KR102484530B1 - Method and Apparatus for Classification and Prediction of Pathology Stage using Decision Tree for Treatment of Prostate Cancer - Google Patents

Abstract

The present invention relates to pathological stage prediction for prostate cancer treatment, and relates to a predictive device using a decision tree using medical data of prostate cancer. A pathological stage prediction method for prostate cancer treatment, which is an aspect of the present invention, includes a first step of collecting a plurality of data related to a plurality of users; a second step of extracting first data related to prostate cancer from among the plurality of collected data; a third step of extracting second data corresponding to a preset factor in relation to the pathology stage of the prostate cancer from among the first data; a fourth step of generating a model for predicting the pathological stage of the prostate cancer using at least one of the second data; a fifth step of generating a plurality of models for predicting the pathological stage of the prostate cancer by repeating the first to fourth steps; a sixth step of collecting third data related to a first user among the plurality of users; a seventh step of selecting a first model to be applied to the first user from among the plurality of models using the third data; and an eighth step of predicting the pathological stage of the prostate cancer of the first user by using the first model.

Description

Method and Apparatus for Classification and Prediction of Pathology Stage using Decision Tree for Treatment of Prostate Cancer}

The present invention relates to pathological stage prediction for prostate cancer treatment, and relates to a predictive device using a decision tree using medical data of prostate cancer. Specifically, the present invention provides clinical decision support for prostate cancer treatment, which can increase reliability and reduce surgical costs, since pathological conditions can be known in advance before prostate cancer treatment because clinical decision support for prostate cancer treatment is possible. It relates to an apparatus and method for predicting pathological stages using data mining for

Worldwide, approximately 680,000 men are diagnosed with prostate cancer each year, making prostate cancer the most common disease in men and the second leading cause of cancer deaths in the United States.

Since 2003 in Korea, the incidence rate has been ranked 5th among male cancers since 2003, and prostate cancer is gradually increasing due to the incidence rate of tumors and the aging population.

There are many methods for treating prostate cancer, and radiation therapy is the most effective treatment method for prostate cancer patients.

However, since it is estimated that 40 to 50% of men have clinically potentially extraprostatic disease, accurate pathological stage prediction is the most appropriate because it informs doctors and patients about prostate cancer before surgery. You can choose a surgical method.

As described above, it is effective to treat prostate cancer by predicting the pathological stage and informing doctors and patients before surgery through early diagnosis.

So far, the predictive model used in the United States has been used to predict the pathological stage, but it is not suitable for Koreans, so it is not easy for experts to determine which part is cancer tissue and which part is not cancer tissue. It was difficult to differentiate between hemorrhage and tumor.

Pathological stage prediction technology is being developed in advanced countries through data analysis, and uses the Katan nomogram, which is the most famous among pathological stage prediction technologies.

The Cartan nomogram is described in US Pat. No. 5,993,388 and US Pat. No. 6,409,664.

The Katan nomogram uses a method of predicting the pathological stage by using a method of determining a specific range as a statistical result, but the technology for Koreans is lacking.

In addition, in the conventional method of predicting the pathological stage by setting a specific range as a statistical result, the range of prediction of the pathological stage was divided using Western prostate cancer-related data.

That is, when new data is input, prediction of whether or not there is a pathological stage is provided according to a certain value of the data.

The method of predicting the pathological stage based on the method of setting a specific range based on these statistical results is the simplest and most effective method when predicting the stage of a Westerner, but requires a lot of data and predicts the pathological stage if it is out of a specific interval according to the input data. There are problems that are difficult to do.

As described above, since Korean prostate cancer clinical data may be different from prostate cancer data of other nationalities according to race and region, there is a problem in generating a prediction device, and a solution to this problem is required.

Korean Intellectual Property Office Registration No. 10-1492710 Korea Intellectual Property Office Registration No. 10-1288082

The present invention relates to pathological stage prediction for prostate cancer treatment, and is intended to provide a prediction device and method using a decision tree using medical data of prostate cancer.

Specifically, the present invention provides clinical decision support for prostate cancer treatment, which can increase reliability and reduce surgical costs, since pathological conditions can be known in advance before prostate cancer treatment because clinical decision support for prostate cancer treatment is possible. To provide a pathological stage prediction device and method using data mining for

However, the technical problems to be achieved in the present invention are not limited to the above-mentioned technical problems, and other technical problems not mentioned will be clear to those skilled in the art from the description below. You will be able to understand.

A pathological stage prediction method for prostate cancer treatment, which is an aspect of the present invention for achieving the above technical problem, includes a first step of collecting a plurality of data related to a plurality of users; a second step of extracting first data related to prostate cancer from among the plurality of collected data; a third step of extracting second data corresponding to a preset factor in relation to the pathology stage of the prostate cancer from among the first data; a fourth step of generating a model for predicting the pathological stage of the prostate cancer using at least one of the second data; a fifth step of generating a plurality of models for predicting the pathological stage of the prostate cancer by repeating the first to fourth steps; a sixth step of collecting third data related to a first user among the plurality of users; a seventh step of selecting a first model to be applied to the first user from among the plurality of models using the third data; and an eighth step of predicting the pathological stage of the prostate cancer of the first user by using the first model.

In addition, in the third step, the preset factors include age, body mass index, initial prostate specific antigen, Gleason score, and percentage of positive tumor core) and clinical pathology (Clinical T stage).

Also, the second data may be data representing the state of the pathological stage derived using the preset factor.

In addition, the second data may include non-metastatic disease data and metastatic disease data representing the state of the pathological stage.

In addition, the preset factors include age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, clinical pathology ( Clinical T stage), and the fifth step may generate the plurality of models using at least one of the preset factors, but using different numbers of factors.

In addition, the seventh step may include a 7-1 step of applying a decision tree-based classification algorithm to the plurality of models; a 7-2 step of applying verification data to a plurality of models to which the decision tree-based classification algorithm is applied; and a 7-3 step of selecting, as the first model, a model best mapped to the verification data among result values of a plurality of models to which the verification data is applied.

In addition, the decision tree-based classification algorithm may be an algorithm in which result values continuously vary according to the decision making of the plurality of users.

In addition, the 7-3 step may include an 8-1 step of calculating accuracy, sensitivity and specificity using a confusion matrix; and an 8-2 step of selecting, as the first model, a model showing the highest value by adding up the accuracy, sensitivity, and specificity among the plurality of models.

In addition, the accuracy may be calculated by Equation 1 below, the sensitivity may be calculated by Equation 2 below, and the specificity may be calculated by Equation 3 below.

Equation 1

Equation 2

Equation 3

In Equations 1 to 3, Accuracy means accuracy, Sensitivity means sensitivity, Specificity means specificity, TP (True Positive) means a value accurately predicted for a non-metastatic disease, FN (False Negative) means a value that incorrectly predicts a non-metastatic disease as a metastatic disease, FP (False Positive) means a value that incorrectly predicts a metastatic disease as a non-metastatic disease, and TN (True Negative) means a metastatic disease means the value accurately predicted for .

The method may further include a ninth step of transmitting the predicted pathological stage of prostate cancer of the first user to the first user or displaying it through a display unit.

On the other hand, another aspect of the present invention for achieving the above technical problem, a pathological stage prediction device for prostate cancer treatment, collects a plurality of data related to a plurality of users, and among the collected plurality of data, prostate cancer and a data cleaning unit that extracts related first data; a data item selection unit for extracting second data corresponding to a preset factor in relation to the pathology stage of the prostate cancer from among the first data; A model for predicting the pathological stage of the prostate cancer is generated using at least one of the second data, and the pathological stage of the prostate cancer is predicted by repeating the operations of the data cleaning unit and the data item selection unit. a data model preparation unit for generating a plurality of models for; and collecting third data related to a first user from among the plurality of users, selecting a first model to be applied to the first user from among the plurality of models using the third data, and selecting the first model. and a classification predictor for predicting the pathological stage of the prostate cancer of the first user by using the first user.

In addition, the preset factors include age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, clinical pathology ( Clinical T stage).

Also, the second data may be data representing the state of the pathological stage derived using the preset factor.

In addition, the second data may include non-metastatic disease data and metastatic disease data representing the state of the pathological stage.

In addition, the preset factors include age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, clinical pathology ( Clinical T stage), and the data model preparation unit may generate the plurality of models using at least one of the preset factors, but using a different number of factors.

In addition, the seventh step may include a 7-1 step of applying a decision tree-based classification algorithm to the plurality of models; a 7-2 step of applying verification data to a plurality of models to which the decision tree-based classification algorithm is applied; and a 7-3 step of selecting, as the first model, a model best mapped to the verification data among result values of a plurality of models to which the verification data is applied.

In addition, the decision tree-based classification algorithm may be an algorithm in which result values continuously vary according to the decision making of the plurality of users.

In addition, the classification prediction unit calculates accuracy, sensitivity, and specificity using a confusion matrix, and a model showing the highest value by summing the accuracy, sensitivity, and specificity among the plurality of models is selected as the first model can be selected.

In addition, the accuracy may be calculated by Equation 1 below, the sensitivity may be calculated by Equation 2 below, and the specificity may be calculated by Equation 3 below.

Equation 1

Equation 2

Equation 3

In Equations 1 to 3, Accuracy means accuracy, Sensitivity means sensitivity, Specificity means specificity, TP (True Positive) means a value accurately predicted for a non-metastatic disease, FN (False Negative) means a value that incorrectly predicts a non-metastatic disease as a metastatic disease, FP (False Positive) means a value that incorrectly predicts a metastatic disease as a non-metastatic disease, and TN (True Negative) means a metastatic disease means the value accurately predicted for .

In addition, the wireless communication unit for transmitting the predicted pathological stage of the prostate cancer of the first user to the first user; and a display unit displaying the predicted pathological stage of prostate cancer of the first user. At least one of them may be further included.

The present invention relates to pathological stage prediction for prostate cancer treatment, and can provide a prediction apparatus and method using a decision tree using medical data of prostate cancer.

Specifically, the present invention provides clinical decision support for prostate cancer treatment, which can increase reliability and reduce surgical costs, since pathological conditions can be known in advance before prostate cancer treatment because clinical decision support for prostate cancer treatment is possible. It is possible to provide a pathological stage prediction device and method using data mining for

The present invention can increase the accuracy of predicting the pathological stage of prostate cancer and reduce costs by generating a predictive model using data obtained from medical data of prostate cancer using a decision tree.

In addition, the present invention provides a pathological stage classification prediction method for the treatment of prostate cancer and a classification prediction model based on a decision tree that classifies and predicts the pathological stage in the prediction of the classification, which can increase the accuracy of classification prediction. Support for clinical decision-making is possible because benefits can be fully utilized, unnecessary medical expenses can be reduced, and accurate treatment can be performed.

In addition, since decision-making is accurately supported through prediction of the pathological stage of prostate cancer according to the present invention, the reliability of prostate cancer treatment can be increased.

However, the effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

1 is a block diagram of a pathological stage classification prediction device for prostate cancer treatment proposed by the present invention.
FIG. 2 is a block diagram of a specific structure of the pathological stage classification prediction unit in FIG. 1 .
FIG. 3 is a flowchart illustrating a pathological stage classification prediction method using a decision tree for prostate cancer treatment through the pathological stage classification prediction apparatus described in FIG. 1 .
4 is a flowchart illustrating a process of generating a pathological stage classification prediction model by learning learning data using the method according to FIG. 3 .
5 is an example showing a decision tree determined according to the step of selecting a classification prediction model of FIG. 4 .

For prostate cancer, it is effective to treat prostate cancer by predicting the pathological stage and informing doctors and patients before surgery through early diagnosis.

So far, the predictive model used in the United States has been used to predict the pathological stage, but it is not suitable for Koreans, so it is not easy for experts to determine which part is cancer tissue and which part is not cancer tissue. It is difficult to differentiate between hemorrhage and tumor.

Pathological stage prediction technology is being developed in advanced countries through data analysis, and uses the Katan nomogram, which is the most famous among pathological stage prediction technologies.

The Cartan nomogram is described in US Pat. No. 5,993,388 and US Pat. No. 6,409,664. The Katan nomogram uses a method of predicting the pathological stage by using a method of determining a specific range as a statistical result, but the technology for Koreans is lacking.

In addition, in the conventional method of predicting the pathological stage by setting a specific range as a statistical result, the range of prediction of the pathological stage was divided using Western prostate cancer-related data.

That is, when new data is input, prediction of whether or not there is a pathological stage is provided according to a certain value of the data.

The method of predicting the pathological stage based on the method of setting a specific range based on these statistical results is the simplest and most effective method when predicting the stage of a Westerner, but requires a lot of data and predicts the pathological stage if it is out of a specific interval according to the input data. There are problems that are difficult to do.

That is, since Korean prostate cancer clinical data may differ from prostate cancer data of other nationalities according to race and region, there is a problem in generating a prediction device.

As described above, pathology prediction technology for prostate cancer treatment is very important, but it is difficult to provide a device and method with high accuracy.

In Korea, a method using an overseas method was used, but in order to predict pathology in Koreans, it is necessary to increase the accuracy of tumor prediction through pathology prediction technology suitable for domestic patients and minimize the error rate of prediction to increase the reliability of cancer treatment. .

On the other hand, Patent Publication No. 10-2013-0023235 is a study for predicting pathological stages in Korea, and proposed a method for predicting prognosis based on Carton nomogram and gene data, and using statistical techniques to develop a predictive model. Improved performance.

The prognosis prediction using the Carton nomogram and genetic data according to the above domestic patent predicted recurrence in the next 5 years according to the genetic data, but there is a problem in that the genetic data analysis takes too long.

On the other hand, the decision tree learning method is the most representative learning method of inductive reasoning and is widely used for classification. It has features that are effective in decision-making support.

The decision tree is composed of nodes. The node located at the top is called the root node, and the root node increases the number of nodes by pruning to lower nodes.

At this time, the lower node is pruned back to the lower node.

In addition, a node at which pruning is no longer performed in a lower node is referred to as a leaf node. And, the step from the root node to the end node is called depth.

This decision tree learning technique is recognized as an excellent automatic classification method because it forms a tree-type decision classification model based on collected data and then classifies the transmitted data according to the decision classification model.

Among them, it is very difficult to predict pathology using prostate cancer data of Koreans, so if it is not properly predicted, treatment may become difficult.

In addition, since cost is consumed for the predictive performance of pathological stage classification depending on the prediction method, an efficient prostate cancer pathological stage classification prediction method using data obtained from prostate cancer data of Koreans and a method that can support decision-making are needed. There are issues that need to be considered.

Therefore, in order to solve the above problems, the present invention intends to provide a prediction apparatus and method using a decision tree using medical data of prostate cancer for predicting the pathological stage of prostate cancer treatment.

Specifically, the present invention provides clinical decision support for prostate cancer treatment, which can increase reliability and reduce surgical costs, since pathological conditions can be known in advance before prostate cancer treatment because clinical decision support for prostate cancer treatment is possible. To provide a pathological stage prediction device and method using data mining for

Prior to the detailed description of the present invention, the basic configuration of a pathological stage classification prediction device for prostate cancer treatment applicable to the present invention will be described.

1 is a block diagram of a pathological stage classification prediction device for prostate cancer treatment proposed by the present invention.

Referring to FIG. 1 , an apparatus for predicting pathological stage classification for prostate cancer treatment 100 includes a wireless communication unit 110, an audio/video (A/V) input unit 120, a user input unit 130, and a sensing unit 140. ), an output unit 150, a memory 160, an interface unit 170, a control unit 180, and a power supply unit 190.

However, since the components shown in FIG. 1 are not essential, an apparatus for predicting pathological stage classification for prostate cancer treatment having more or fewer components may be implemented.

Hereinafter, the above components are examined in turn.

The wireless communication unit 110 may include one or more modules enabling wireless communication between a pathological stage classification prediction device for prostate cancer treatment and a wireless communication system or between a device and a network in which the device is located.

For example, the wireless communication unit 110 may include a broadcast receiving module 111, a mobile communication module 112, a wireless Internet module 113, a short-distance communication module 114, a location information module 115, and the like. .

The broadcast reception module 111 receives a broadcast signal and/or broadcast-related information from an external broadcast management server through a broadcast channel.

The broadcast channel may include a satellite channel and a terrestrial channel. The broadcast management server may be a server that generates and transmits broadcast signals and/or broadcast-related information or a server that receives previously generated broadcast signals and/or broadcast-related information and transmits the generated broadcast signals and/or broadcast-related information to a pathological stage classification prediction device for prostate cancer treatment. can mean The broadcast signal includes not only a TV broadcast signal, a radio broadcast signal, and a data broadcast signal, but also a TV broadcast signal or a broadcast signal in which a data broadcast signal is combined with a radio broadcast signal.

The broadcast-related information may refer to information related to a broadcast channel, a broadcast program, or a broadcast service provider. The broadcast-related information may also be provided through a mobile communication network. In this case, it can be received by the mobile communication module 112.

The broadcast related information may exist in various forms. For example, it may exist in the form of an Electronic Program Guide (EPG) of Digital Multimedia Broadcasting (DMB) or an Electronic Service Guide (ESG) of Digital Video Broadcast-Handheld (DVB-H).

The broadcast receiving module 111 includes, for example, Digital Multimedia Broadcasting-Terrestrial (DMB-T), Digital Multimedia Broadcasting-Satellite (DMB-S), Media Forward Link Only (MediaFLO), and Digital Video Broadcasting (DVB-H). -Handheld), ISDB-T (Integrated Services Digital Broadcast-Terrestrial), etc. can be used to receive digital broadcasting signals. Of course, the broadcast receiving module 111 may be configured to be suitable for other broadcasting systems as well as the above-described digital broadcasting system.

Broadcast signals received through the broadcast receiving module 111 and/or broadcast related information may be stored in the memory 160 .

The mobile communication module 112 transmits and receives a radio signal with at least one of a base station, an external device, and a server on a mobile communication network.

It may include various types of data according to text/multimedia message transmission and reception.

The wireless Internet module 113 refers to a module for wireless Internet access, and may be built into or external to a pathological stage classification prediction device for prostate cancer treatment. Wireless Internet technologies include wireless LAN (WLAN) (Wi-Fi), wireless broadband (Wibro), world interoperability for microwave access (Wimax), high speed downlink packet access (HSDPA), and the like.

The short-distance communication module 114 refers to a module for short-distance communication. Short range communication technologies include Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra Wideband (UWB), ZigBee, and Wireless Fidelity (Wi-Fi). can be used

The location information module 115 is a module for acquiring the location of a pathological stage classification prediction device for prostate cancer treatment, and a representative example thereof is a Global Position System (GPS) module.

Referring to FIG. 1 , an audio/video (A/V) input unit 120 is for inputting an audio signal or a video signal, and may include a camera 121 and a microphone 122. The camera 121 processes an image frame such as a still image or a moving image obtained by an image sensor in a photographing mode. The processed image frame may be displayed on the display unit 151 .

An image frame processed by the camera 121 may be stored in the memory 160 or transmitted to the outside through the wireless communication unit 110 . Two or more cameras 121 may be provided according to the use environment.

The microphone 122 receives external sound signals through a microphone in a recording mode, a voice recognition mode, and the like, and processes them into electrical voice data. The processed voice data may be converted into a form transmittable to a mobile communication base station through the mobile communication module 112 and then output. Various noise cancellation algorithms for removing noise generated in the process of receiving an external sound signal may be implemented in the microphone 122 .

The user input unit 130 generates input data for the user to control the operation of the pathological stage classification prediction device for prostate cancer treatment. The user input unit 130 may include a key pad, a dome switch, a touch pad (static pressure/capacity), a jog wheel, a jog switch, and the like.

The sensing unit 140 includes an open/closed state of the pathological stage classification prediction device for prostate cancer treatment, a position of the pathological stage classification prediction device for prostate cancer treatment, whether there is a user contact, and a pathological stage classification prediction device for prostate cancer treatment. Operation of the pathological stage classification prediction device for prostate cancer treatment by detecting the current state of the pathological stage classification prediction device for prostate cancer treatment, such as the orientation of the pathological stage classification prediction device for prostate cancer treatment, acceleration / deceleration of the pathological stage classification prediction device for prostate cancer treatment Generates a sensing signal to control.

The sensing unit 140 may sense whether or not the power supply unit 190 is supplying power and whether or not the interface unit 170 is connected to an external device.

Meanwhile, the sensing unit 140 may include a proximity sensor 141 .

The output unit 150 is for generating an output related to sight, hearing or touch, and includes a display unit 151, a sound output module 152, an alarm unit 153, a haptic module 154, and a projector module ( 155), etc. may be included.

The display unit 151 displays (outputs) information processed by the pathological stage classification prediction device for prostate cancer treatment.

The display unit 151 may include a liquid crystal display (LCD), a thin film transistor-liquid crystal display (TFT LCD), an organic light-emitting diode (OLED), and a flexible display. display) and a 3D display.

Some of these displays may be of a transparent type or a light transmission type so that the outside can be seen through them. This may be referred to as a transparent display, and a representative example of the transparent display is TOLED (Transparent OLED) and the like. A rear structure of the display unit 151 may also be configured as a light transmissive structure. With this structure, the user can see an object located behind the body of the pathological stage classification prediction device for prostate cancer treatment through the area occupied by the display unit 151 of the body of the device for predicting pathological stage classification for treatment of prostate cancer. there is.

Depending on the implementation form of the pathological stage classification prediction device for prostate cancer treatment, two or more display units 151 may exist. For example, in an apparatus for predicting pathological stage classification for prostate cancer treatment, a plurality of display units may be spaced apart or integrally disposed on one surface, or may be respectively disposed on different surfaces.

When the display unit 151 and a sensor for detecting a touch operation (hereinafter referred to as a 'touch sensor') form a mutual layer structure (hereinafter referred to as a 'touch screen'), the display unit 151 may be used in addition to an output device. It can also be used as an input device. The touch sensor may have a form of, for example, a touch film, a touch sheet, or a touch pad.

The touch sensor may be configured to convert a pressure applied to a specific area of the display unit 151 or a change in capacitance generated in a specific area of the display unit 151 into an electrical input signal. The touch sensor may be configured to detect not only the touched position and area, but also the pressure upon touch.

When there is a touch input to the touch sensor, the corresponding signal(s) are sent to the touch controller. The touch controller processes the signal(s) and then transmits corresponding data to the controller 180. Thus, the controller 180 can know which area of the display unit 151 has been touched.

The proximity sensor 141 may be disposed in an inner area of an apparatus for predicting pathological stage classification for prostate cancer treatment covered by the touch screen or near the touch screen. The proximity sensor refers to a sensor that detects the presence or absence of an object approaching a predetermined detection surface or an object existing in the vicinity without mechanical contact by using the force of an electromagnetic field or infrared rays. Proximity sensors have a longer lifespan than contact sensors, and their utilization is also high.

Examples of the proximity sensor include a transmissive photoelectric sensor, a direct reflection photoelectric sensor, a mirror reflection photoelectric sensor, a high frequency oscillation type proximity sensor, a capacitive proximity sensor, a magnetic proximity sensor, an infrared proximity sensor, and the like. When the touch screen is capacitive, the proximity of the pointer is detected by a change in electric field according to the proximity of the pointer. In this case, the touch screen (touch sensor) may be classified as a proximity sensor.

Hereinafter, for convenience of explanation, an act of approaching the touch screen without contacting the pointer to recognize that the pointer is located on the touch screen is referred to as a "proximity touch", and the touch An act of actually touching a pointer on a screen is referred to as "contact touch". The position at which the pointer touches on the touch screen means a position at which the pointer vertically corresponds to the touch screen when the pointer is touched.

The proximity sensor detects a proximity touch and a proximity touch pattern (eg, proximity touch distance, proximity touch direction, proximity touch speed, proximity touch time, proximity touch location, proximity touch movement state, etc.). Information corresponding to the sensed proximity touch operation and proximity touch pattern may be output on the touch screen.

The audio output module 152 may output audio data received from the wireless communication unit 110 or stored in the memory 160 in a recording mode, a voice recognition mode, a broadcast reception mode, or the like. The sound output module 152 also outputs sound signals related to functions performed in the pathological stage classification prediction device for prostate cancer treatment. The sound output module 152 may include a receiver, a speaker, a buzzer, and the like.

The alarm unit 153 outputs a signal for notifying occurrence of an event of the pathological stage classification prediction device for prostate cancer treatment.

The alarm unit 153 may output a signal for notifying the occurrence of an event in a form other than a video signal or an audio signal, for example, vibration.

The video signal or audio signal may be output through the display unit 151 or the audio output module 152, so they 151 and 152 may be classified as part of the alarm unit 153.

A haptic module 154 generates various tactile effects that a user can feel. A representative example of the tactile effect generated by the haptic module 154 is vibration. The strength and pattern of vibration generated by the haptic module 154 can be controlled.

For example, different vibrations may be synthesized and output or sequentially output.

In addition to vibration, the haptic module 154 responds to stimuli such as a pin array vertically moving with respect to the contact skin surface, air blowing force or suction force through a nozzle or suction port, rubbing against the skin surface, contact of an electrode, electrostatic force, and the like. It is possible to generate various tactile effects, such as the effect of heat absorption and the effect of reproducing the feeling of coolness and warmth using an element capable of absorbing heat or generating heat.

The haptic module 154 not only transmits a tactile effect through direct contact, but also can be implemented so that a user can feel a tactile effect through a muscle sense such as a finger or an arm. Two or more haptic modules 154 may be provided according to configuration aspects of the pathological stage classification prediction device for prostate cancer treatment.

The projector module 155 is a component for performing an image project function using a pathological stage classification prediction device for prostate cancer treatment, and displays the image on the display unit 151 according to a control signal from the controller 180. An image identical to or at least partially different from the image displayed on the screen may be displayed on an external screen or wall.

Specifically, the projector module 155 generates an image to be output to the outside using a light source (not shown) that generates light (eg, laser light) for outputting an image to the outside and light generated by the light source. It may include an image generating unit (not shown) for generating an image, and a lens (not shown) for enlarging and outputting an image to the outside at a certain focal distance. In addition, the projector module 155 may include a device (not shown) capable of adjusting an image projection direction by mechanically moving a lens or an entire module.

The projector module 155 may be divided into a cathode ray tube (CRT) module, a liquid crystal display (LCD) module, and a digital light processing (DLP) module according to the type of display device. In particular, the DLP module may be advantageous in reducing the size of the projector module 151 by magnifying and projecting an image generated by reflecting light generated from a light source to a digital micromirror device (DMD) chip.

Preferably, the projector module 155 may be provided in the longitudinal direction on the side, front, or rear surface of the pathological stage classification prediction device for prostate cancer treatment. Of course, it is natural that the projector module 155 may be provided at any position of the pathological stage classification prediction device for prostate cancer treatment, if necessary.

The memory unit 160 may store programs for processing and control by the control unit 180, and functions for temporarily storing input/output data (eg, messages, audio, still images, moving images, etc.) can also be performed. The memory unit 160 may also store the frequency of use of each of the pieces of data. In addition, the memory unit 160 may store data related to vibration and sound of various patterns output when a touch is input on the touch screen.

The memory 160 may be a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only Memory, ROM), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic It may include at least one type of storage medium among a disk and an optical disk. The apparatus for predicting pathological stage classification for treatment of prostate cancer may operate in relation to a web storage that performs a storage function of the memory 160 on the Internet.

The interface unit 170 serves as a passage for all external devices connected to the pathological stage classification prediction device for prostate cancer treatment. The interface unit 170 receives data from an external device, receives power, and transmits it to each component inside the pathological stage classification prediction device for prostate cancer treatment, or inside the pathological stage classification prediction device for prostate cancer treatment. of data to be transmitted to an external device. For example, a wired/wireless headset port, an external charger port, a wired/wireless data port, a memory card port, a port for connecting a device having an identification module, an audio I/O (Input/Output) port, A video I/O (Input/Output) port, an earphone port, and the like may be included in the interface unit 170 .

The identification module is a chip that stores various information for authenticating the right to use the pathological stage classification prediction device for prostate cancer treatment, and includes a user identification module (UIM) and a subscriber identification module (SIM). , Universal Subscriber Identity Module (USIM), and the like. A device equipped with an identification module (hereinafter referred to as 'identification device') may be manufactured in the form of a smart card. Accordingly, the identification device may be connected to a pathological stage classification prediction device for prostate cancer treatment through the port.

The interface unit becomes a path through which power from the cradle is supplied to the pathological stage classification prediction device for prostate cancer treatment when the device for predicting pathological stage classification for prostate cancer treatment is connected to an external cradle, or is provided to the user. It can become a passage through which various command signals input from the cradle are transmitted to the mobile device. Various command signals or the power input from the cradle may be operated as a signal for recognizing that the mobile device is correctly mounted on the cradle.

A controller (180) controls the overall operation of the pathological stage classification prediction device for prostate cancer treatment.

The control unit 180 may further include a pathological stage classification prediction unit 200 .

The pathological stage classification prediction unit 200 includes a data pre-processing unit and a classification prediction unit.

The pathological stage classification prediction unit 200 receives medical data of prostate cancer through a data pre-processing unit and analyzes data related to pathology prediction of prostate cancer.

Specifically, the pathological stage classification prediction unit 200 removes unnecessary data from prostate cancer medical data, selects data related to pathological stage prediction from the medical data from which unnecessary data has been removed, and collects data necessary for classification prediction. By combining several items, various data models related to pathology prediction can be prepared.

A detailed description of the pathological stage classification prediction unit 200 will be described later with reference to FIG. 2 .

The power supply unit 190 receives external power and internal power under the control of the controller 180 and supplies power necessary for the operation of each component.

Various embodiments described herein may be embodied in a recording medium readable by a computer or similar device using, for example, software, hardware, or a combination thereof.

According to hardware implementation, the embodiments described herein include application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), It may be implemented using at least one of processors, controllers, micro-controllers, microprocessors, and electrical units for performing other functions. The described embodiments may be implemented by the controller 180 itself.

According to software implementation, embodiments such as procedures and functions described in this specification may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein. The software code may be implemented as a software application written in any suitable programming language. The software code may be stored in the memory 160 and executed by the controller 180.

Next, FIG. 2 is a block diagram of a specific structure of the pathological stage classification prediction unit in FIG. 1 .

Referring to FIG. 2 , the pathological stage classification prediction unit 200 proposed by the present invention may include a data pre-processing unit 210 and a classification prediction unit 220.

Here, the data pre-processing unit 210 may include a data cleaning unit 211 , a data item selection unit 212 and a data model preparation unit 213 .

The data pre-processing unit 210 provides a function of receiving medical data of prostate cancer and analyzing data related to pathology prediction of prostate cancer.

First, the data cleaning unit 211 provides a function of removing unnecessary data from prostate cancer medical data.

The data cleaning unit 211 removes unnecessary data from the data obtained from the prostate cancer medical data and removes erroneously input data values from the input medical data.

In addition, the data cleaning unit 211 removes data unnecessary for pathological stages, such as patient ID, radiation treatment, and rehabilitation information, which are unnecessary information. Then, the missing data is deleted, and all rows and columns of the corresponding data are deleted.

Next, the data item selection unit 212 provides a function of selecting data related to pathological stage prediction with respect to the data transmitted from the data cleaning unit.

That is, the data item selection unit 212 extracts data related to pathological stage prediction with respect to the data transmitted from the data cleaning unit 211 and transforms the data.

In addition, the data model preparation unit 213 provides a function of preparing various data models related to pathology prediction by combining data necessary for classification prediction into several items.

That is, the data model preparation unit 213 is a step of generating various data models for the data transmitted from the data item selection unit 212 .

In addition, the classification prediction unit 220 provides a function of generating a classification prediction model for the analyzed data.

The classification prediction unit 220 receives data models prepared through data analysis from the data pre-processing unit 210, prepares learning data for the prostate cancer medical data, learns the training data, and applies a decision tree algorithm. After learning, a pathological stage classification prediction model is created.

Based on the structure of the data preprocessing unit 210 and the classification prediction unit 220 of the above-described pathological stage classification prediction unit 200, pathological stage classification prediction using the decision tree for prostate cancer treatment proposed by the present invention explain how

FIG. 3 is a flowchart illustrating a pathological stage classification prediction method using a decision tree for prostate cancer treatment through the pathological stage classification prediction apparatus described in FIG. 1 .

Referring to FIG. 3 , in the pathological stage classification prediction method using a decision tree for prostate cancer treatment proposed by the present invention, first, the data cleaning unit 211 removes unnecessary items from prostate cancer medical data. A step of removing data (S311) is performed.

Thereafter, the data item selection unit 212 selects data related to pathology prediction among the data transmitted from the data cleaning unit 211 (S312).

In step S312, the classification prediction unit 220 selects the basic data and biopsy data from among the patient's basic data (Demographic) and biopsy data and pathology data among the data for predicting the pathological stage, and uses a predictive model. It can be set as an input value to generate, and pathological stage data can be set as an output value.

Here, the basic information data uses the patient's age and body mass index, and the biopsy data uses the initial prostate specific antigen, Gleason score, and percentage of positive tumors. core) and clinical pathology (Clinical T stage) items can be used.

That is, the final input data is a total of six items: age, body mass index, early prostate specific antigen, Gleason score, tumor occupancy rate, and clinical pathology. As an output value, T stage and N stage items of pathology stage among pathology data can be used.

Here, the T stage and the N stage are divided into pT2a, pT2b, pT2c, pT3a, pT3b, pT3c, and pN1. Here, pT2a, pT2b, and pT2c are determined as organ confined diseases (OCD), and pT3a, pT3b, pT3c, and pN1 are determined as non-organ confined diseases (NOCD).

That is, the final output data may be a non-metastatic disease class (0) and a metastatic disease class (1), which are pathological stages.

After step S312, the data model preparation unit 213 generates and prepares data models of various combinations from among the data received from the data item selection unit 212 to be applied to the prediction model (S313).

In addition, the classification prediction unit 220 prepares training data from the data models prepared by the data pre-processing unit 210 and medical data of prostate cancer, and generates prediction models according to the data model (S314).

In step S314, the data model preparation unit 213 may prepare various models because there may be unnecessary items in the input data for pathological stage prediction, and if there are unnecessary data, classification prediction accuracy may be lowered.

A total of 5 types of data models between input data and output data can be prepared as shown in Table 1.

In Table 1, each data model has 3 to 5 input values and 1 output value.

In addition, data model 1 consists of 3 input values, data model 2 consists of 4 input values, data models 3 and 4 consist of 5 input values, and data model 5 consists of 6 input values.

Thereafter, a step of selecting a prediction model having the highest accuracy from among prediction models generated according to the data model (S315) is performed.

In the Internet game addiction classification prediction method using the decision tree for Internet game addiction treatment, the data cleaning unit 211 removes unnecessary data from the Internet game lifestyle data of Internet game addiction, starting from step S100.

After step S100, the data item selector 212 selects data related to Internet game addiction prediction from among the data transmitted from the data cleaner 211 (S200).

In step S200, the classification prediction unit 220 inputs the patient's basic data (Demographic), internet game usage time, internet game cost, and offline meeting participation data among data for predicting internet game addiction to generate a predictive model. value, and Internet game addiction data can be set as an output value.

Through the process described in FIG. 3, FIG. 4 is a flowchart for explaining a process of generating a classification prediction model for a pathological stage by learning learning data according to an embodiment of the present invention.

Since steps S311 to S315 in FIG. 4 correspond to steps S311 to S315 in FIG. 3 , they are omitted for brevity of the specification.

Steps of generating a predictive model by learning learning data in FIG. 3 are steps S314 and S315.

Referring to FIG. 4, after the steps of learning learning data to generate a predictive model (S314, S315), a decision tree-based classification algorithm is applied to generate a predictive model by learning the learning data according to data models. The step of learning by doing (S510) proceeds.

In step S510, in order to generate a classification prediction model through the classification prediction unit 220, after receiving the five data models from the data preprocessing unit 210, the five data models in Table 1 may be prepared.

Here, a decision tree-based learning model is generated using 70% of the entire learning data, and the remaining 30% of the data is transmitted to step S520 to be used for verification.

Here, in order to create a decision tree-based learning model, the five data models can be used as input values, and the output data items can be used as output values for learning.

At this time, five prediction models may be generated using the five data models.

Thereafter, a step of applying verification data to the predictive models to which the generated decision tree-based classification algorithm is applied (S520) proceeds.

In step S520, the classification prediction unit 220 applies verification data to the predictive models, and the accuracy, sensitivity, and specificity of the five generated predictive models can be evaluated.

Here, accuracy, sensitivity, and specificity use a confusion matrix, and parameters shown in Table 2 below may be applied.

In Table 2, TP (True Positive) means a value that is accurately predicted for a non-metastatic disease, FN (False Negative) means a value that incorrectly predicts a non-metastatic disease as a metastatic disease, and FP (False Positive) means the value of incorrectly predicting metastatic disease as non-metastatic disease, and TN (True Negative) means the value of accurately predicting metastatic disease.

Here, accuracy, sensitivity, and specificity can be calculated through the following equations.

Calculation results for the five prediction models using the above equation are transmitted to step 530.

After the step S520, a step of selecting a classification prediction model for the pathological stage of prostate cancer by selecting a prediction model having the highest accuracy after applying the verification data (S530) is sequentially performed.

In step S530, in order to select a prediction model with the highest accuracy in the classification prediction unit 220 and select a classification prediction model for the pathological stage of prostate cancer, the highest accuracy value among the calculation result values transmitted from step S520 is selected. can decide

According to an embodiment, the calculation result value transmitted from step S520 may be shown in Table 3 below.

Among the calculation results of Table 3, the classification prediction model with the highest accuracy, sensitivity, and specificity is the fourth classification prediction model 4.

5 is an exemplary view showing a decision tree according to the result of step 530 of FIG. 4 .

In the step S530, the classification prediction model 4, which is the highest among the calculation results, is displayed. When new data is received, predictions can be made for non-metastatic and metastatic diseases using the classification prediction model 4.

For example, if the tumor occupancy rate is 40, the clinical pathology is T2c, and the age is 64 years old, the root node, node 1, node 4, node 10, and final node 15 can be predicted as a non-metastatic disease in the order.

Accordingly, when the method and apparatus for predicting pathological stages for treatment of prostate cancer of the present invention are applied, the apparatus and method for predicting using a decision tree using medical data of prostate cancer can be provided.

Specifically, the present invention provides clinical decision support for prostate cancer treatment, which can increase reliability and reduce surgical costs, since pathological conditions can be known in advance before prostate cancer treatment because clinical decision support for prostate cancer treatment is possible. It is possible to provide a pathological stage prediction device and method using data mining for

The present invention can increase the accuracy of predicting the pathological stage of prostate cancer and reduce costs by generating a predictive model using data obtained from medical data of prostate cancer using a decision tree.

In addition, the present invention provides a pathological stage classification prediction method for the treatment of prostate cancer and a classification prediction model based on a decision tree that classifies and predicts the pathological stage in the prediction of the classification, which can increase the accuracy of classification prediction. Support for clinical decision-making is possible because benefits can be fully utilized, unnecessary medical expenses can be reduced, and accurate treatment can be performed.

In addition, since decision-making is accurately supported through prediction of the pathological stage of prostate cancer according to the present invention, the reliability of prostate cancer treatment can be increased.

The above-described embodiments of the present invention may be implemented through various means. For example, embodiments of the present invention may be implemented by hardware, firmware, software, or a combination thereof.

In the case of hardware implementation, the method according to the embodiments of the present invention includes one or more ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices) , Field Programmable Gate Arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, etc.

In the case of implementation by firmware or software, the method according to the embodiments of the present invention may be implemented in the form of a module, procedure, or function that performs the functions or operations described above. The software codes may be stored in a memory unit and driven by a processor. The memory unit may be located inside or outside the processor and exchange data with the processor by various means known in the art.

Detailed descriptions of the preferred embodiments of the present invention disclosed as described above are provided to enable those skilled in the art to implement and practice the present invention. Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the scope of the present invention. For example, those skilled in the art can use each configuration described in the above-described embodiments in a manner of combining with each other. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The present invention may be embodied in other specific forms without departing from the spirit and essential characteristics of the present invention. Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered as illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention. The invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. In addition, claims that do not have an explicit citation relationship in the claims may be combined to form an embodiment or may be included as new claims by amendment after filing.

Claims (20)

A first step of collecting a plurality of data related to a plurality of users;
a second step of extracting first data related to prostate cancer from among the plurality of collected data;
a third step of extracting second data corresponding to a preset factor in relation to the pathology stage of the prostate cancer from among the first data;
a fourth step of generating a model for predicting the pathological stage of the prostate cancer using at least one of the second data;
a fifth step of generating a plurality of models for predicting the pathological stage of the prostate cancer by repeating the first to fourth steps;
a sixth step of collecting third data related to a first user among the plurality of users;
a seventh step of selecting a first model to be applied to the first user from among the plurality of models using the third data; and
An eighth step of predicting the pathological stage of the prostate cancer of the first user using the first model;

The seventh step,
a 7-1 step of applying a decision tree-based classification algorithm to the plurality of models;
a 7-2 step of applying verification data to a plurality of models to which the decision tree-based classification algorithm is applied; and
A 7-3 step of selecting, as the first model, a model best mapped to the verification data among result values of a plurality of models to which the verification data is applied;

In the 7-3 step,
Step 8-1 of calculating accuracy, sensitivity, and specificity using a confusion matrix; and
An 8-2 step of selecting a model showing the highest value by adding up the accuracy, sensitivity, and specificity among the plurality of models as the first model;

The accuracy is calculated by Equation 1 below, the sensitivity is calculated by Equation 2 below, and the specificity is calculated by Equation 3 below. Stage prediction method.

Equation 1

Equation 2

Equation 3

In Equations 1 to 3, Accuracy means accuracy, Sensitivity means sensitivity, Specificity means specificity, TP (True Positive) means a value accurately predicted for a non-metastatic disease, FN (False Negative) means a value that incorrectly predicts a non-metastatic disease as a metastatic disease, FP (False Positive) means a value that incorrectly predicts a metastatic disease as a non-metastatic disease, and TN (True Negative) means a metastatic disease means the value accurately predicted for .
According to claim 1,
in the third step
The preset factor is,
Including age, Body Mass Index, initial Prostate Specific Antigen, Gleason score, Percentage of positive core, and Clinical T stage Pathological stage prediction method for prostate cancer treatment characterized by. According to claim 2,
The second data,
The pathological stage prediction method for prostate cancer treatment, characterized in that the data representing the state of the pathological stage derived using the preset factor. According to claim 3,
The second data,
A pathological stage prediction method for treating prostate cancer, comprising non-metastatic disease data and metastatic disease data representing the state of the pathological stage. According to claim 1,
The preset factor is,
Including age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, clinical pathology,
The fifth step is
Pathological stage prediction method for prostate cancer treatment, characterized in that for generating the plurality of models using at least one of the preset factors, but using a different number of factors. delete According to claim 1,
The pathological stage prediction method for prostate cancer treatment, characterized in that the decision tree-based classification algorithm is an algorithm in which result values continuously vary according to the decision making of the plurality of users.
delete delete According to claim 1,
The pathological stage prediction method for prostate cancer treatment, further comprising a ninth step of transmitting the predicted pathological stage of the first user to the first user or displaying the predicted pathological stage of prostate cancer through a display unit. ◈Claim 11 was abandoned when the registration fee was paid.◈ a data cleaning unit that collects a plurality of data related to a plurality of users and extracts first data related to prostate cancer from among the collected data;
a data item selection unit for extracting second data corresponding to a preset factor in relation to the pathology stage of the prostate cancer from among the first data;
A model for predicting the pathological stage of the prostate cancer is generated using at least one of the second data, and the pathological stage of the prostate cancer is predicted by repeating the operations of the data cleaning unit and the data item selection unit. a data model preparation unit for generating a plurality of models for; and
Third data related to a first user among the plurality of users is collected, and a first model to be applied to the first user is selected from among the plurality of models using the third data, and the first model is used. and a classification prediction unit for predicting the pathological stage of the prostate cancer of the first user by doing so;

The classification prediction unit,
Applying a decision tree-based classification algorithm to the plurality of models,
Applying verification data to a plurality of models to which the decision tree-based classification algorithm is applied,
Selecting a model best mapped to the verification data among result values of a plurality of models to which the verification data is applied as the first model;

The classification prediction unit,
Calculate accuracy, sensitivity and specificity using a confusion matrix,
Among the plurality of models, a model showing the highest value by summing the accuracy, sensitivity, and specificity is selected as the first model,

The accuracy is calculated by Equation 1 below, the sensitivity is calculated by Equation 2 below, and the specificity is calculated by Equation 3 below. stage prediction device.

Equation 1

Equation 2

Equation 3

In Equations 1 to 3, Accuracy means accuracy, Sensitivity means sensitivity, Specificity means specificity, TP (True Positive) means a value accurately predicted for a non-metastatic disease, FN (False Negative) means a value that incorrectly predicts a non-metastatic disease as a metastatic disease, FP (False Positive) means a value that incorrectly predicts a metastatic disease as a non-metastatic disease, and TN (True Negative) means a metastatic disease means the value accurately predicted for .
◈Claim 12 was abandoned when the registration fee was paid.◈ According to claim 11,
The preset factor is,
Characterized by age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, and clinical pathology A pathological stage prediction device for the treatment of prostate cancer. ◈Claim 13 was abandoned when the registration fee was paid.◈ According to claim 12,
The second data,
Pathological stage prediction device for prostate cancer treatment, characterized in that the data representing the state of the pathological stage derived using the preset factor. ◈Claim 14 was abandoned when the registration fee was paid.◈ According to claim 13,
The second data,
A pathological stage predictive device for treating prostate cancer, characterized in that it comprises non-metastatic disease data and metastatic disease data representing the state of the pathological stage. ◈Claim 15 was abandoned when the registration fee was paid.◈ According to claim 11,
The preset factor is,
Including age, body mass index, initial prostate specific antigen, Gleason score, percentage of positive core, clinical pathology,
The data model preparation unit,
Pathological stage prediction apparatus for prostate cancer treatment, characterized in that for generating the plurality of models using at least one of the preset factors, but using a different number of factors. delete ◈Claim 17 was abandoned when the registration fee was paid.◈ According to claim 11,
The pathological stage prediction device for prostate cancer treatment, characterized in that the decision tree-based classification algorithm is an algorithm in which result values continuously vary according to the decision making of the plurality of users.
delete delete ◈Claim 20 was abandoned when the registration fee was paid.◈ According to claim 11,
a wireless communication unit which transmits the predicted pathological stage of prostate cancer of the first user to the first user; and a display unit displaying the predicted pathological stage of prostate cancer of the first user. Pathological stage prediction device for prostate cancer treatment, characterized in that it further comprises at least one of.

Images