of proteins having the highest prediction accuracy. The protein with the highest prediction accuracy is determined using a survival function, where the survival function S (t) is defined as: &quot; (4) &quot;

Where t is the time variable, T is the time to death, and Pr is the probability function. That is, the survival function S (t) means the probability of surviving longer than a specific time t.

Thus, through the survival function, the optimal protein associated with oral cancer is determined to include at least p53 and CA9, and the optimal clinical parameter among the clinical parameters is determined to include the age of the patient.

Hereinafter, a process of predicting the incidence of oral cancer based on the optimal protein and the optimal clinical parameters will be described.

5 illustrates an exemplary model nomogram according to another embodiment of the present invention. FIGS. 6A and 6B illustrate an exemplary optimal input / output nomogram showing the probability of developing oral cancer according to another embodiment of the present invention. FIG. The optimal input / output nomograms of FIGS. 6A and 6B are those in which some items are deleted in the input / output nomogram 300 shown in FIG. 3, and redundant description of the components already described with reference to FIG. 3 is omitted.

Referring to FIG. 5, the optimal model nomogram 500 includes a score axis 510, an optimal protein parameter axis 520, an optimal clinical parameter axis 530, a predicted oral cancer incidence score axis 540, And a probability axis 550. The optimal protein parameter axis 520 includes the p53 protein parameter axis 521 and the CA9 protein parameter axis 522 and the optimal clinical parameter axis 530 represents a parameter for the age of the patient. In addition, the periodical oral cancer incidence probability axis 550 includes the probability axis 551 of oral cancer after 5 years, the probability axis 552 of oral cancer 10 years later, and the probability axis 553 of oral cancer 15 years later.

Here, the score axis 510 is an axis for converting the parameters indicated in the optimal protein parameter axis 520 and the optimal clinical parameter axis 530 into points. For example, when 'low' is displayed on the p53 protein parameter axis 521, the parameter value corresponding to the p53 protein becomes 100 points. If 'low' is displayed on the CA9 protein parameter axis 522, Becomes 49 points. Similarly, if &lt; 54 &gt; is indicated on the optimal clinical parameter axis 530, then the parameter value for the patient's age is 35 points. However, without being limited thereto, the conversion of the scores may be variously variable.

If all of the parameter values shown in the optimal model nomogram 500 are summed up, the predicted score of the onset of oral cancer can be calculated, and the calculated predicted score of the onset of oral cancer is displayed on the predicted score axis of oral cancer 540. Based on the prediction score of oral cancer incidence predicted on the oral cancer incidence prediction axis 540, the probability of occurrence of the oral cancer incidence axis 551, the incidence probability of oral cancer incidence 552 after 10 years, and the incidence probability of oral cancer incidence 553 after 15 years The probabilities at the point of contact with each other when the waterline is lowered are calculated as the probability of developing oral cancer according to each period.

6A and 6B, an optimal input / output nomogram 600 includes an optimal input nomogram 610 and an optimal output nomogram 620, and an optimal input nomogram 610 includes an optimal clinical parameter input 610a and an optimal protein parameter input unit 610b. The optimum output nomogram 620 includes an oral cancer incidence prediction score output unit 621, a 5-year oral cancer incidence probability output unit 622, An output unit 623 and a mouth cancer incidence probability output unit 624 after 15 years. The optimal protein parameter input unit 610b includes a p53 input unit 611 and a CA9 input unit 612. [

Referring to FIG. 6A, when the patient's age is under 54 years old and the detection level of p53 is 'high' and the detection level of CA9 is 'low' among the proteins related to oral cancer, the optimal model nomogram 54 corresponding to 'low' in the CA9 protein parameter axis 522 and '54' in the optimal clinical parameter axis parameter 530, , And the predicted score of oral cancer incidence is calculated as a total of 84 points. The probability of developing oral cancer corresponding to the predicted oral cancer incidence score of '84' calculated in the optimal model nomogram 500 is output to the optimal input / output nomogram 600. For example, the probability of developing oral cancer is 0.8702214468704769, the probability of developing oral cancer is 0.7872025763580728 after 10 years, and the probability of developing oral cancer is 0.7345298372589231 after 15 years, corresponding to the predicted score of 84 for oral cancer.

6B, when the patient's age is under 54 years of age and the detection level of p53 is 'low' and the detection level of CA9 is 'high' among the proteins related to oral cancer, the optimal model nomogram 500 corresponding to 'low' at the p53 protein parameter axis 521, 0 at the CA9 protein parameter axis 522 corresponding to 'high' and '54' at the optimal clinical parameter axis parameter 530, , And the predicted score of oral cancer incidence is calculated as a total of 135 points. The probability of developing oral cancer corresponding to the predicted oral cancer incidence score of '135' calculated in the optimal model nomogram 500 is output to the optimal input / output nomogram 600. For example, the probability of developing oral cancer is 0.9774999999999999, the probability of developing oral cancer is 0.0.9624999999999999, and the probability of developing oral cancer after 15 years is 0.9541666666666666, corresponding to the predicted score of oral cancer, which is '135'.

According to the embodiment of the present invention, the patient inputs the optimal clinical parameters and optimal protein parameters for predicting the incidence of oral cancer through the input / output nomogram 600, The probability of the onset of oral cancer can be predicted with a high accuracy.

7 illustrates an exemplary risk graph depicting the risk of developing oral cancer according to another embodiment of the present invention.

Referring to FIG. 7, the onset risk graph 700 represents the probability of onset over time. Specifically, in the onset risk graph 700, the x-axis represents the change in time, expressed in terms of the month, and the y-axis represents the onset probability, 1.0 means no disease, And the probability of developing oral cancer is improved.

The onset risk graph 700 shows the probability of developing oral cancer according to the change of time, which is calculated by the optimal model nomogram 500 based on the parameters inputted through the input / output nomogram 600. The onset risk graph 700 includes a first graph 710 and a second graph 720. In the first graph 710, it is predicted that the oral cancer incidence probability is maintained at about 1.0 even if the time passes, and the oral cancer incidence is predicted not to occur. In the second graph 720, the probability of oral cancer incidence is about 0.7 It is predicted that oral cancer will develop with a high probability.

Accordingly, according to the embodiment of the present invention, when the probability of developing oral cancer is predicted over time, the probability of developing oral cancer may be different depending on the parameter as shown in the graph 700 of risk of developing oral cancer. As shown in the graph 700 of risk of onset of oral cancer, the probability of developing oral cancer of the patient according to the change of time can be predicted in the long term.

In this specification, each block or each step may represent a part of a module, segment or code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative embodiments, the functions mentioned in the blocks or steps may occur out of order. For example, two blocks or steps shown in succession may in fact be performed substantially concurrently, or the blocks or steps may sometimes be performed in reverse order according to the corresponding function.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, which is capable of reading information from, and writing information to, the storage medium. Alternatively, the storage medium may be integral with the processor. The processor and the storage medium may reside within an application specific integrated circuit (ASIC). The ASIC may reside within the user terminal. Alternatively, the processor and the storage medium may reside as discrete components in a user terminal.

Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, it is to be understood that the present invention is not limited to those embodiments and various changes and modifications may be made without departing from the scope of the present invention. . Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. Therefore, it should be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

10 patients
100 Oral cancer probability prediction device
110 receiver
120 score calculation unit
130 probability calculator
300 input and output nomogram
310 input nomogram
310a clinical parameter inputting unit
310b protein parameter input unit
311 Gender Input
312 Age input section
313, 611 p53 input unit
314, 612 CA9 input unit
315 P16 input unit
320 output nomogram
321 Oral cancer incidence prediction score output unit
322 5-year probability of oral cancer incidence output
323 10 years later Oral cancer incidence probability output
324 Percentage of oral cancer outbreak output after 15 years
400 Accuracy Distribution Graph
410 Prediction Probability Accuracy for First Combination
420 Prediction probability accuracy for the second combination
430 Prediction probability accuracy for the third combination
440 Prediction probability accuracy for the fourth combination
441 accuracy maximum
442 accuracy minimum
443 Accuracy 75% Value
444 Accuracy 25% Value
445 Accuracy Median
450 Prediction Probability Accuracy for Fifth Combination
460 Prediction probability accuracy for the sixth combination
500 Optimal Model Nomogram
510 score axis
520 Optimal protein parameter axis
521 p53 protein parameter axis
522 CA9 protein parameter axis
530 Optimal clinical parameter axis
540 Oral Cancer Prediction Score Axis
550 Probability axis of oral cancer development by period
551 Probability axis of oral cancer after 5 years
552 Probability axis of oral cancer after 10 years
553 Probability axis of oral cancer after 15 years
600 Optimal Input / Output Nomogram
610 Optimal input nomogram
610a optimal clinical parameter inputting part
610b Optimal protein parameter input unit
620 Optimal output nomogram
621 Oral cancer incidence prediction score output unit
622 Probability of Oral Cancer After 5 Years Output
623 Percentage of oral cancer outbreak output after 10 years
624 Probability of Oral Cancer after 15 Years Output
700 outbreak risk graph
710 1st graph
720 second graph

Claims (18)

Receiving a clinical parameter of a patient and a parameter for each of a plurality of proteins associated with oral cancer of the patient;
Calculating the predicted oral cancer incidence score for the patient based on the parameter; And
And calculating the onset probability of the oral cancer within a predetermined period based on the predicted oral cancer onset score. The method according to claim 1,
The step of receiving the parameter comprises:
And outputting an input nomogram capable of inputting the parameter. The method according to claim 1,
The step of calculating the predicted oral cancer incidence score may include:
And outputting a risk graph indicating the risk of developing oral cancer based on the calculated prediction score of developing oral cancer. The method according to claim 1,
Further comprising the step of outputting a probability nomogram that is configured to calculate the onset probability of the oral cancer within at least the predetermined period of time. The method according to claim 1,
Wherein the clinical parameter comprises the age, sex, gum condition, and tongue status of the patient, and wherein the protein associated with the oral cancer comprises at least two types of proteins. 6. The method of claim 5,
Further comprising determining an optimal clinical parameter and an optimal protein using a survival function prior to receiving the parameter. The method according to claim 1,
Wherein the clinical parameter comprises the age of the patient. The method according to claim 1,
Wherein said plurality of proteins comprises at least p53 and CA9. A receiver for receiving a clinical parameter of a patient and a parameter for each of a plurality of proteins associated with oral cancer of the patient;
A score calculating unit for calculating the predicted oral cancer incidence score for the patient based on the clinical parameters and the parameters; And
And a probability calculation unit for calculating a probability of occurrence of the oral cancer within a predetermined period based on the prediction score of development of the oral cancer. 10. The method of claim 9,
And an output unit connected to at least one of the receiving unit, the score calculating unit, and the probability calculating unit. 11. The method of claim 10,
Wherein the output unit outputs an input nomogram capable of inputting the parameter. 11. The method of claim 10,
Wherein the output unit outputs a risk graph indicating a risk of developing oral cancer based on the calculated predicted oral cancer incidence score. 11. The method of claim 10,
Wherein the output unit outputs a probability nomogram that is configured to calculate the onset probability of the oral cancer within at least the predetermined period. 10. The method of claim 9,
Wherein the clinical parameters include the age, sex, gum condition and tongue status of the patient, and the protein associated with the oral cancer comprises at least two types of proteins. 15. The method of claim 14,
Further comprising an optimizer to determine an optimal clinical parameter and an optimal protein using a survival function prior to receiving the parameter. 10. The method of claim 9,
Wherein the clinical parameter comprises the age of the patient. 10. The method of claim 9,
Wherein the plurality of proteins comprises at least p53 and CA9. Receiving a clinical parameter of the patient and a parameter for each of a plurality of proteins associated with oral cancer of the patient,
Calculating a predicted oral cancer incidence prediction score for the patient based on the parameter,
And calculating a probability of occurrence of the oral cancer within a predetermined period based on the predicted oral cancer incidence score.

Images