Artificial intelligence system

Abstract

To disclose an artificial intelligence system and prediction method of the same.SOLUTION: An artificial intelligence system includes: a data module that receives personal health record (PHR) data on a patient, and performs preprocessing of the PHR data to generate final data; a model module receives the final data and an instruction, and derives a pre-prediction of the patient via at least one artificial intelligence model selected according to the instruction; and a Reliable and Interpretable AI System (RIAS) that operates separately and independently the data module and the model module, receives the pre-prediction and the final data, and converts the pre-prediction into a final prediction.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an artificial intelligence system and prediction methods thereof. Specifically, the present invention relates to an artificial intelligence system that can use various data and artificial intelligence models and prediction methods thereof.

Background Art

[0002] PHR (Personal Health Record) data can cover various information as records of an individual's health. Using such PHR data to derive medical judgments such as the current patient's mortality rate is an expansion from the field of physicians to the field of artificial intelligence. Medical diagnosis artificial intelligence usually takes the method of developing one artificial intelligence model for a specific type of data to draw conclusions. That is, it is a method of constructing an independent pipeline according to the data and constructing a new pipeline when the artificial intelligence model is updated.

[0003] Such a method has the inefficiency of discarding the overlapping parts for each pipeline without utilizing them even if they exist, and having to use a new pipeline from the beginning. Therefore, research has constantly existed to eliminate such inefficiencies.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object of the present invention is to provide an artificial intelligence system that does not need to be modified even when data and an artificial intelligence model change.

[0006] Another object of the present invention is to provide a prediction method for an artificial intelligence system that does not need to be modified even when data and an artificial intelligence model change.

[0007] Another object of the present invention is to provide a computer program recorded on a recording medium that causes a prediction method for an artificial intelligence system that does not need to be modified even when data and an artificial intelligence model change to be executed.

[0008] The object of the present invention is not limited to the objects mentioned above, and other objects and advantages of the present invention that have not been mentioned will be understood from the following description and will be more clearly understood from the embodiments of the present invention. Furthermore, it will be easily understood that the objects and advantages of the present invention can be realized by the means shown in the claims and combinations thereof.

Means for Solving the Problems

[0009] An artificial intelligence system according to some embodiments of the present invention for solving the above problems includes a data module that receives a patient's PHR (Personal Health Record) data and executes preprocessing of the PHR data to generate final data, receives the final data and instructions, and derives a pre-prediction of the patient through at least one artificial intelligence model selected according to the instructions, and a model module, and a RIAS (Reliable and Interpretable AI System) that is separated from the data module and the model module and operates independently, receives the pre-prediction and the final data, and converts the pre-prediction into a final prediction.

[0010] Also, the patient can include a myocardial infarction patient.

[0011] In addition, the final prediction can include the mortality rate of the patient after a specific period.

[0012] In addition, the final prediction can include the proportion of factors that positively contribute to the mortality rate and the proportion of factors that negatively contribute to the mortality rate.

[0013] In addition, the PHR data can include the mortality rate of the patient after acute myocardial infarction or the reverse remodeling data of the patient after acute myocardial infarction.

[0014] In addition, the PHR data can further include data obtained by modifying or manipulating the mortality rate or the remodeling data.

[0015] In addition, the data module can include a preprocessing module that receives instructions including a preset data frame, preprocesses the PHR data according to the data frame to generate preprocessed data, and an imputation module that processes missing values of the preprocessed data to generate final data.

[0016] In addition, the imputation module can receive the instructions and process the missing values according to the data frame.

[0017] In addition, the imputation module can process the missing values through at least one of mean substitution, median substitution, mode substitution, regression substitution, K-nearest neighbor substitution, and multiple substitution.

[0018] In addition, the instructions can include information for a predetermined artificial intelligence model.

[0019] In addition, the at least one artificial intelligence model can be classified into one type if they have common functions with each other, and the model module can objectify the common functions according to the type.

[0020] Also, the type can include at least one of GBDT (Gradient Boosted Decision Trees), pytorch tabula, and Saint.

[0021] Also, the RIAS can include at least one of a training module that trains the artificial intelligence model, tunes hyperparameters of the artificial intelligence model, and evaluates it, a calibration module that corrects the pre-prediction to derive a final prediction, a local explanation module that executes a local explanations method for the contribution degree of each feature with respect to the final data, and a counterfactual explanation module that executes a counterfactual explanations method for the final data.

[0022] Also, the pre-prediction is a prediction value between 0 and 1, and the calibration module can adjust the prediction value so as to mimic the occurrence probability of an actual event to derive a final prediction.

[0023] An artificial intelligence system according to some embodiments of the present invention for solving the above problems includes a memory and a processor that executes operations in conjunction with the memory, and the processor preprocesses patient PHR data to generate final data, inputs the final data into a selected artificial intelligence model to derive a pre-prediction, and corrects the pre-prediction to derive a final prediction.

[0024] A prediction method of an artificial intelligence system according to some embodiments of the present invention for solving the above other problems is a prediction method of an artificial intelligence system, including that the artificial intelligence system preprocesses patient PHR data to generate final data, inputs the final data into a selected artificial intelligence model to derive a pre-prediction, and corrects the pre-prediction to derive a final prediction.

[0025] In addition, generating the final data can include changing the PHR data into a preset data frame to generate preprocessed data, and processing missing values in the preprocessed data to generate the final data.

[0026] Furthermore, it further includes the artificial intelligence system receiving an instruction, and the instruction can include the above-mentioned preset data frame.

[0027] In addition, deriving the pre-prediction can include selecting an artificial intelligence model according to the instruction, and deriving the pre-prediction through the selected artificial intelligence model.

Advantages of the Invention

[0028] For the artificial intelligence system and its prediction method of the present invention, it is not necessary to modify the pipeline even if the input data changes.

[0029] Moreover, even when the artificial intelligence module changes, it is not necessary to modify the pipeline, and an efficient pipeline system can be implemented.

[0030] Furthermore, since the degree of contribution to the factors contributing to the prediction can be expressed, it is possible to facilitate the understanding and explanation of the prediction for patients.

[0031] In addition to the above-mentioned content, the specific effects of the present invention will be described together while explaining the specific matters for implementing the following invention.

Brief Description of the Drawings

[0032]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

DETAILED DESCRIPTION OF THE INVENTION

[0033] The terms or words used in this specification and the claims shall not be construed as being limited to their general or dictionary meanings. In accordance with the principle that the inventor can define the concept of a term or word in order to best explain his or her own invention, they should be construed as meanings and concepts consistent with the technical idea of the present invention. Also, the embodiments described in this specification and the configurations shown in the drawings are merely one embodiment in which the present invention is realized and do not represent all of the technical ideas of the present invention. Therefore, it should be understood that there may be various equivalents, modifications, and applicable examples that can replace them at the time of this application.

[0034] The terms such as first, second, A, B, etc. used in this specification and the claims can be used to describe various components, but the said components should not be limited by the said terms. The said terms are only used for the purpose of distinguishing one component from another. For example, without departing from the scope of the rights of the present invention, the first component can be named the second component, and similarly, the second component can also be named the first component. The term "and / or" includes a combination of a plurality of relatedly described items or any one of a plurality of relatedly described items.

[0035] The terms used in this specification and the claims are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. It should be understood that terms such as "including" or "having" in this application do not preclude the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification in advance.

[0036] Unless otherwise defined, all terms used in this specification, including technical or scientific terms, have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0037] Terms such as those defined in commonly used dictionaries should be construed to have a meaning consistent with their meaning in the context of the relevant art and should not be construed in an idealized or overly formal sense unless clearly defined in this application.

[0038] Furthermore, each configuration, process, step, or method included in each embodiment of the present invention can be shared within a range that is not technically mutually contradictory.

[0039] Hereinafter, an artificial intelligence system according to some embodiments of the present invention will be described with reference to FIGS. 1 to 9.

[0040] FIG. 1 is a block diagram for explaining an artificial intelligence system according to some embodiments of the present invention.

[0041] Referring to FIG. 1, an artificial intelligence system (10) according to some embodiments of the present invention can receive PHR data (DATA_PHR) and instructions (Inst). The artificial intelligence system (10) can internally generate a prediction (Prd) via the PHR data (DATA_PHR) and the instructions (Inst). The artificial intelligence system (10) can transmit the generated prediction (Prd) externally.

[0042] FIG. 2 is a block diagram for explaining the PHR data of FIG. 1.

[0043] Referring to FIG. 2, the PHR data (DATA_PHR) can be data for various personal health records of a patient. The PHR data (DATA_PHR) can be, for example, personal health record data of a myocardial infarction patient. However, the present embodiment is not limited thereto. The PHR data (DATA_PHR) can include, for example, at least one of first data (d1) for the mortality rate within 6 months from the onset of acute myocardial infarction (AMI), second data (d2) for reverse remodeling within 12 months from the onset of myocardial infarction, third data (d3) for the mortality rate within 5 years from the onset of myocardial infarction, and fourth data (d4) for the data obtained by performing a change operation on the above data.

[0044] At this time, reverse remodeling is a process in which the heart recovers from the damage caused by myocardial infarction, and thereby, it can be shown that the size, shape, function, etc. of the heart try to return to normal again. That is, the reverse remodeling process can include a medical strategy for minimizing damage to heart tissue and optimizing heart function to improve the quality of life of the patient.

[0045] The periods mentioned in the above first data (d1), second data (d2), and third data (d3) are all exemplary periods, and different periods can be adopted as needed and according to the purpose.

[0046] FIG. 3 is a block diagram for explaining the instruction of FIG. 1.

[0047] Referring to FIGS. 1 to 3, the instruction (Inst) can include a plurality of information necessary for the operation of the artificial intelligence system (10) according to the present embodiment.

[0048] The instruction (Inst) can include, for example, at least one of first information (I1) for a data frame, second information (I2) for an artificial intelligence model, and third information (I3) for evaluation parameters for the evaluation of the artificial intelligence model.

[0049] At this time, the first information (I1) can be information on how to preprocess the data frame of the PHR data (DATA_PHR). That is, since the PHR data (DATA_PHR) can exist in various forms, if it is not converted into a single data frame, there may be inefficiencies where a new pipeline system has to be newly configured. The artificial intelligence system (10) according to this embodiment can convert various PHR data (DATA_PHR) into a preset data frame included in the instruction (Inst) and transmit it to the artificial intelligence model, so there is no need to newly configure the pipeline system and the efficiency can be maximized.

[0050] The second information (I2) can be designation information for the artificial intelligence model. That is, the second information (I2) can include information on what artificial intelligence model to use to derive the prediction (Prd). Existing prediction systems may have inefficiencies where a new pipeline system has to be newly configured in order to newly introduce the latest artificial intelligence model. The artificial intelligence system (10) according to this embodiment can select the artificial intelligence model to the preset model included in the instruction (Inst) and can use the functions overlapping with other artificial intelligence models as they are, so the efficiency can be improved.

[0051] The third information (I3) may be necessary when presetting the parameters required for evaluating the artificial intelligence model. At this time, the third information (I3) can be preset so that it can be applied regardless of what kind of artificial intelligence model is selected from various artificial intelligence models.

[0052] FIG. 4 is an illustrative diagram showing a prediction when the beta-blocker in FIG. 1 is not administered, and FIG. 5 is an illustrative diagram showing a prediction when the beta-blocker in FIG. 1 is administered.

[0053] Referring to FIGS. 1, 4, and 5, the prediction (Prd) can indicate a probabilistic value such as the mortality rate of a patient 12 months after the onset of myocardial infarction. FIG. 4 shows, as an illustration of the prediction (Prd), that the mortality rate of a patient in a situation where a beta-blocker was not administered after the onset of myocardial infarction is 95%.

[0054] In FIG. 4, the red portion can mean the state of the patient that contributes to the death of the patient, and the blue portion can mean the state of the patient that contributes to the survival of the patient. The prediction (Prd) of the artificial intelligence system (10) of this embodiment can provide the degree of contribution to the current prediction (Prd) value in this way, so that analysis and understanding can be facilitated.

[0055] Referring to FIG. 5, it can be seen that the mortality rate decreases to 38% when a beta-blocker is administered. That is, medical staff in charge of treating the patient can receive assistance in diagnosis and prescription by comparing the prediction (Prd) in FIG. 4 with the prediction (Prd) in FIG. 5.

[0056] FIG. 6 is a block diagram for explaining the configuration of the artificial intelligence system of FIG. 1.

[0057] Referring to FIG. 6, the artificial intelligence system (10) can include a data module (100), a RIAS (Reliable and Interpretable AI System) (200), and a model module (300).

[0058] The data module (100) can receive PHR data (DATA_PHR) and instructions (Inst) and perform preprocessing. The data module (100) can preprocess the PHR data (DATA_PHR) according to the instructions (Inst) to generate final data (Data_F). The data module (100) can transmit the final data (Data_F) to the RIAS (200) and the model module (300).

[0059] The model module (300) can receive an instruction (Inst) and final data (Data_F), learn an artificial intelligence model, and execute a prediction. The model module (300) can generate a pre-prediction (Prd_P) through the artificial intelligence model. The model module (300) can transmit the pre-prediction (Prd_P) to the RIAS (200). At this time, the pre-prediction (Prd_P) can be a value between 0 and 1.

[0060] The RIAS (200) can receive the instruction (Inst) and the final data (Data_F). The RIAS (200) can also receive the pre-prediction (Prd_P) from the model module (300). The RIAS (200) can derive the pre-prediction (Prd_P) into a final prediction (Prd) using the final data (Data_F) and the instruction (Inst). The RIAS (200) can evaluate and tune the pre-prediction (Prd_P).

[0061] At this time, the data module (100), the model module (300), and the RIAS (200) can operate in a pipeline system regardless of the type of data and the type of model. That is, a general artificial intelligence system generates a prediction using an artificial intelligence model determined according to the type of data. When the data changes or the artificial intelligence model changes, a new pipeline system has to be implemented again instead of the existing pipeline system.

[0062] In contrast, for the artificial intelligence system according to this embodiment, the data module (100) converts the changed data into the same data frame and provides it. The model module (300) objectifies it so that it can utilize common functions among various artificial intelligence models. The RIAS (200) can derive the final prediction (Prd) without a new pipeline system configuration independently of the data and the artificial intelligence model.

[0063] FIG. 7 is a block diagram for explaining the structure of the data module of FIG. 6.

[0064] Referring to FIG. 7, the data module (100) can include a preprocessing module (110) and an imputation module (120).

[0065] The preprocessing module (110) can receive PHR data (DATA_PHR) and instructions (Inst). At this time, the instructions (Inst) can include information on the data frame. The preprocessing module (110) can preprocess the PHR data (DATA_PHR) via the information on the data frame included in the instructions (Inst). Thereby, the preprocessing module (110) can generate preprocessed data (Data_P).

[0066] The imputation module (120) can receive the preprocessed data (Data_P) from the preprocessing module (110). Furthermore, the imputation module (120) can receive the instructions (Inst) and determine a method for handling missing values included in the instructions (Inst).

[0067] The imputation module (120) can handle missing values in at least one of, for example, mean substitution, median substitution, mode substitution, regression substitution, k-nearest neighbor substitution, and multiple substitution. At this time, the method selected by the imputation module (120) may be a preset method or a method specified by the instructions (Inst). The imputation module (120) can remove the missing values in the preprocessed data (Data_P) to generate final data (Data_F).

[0068] FIG. 8 is a block diagram for explaining the structure of RIAS of FIG. 6.

[0069] Referring to FIG. 8, the RIAS (200) can include at least one of a training module (210), a calibration module (220), a regional explanation module (230), and an alternative factual explanation module (240).

[0070] The training module (210) can inherit the model of the pre-prediction (Prd_P) received by the RIAS (200) from the model module (300), and tune, learn, and evaluate the hyperparameters of the model. The training module (210) can still operate as it is even if the artificial intelligence model selected by the model module (300) changes.

[0071] The calibration module (220) can convert the pre-prediction (Prd_P) into the final prediction (Prd). The calibration module (220) can process the pre-prediction (Prd_P) to mimic the likelihood of the actual event occurring and convert it into the prediction (Prd) so that the user can accept it with the predicted probability of the actual event occurring.

[0072] The regional explanation module (230) can inform how much the features of the final data (Data_F) utilized in the prediction (Prd) contributed to the prediction (Prd). That is, the regional explanation module (230) can generate the contribution parts of FIGS. 4 and 5.

[0073] The alternative factual explanation module (240) can provide insights on what should be done to obtain the result desired by the user based on the pre-prediction (Prd_P). Therefore, the prediction (Prd) can include the said insights. For example, the alternative factual explanation module (240) can provide insights to confirm the mortality rate with or without beta-blocker administration in FIGS. 4 and 5 and propose beta-blocker administration.

[0074] FIG. 9 is a block diagram for explaining the model module of FIG. 6.

[0075] Referring to FIG. 9, the model module (300) can select any one of various artificial intelligence models to perform learning and inference. Each model can be classified into the same type among models having common functions with each other. Specifically, the artificial intelligence models can include a first type (310), a second type (320), and a third type (330). In FIG. 9, the number of types is shown as three, but this embodiment is not limited thereto.

[0076] The first type (310), the second type (320), and the third type (330) can be, for example, GBDT (Gradient Boosted Decision Trees), Pytorch tabula, and Saint, respectively. However, this is only an example, and this embodiment is not limited thereto.

[0077] Therefore, the first model (311), the second model (312), and the third model (313) belonging to the first type (310) can be XGBoost, LightGBM, and CatBoost, respectively. However, this embodiment is not limited thereto. The fourth model (321), the fifth model (322), the sixth model (323), the seventh model (324), and the eighth model (325) belonging to the second type (320) can be FTTransformer, TabTransformer, TabNet, AutoInt, and MLP, respectively. However, this embodiment is not limited thereto.

[0078] In this way, the model module (300) can specify a common object for the same type of artificial intelligence models having the same function so that the pipelines can be shared as much as possible.

[0079] This embodiment operates independently of data or an artificial intelligence model and can use the same flow regardless of which data or model is used. As a result, different from existing artificial intelligence systems, predictions can be derived much more efficiently.

[0080] Hereinafter, with reference to FIGS. 6 to 8 and FIGS. 10 to 12, a prediction method of an artificial intelligence system according to some embodiments of the present invention will be described. Parts overlapping with the above-described embodiments will be simplified or omitted.

[0081] The prediction method of an artificial intelligence system according to some embodiments of the present invention can be executed by a computer program stored in a recording medium. In this case, it can operate in an organic combination with the artificial intelligence system. However, this embodiment is not limited thereto.

[0082] FIG. 10 is a flowchart for explaining a prediction method of an artificial intelligence system according to some embodiments of the present invention, and FIG. 11 is a flowchart for explaining in detail the final data generation stage of FIG. 10. FIG. 12 is a flowchart for explaining in detail the pre-prediction derivation stage of FIG. 10.

[0083] Referring to FIG. 10, PHR data is preprocessed to generate final data (S100).

[0084] Referring in detail to FIG. 11, the PHR data is changed to a preset data frame to generate preprocessed data (S110).

[0085] Specifically, referring to FIG. 7, the preprocessing module (110) can receive PHR data (DATA_PHR) and instructions (Inst). At this time, the instructions (Inst) can include information about the data frame. The preprocessing module (110) can preprocess the PHR data (DATA_PHR) through the information about the data frame included in the instructions (Inst). Thereby, the preprocessing module (110) can generate preprocessed data (Data_P).

[0086] Referring to FIG. 11 again, missing values in the preprocessed data are processed to generate final data (S120).

[0087] Specifically, referring to FIG. 7, the imputation module (120) can receive the preprocessed data (Data_P) from the preprocessing module (110). Further, the imputation module (120) can receive the instructions (Inst) and determine the method for missing value processing included in the instructions (Inst). The imputation module (120) can remove the missing values in the preprocessed data (Data_P) to generate final data (Data_F).

[0088] Referring to FIG. 10 again, the final data is input into the selected artificial intelligence model to derive a pre-prediction (S200).

[0089] Specifically, referring to FIG. 12, an artificial intelligence model is selected by the instructions (S210), and a pre-prediction is derived through the selected artificial intelligence model (S220).

[0090] Specifically, referring to FIG. 6, the model module (300) can receive an instruction (Inst) and final data (Data_F), learn an artificial intelligence model, and execute a prediction. The model module (300) can generate a pre-prediction (Prd_P) via the artificial intelligence model. The model module (300) can transmit the pre-prediction (Prd_P) to the RIAS (200).

[0091] Referring to FIG. 10 again, the pre-prediction is corrected to derive a final prediction (S300).

[0092] Specifically, referring to FIG. 8, the calibration module (220) can convert the pre-prediction (Prd_P) into a final prediction (Prd). The calibration module (220) can process the pre-prediction (Prd_P) to mimic the occurrence probability of an actual event and convert it into a prediction (Prd) that the user can accept with the occurrence prediction probability of the actual event.

[0093] FIG. 13 is a drawing for explaining the hardware configuration of an artificial intelligence system according to some embodiments of the present invention.

[0094] The artificial intelligence system according to some embodiments of the present invention can be implemented in an electronic device (1000). The electronic device (1000) can include a processor (1010), input / output devices (1020, I / O), a memory (1030, memory), an interface (1040), and a bus (1050, bus). The processor (1010), input / output devices (1020), memory (1030), and / or interface (1040) can be coupled to each other via the bus (1050). The bus (1050) corresponds to a path through which data moves.

[0095] Specifically, the processor (1010) can include at least one of a CPU (Central Processing Unit), MPU (Micro Processor Unit), MCU (Micro Controller Unit), GPU (Graphic Processing Unit), microprocessor, digital signal processing, microcontroller, financial application processor (AP, application processor), and logic elements capable of performing similar functions thereto.

[0096] The processor (1010) and memory (1030) in FIG. 1 can be configured by software on the electronic device (1000) or separate hardware.

[0097] The input / output device (1020) can include at least one of a keypad, keyboard, touch screen, and display device.

[0098] The memory (1030) can load data and / or programs, etc. At this time, the memory (1030) can include a high-speed dynamic random access memory (DRAM) and / or static random access memory (SRAM), etc., as an operating memory for improving the operation of the processor (1010). The memory (1030) can include one or more volatile memory devices such as DDR SDRAM (Double Data Rate Static DRAM), SDR SDRAM (Single Data Rate SDRAM), and / or one or more non-volatile memory devices such as EEPROM (Electrical Erasable Programmable ROM), flash memory.

[0099] The interface (1040) can perform the function of transmitting data to or receiving data from a communication network. The interface (1040) can be in a wired or wireless form. For example, the interface (1040) can include an antenna or a wired / wireless transceiver, etc.

[0100] The storage can store data and / or programs, etc. The storage can include one or more non-volatile memory devices such as a semiconductor drive (SSD, Solid State Drive), a hard drive, or a flash memory. In the present invention, the storage can store a computer program composed of instructions for executing the above-described method for providing a digital will service.

[0101] The input / output device (1020) can be applied to personal digital assistants (PDAs), portable computers, web tablets, wireless phones, mobile phones, digital music players, memory cards, or all electronic products capable of transmitting and / or receiving information in a wireless environment.

[0102] Also, the quantum circuit simulation device according to an embodiment of the present invention can be a device formed by connecting a plurality of electronic devices (1000) to each other via a network. In such a case, each module or combination of modules can be implemented as an electronic device (1000). However, this embodiment is not limited thereto.

[0103] Additionally, the quantum circuit simulation device can be implemented by at least one of a workstation, a data center, an internet data center (IDC), a direct attached storage (DAS) system, a storage area network (SAN) system, a network attached storage (NAS) system, and a redundant array of inexpensive disks or redundant array of independent disks (RAID) system, but the present embodiment is not limited thereto.

[0104] Furthermore, the quantum circuit simulation device can transmit data via a network. The network can include a network based on wired internet technology, wireless internet technology, and short-range communication technology. The wired internet technology can include, for example, at least one of a local area network (LAN) and a wide area network (WAN).

[0105] The above description merely exemplarily explains the technical idea of the present embodiment. Those with ordinary knowledge in the technical field to which the present embodiment belongs can make various modifications and variations without departing from the essential characteristics of the present embodiment. Therefore, the present embodiment is not for limiting the technical idea of the present embodiment but for explaining it, and the scope of the technical idea of the present embodiment is not limited by such an embodiment. The protection scope of the present embodiment should be interpreted by the following claims, and all technical ideas within the equivalent scope should be construed as being included in the scope of rights of the present embodiment.

Claims

1. A data module that receives patient PHR (Personal Health Record) data and performs pre-processing of the PHR data to generate final data; a model module that receives the final data and instructions and derives a pre-prediction for the patient via at least one artificial intelligence model selected according to the instructions; and A Reliable and Interpretable AI System (RIAS) that is separated from the data module and the model module and operates independently, receives the pre-prediction and the final data, and converts the pre-prediction into a final prediction; Artificial intelligence system.

2. The artificial intelligence system of claim 1 , wherein the patient comprises a myocardial infarction patient.

3. The artificial intelligence system of claim 1 , wherein the final prediction includes a mortality rate for the patient after a specified period of time.

4. 4. The artificial intelligence system of claim 3, wherein the final prediction includes a weighting of factors that contribute positively to the mortality rate and a weighting of factors that contribute negatively to the mortality rate.

5. The artificial intelligence system of claim 1 , wherein the PHR data includes a patient's post-acute myocardial infarction mortality rate or a patient's post-acute myocardial infarction reverse remodeling data.

6. The artificial intelligence system of claim 5 , wherein the PHR data further includes data that modifies or manipulates the mortality or remodeling data.

7. The data module includes: A pre-processing module receives an instruction including a preset data frame, and pre-processes the PHR data according to the data frame to generate pre-processed data; an imputation module for processing missing values ​​of the pre-processed data to generate final data; 2. The artificial intelligence system of claim 1.

8. The artificial intelligence system of claim 7 , wherein the imputation module receives the instructions and processes the missing values ​​according to the data frame.

9. 8. The artificial intelligence system of claim 7, wherein the imputation module handles the missing values ​​via at least one of mean substitution, median substitution, mode substitution, regression substitution, K-nearest neighbor substitution, and multiple imputation.

10. The artificial intelligence system of claim 1 , wherein the instructions include information for a predetermined artificial intelligence model.

11. The artificial intelligence system of claim 10 , wherein the at least one artificial intelligence model can be classified into one type if they have common functions, and the model module objectifies the common functions according to the type.

12. 12. The artificial intelligence system of claim 11, wherein the types include at least one of Gradient Boosted Decision Trees (GBDT), pytorch tabula, and Saint.

13. The RIAS is a training module that trains the artificial intelligence model, tunes hyperparameters of the artificial intelligence model, and evaluates the training module; a calibration module for correcting the pre-prediction to derive a final prediction; A local explanation module that performs a local explanation method for the contribution of each feature to the final data; at least one counterfactual explanation module that performs a counterfactual explanation scheme on the final data; 2. The artificial intelligence system of claim 1.

14. The pre-prediction is a predicted value between 0 and 1; The calibration module adjusts the pre-prediction to mimic the likelihood of an actual event occurring to derive a final prediction.

14. The artificial intelligence system of claim 13.

15. memory; and a processor in communication with the memory for performing operations; The processor, Pre-processing the patient's PHR data to generate the final data, inputting the final data into a selected artificial intelligence model to derive a pre-prediction; correcting the pre-prediction to derive a final prediction; Artificial intelligence system.

Images