System and method for predicting disease risk of companion animal and providing insurance

Abstract

The present application relates to a system and a method for predicting a disease risk of a companion animal and providing insurance based thereon, the system comprising: a data collection unit for collecting data on a companion animal; a data preprocessing unit for preprocessing the collected data; a model training unit for training a machine learning- or deep learning-based disease risk prediction model by using the preprocessed data; a risk prediction unit for predicting the disease risk of the companion animal by using the trained model; an insurance premium calculation unit for calculating an insurance premium suitable for the current health condition of the companion animal on the basis of the predicted disease risk; and an insurance proposal unit for proposing an insurance product on the basis of the calculated insurance premium, and providing an insurance premium discount according to removal of confirmed health risk factors or efforts to remove such health risk factors.

Description

System and Method for Predicting Pet Disease Risk and Providing Insurance

[0001] The present application relates to a system and method for predicting disease risk in pets and providing insurance. More specifically, the present application relates to a system and method for predicting disease risk in pets using artificial intelligence technology and providing customized insurance products based on this.

[0002] Cross-reference regarding related applications

[0003] This application claims priority to Korean Patent Application No. 10-2024-0161821 filed on November 14, 2024, the entire contents of which are incorporated by reference into this application.

[0004] With the recent increase in households raising pets, interest in pet health management and medical expenses is growing. In particular, as pets' lifespans increase, the importance of managing chronic diseases is being highlighted, leading to a rise in the burden of medical costs.

[0005] Existing pet insurance products typically feature standardized coverage and premium structures, which have limitations in adequately reflecting the individual health status or disease risk of pets. Consequently, owners of healthy pets have faced the burden of relatively high premiums, while conversely, pets at high risk of disease often found it difficult to enroll or faced limited coverage.

[0006] Furthermore, existing pet health management systems were limited to diagnosing only the current health status, and thus had limitations in predicting the likelihood of future disease occurrence and suggesting preventive management measures to prepare for it.

[0007] Meanwhile, while systems capable of predicting human disease risks and providing personalized health management services are being developed through the advancement of artificial intelligence and big data technologies, there are few instances of their application to companion animals.

[0008] Therefore, there is a need to develop new systems and methods that can accurately predict disease risks by comprehensively considering the individual health status and living environment of pets, provide customized insurance products based on this, and simultaneously promote the health of pets.

[0009] Accordingly, the present application, conceived from the above-mentioned problems, aims to provide a system and method capable of accurately predicting the disease risk of individual pets by comprehensively analyzing various health data and living environment information of pets.

[0010] This application proposes a customized insurance product tailored to the current health condition of pets based on predicted disease risk, and thereby aims to improve the equity and efficiency of insurance services.

[0011] This application aims to promote the health of companion animals by identifying health risk factors and presenting specific measures to improve them.

[0012] This application aims to encourage active health management and reduce the burden of medical expenses in the long term by providing a discount option that reflects the pet owner's efforts in health management in the insurance premium.

[0013] This application aims to provide a method for offering suitable insurance products to pets that have had difficulty obtaining insurance due to old age or pre-existing diseases.

[0014] This application aims to provide dynamic and flexible insurance services by continuously monitoring changes in the health status and disease risk of companion animals and reflecting this in insurance products.

[0015] This application aims to raise pet owners' awareness of health management and encourage active preventive activities by providing them with easy-to-understand information regarding disease risks.

[0016] By solving these problems, this application aims to simultaneously achieve the promotion of pet health and the qualitative improvement of insurance services, and ultimately contribute to the improvement of the quality of life for pets and their guardians.

[0017] A system for predicting the disease risk of a pet and providing insurance based thereon according to one embodiment of the present application may include: a data collection unit for collecting data regarding a pet; a data preprocessing unit for preprocessing the collected data; a model learning unit for learning a machine learning or deep learning-based disease risk prediction model using the preprocessed data; a risk prediction unit for predicting the disease risk of a pet using the learned model; a premium calculation unit for determining a premium suitable for the current health condition of the pet based on the predicted disease risk; and an insurance proposal unit for proposing an insurance product based on the calculated premium and providing a premium discount based on the removal of identified health risk factors or efforts to remove them.

[0018] In one embodiment, the system may further include a grade calculation unit that calculates a grade based on the predicted disease risk.

[0019] In one embodiment, the grade calculation unit can calculate the disease risk level as a grade ranging from 1 to 100.

[0020] In one embodiment, the system includes an external server and a personal terminal, and the data collection unit can collect data through the personal terminal.

[0021] In one embodiment, the external server may include at least one of a pharmaceutical company server, an insurance company server, and a clinical trial center server.

[0022] In one embodiment, the system may further include a data transmission unit that transmits the predicted disease risk to the external server.

[0023] In one embodiment, the data collection unit may collect data including at least one of pet guardian input information, medical record reception information, pet tech product information, and life logs.

[0024] In one embodiment, the pet owner input information may include at least one of the following: the pet's registration number, breed, front and side photos, neutering status, environment, weight, obesity level, medical history, walking information, health checkup results, disease name, surgery history, prescription history, medication history, hospital receipt, food information, eye photos, walking video, and insurance product information.

[0025] In one embodiment, the medical record receiving information may include at least one of health checkup, treatment, prescription, diagnosis name, and genomic information.

[0026] In one embodiment, the pet tech product information may include at least one of the following: frequency of urination, amount of urine, frequency of defecation, amount of defecation, amount of exercise, behavioral pattern, amount of food intake, and CCTV information.

[0027] In one embodiment, the lifelog may include at least one of activity volume, visited locations, blogs, app logs, and search information.

[0028] In one embodiment, the data preprocessing unit preprocesses data collected by the data collection unit, and the data may preprocess at least one of text data including electronic health records, image data including X-ray, MRI, ultrasound images, ocular images and images, dental images and images, image data, voice data, biosignal data including ECG, blood pressure, heart rate and respiratory rate, and data collected from a wearable device.

[0029] In one embodiment, the data preprocessing unit can extract features of the text using a natural language processing model including an LSTM or a Transformer for the text data.

[0030] In one embodiment, the data preprocessing unit can extract image features from the image data using an image processing model including a CNN or ResNet.

[0031] In one embodiment, the data preprocessing unit can extract features of the biosignal using a time series data processing model including 1D-CNN or LSTM for the biosignal data.

[0032] In one embodiment, the disease risk prediction model can predict the risk of at least one of heart disease, kidney disease, joint disease, skin disease, cancer, neurological disease, and endocrine disease.

[0033] In one embodiment, the system may further include a report generation unit that generates a report based on the predicted disease risk.

[0034] In one embodiment, the insurance proposal unit may perform the steps of: classifying the health status of a pet based on a predicted disease risk; calculating a premium based on the current health status according to the classified health status; adjusting coverage details based on the calculated premium; creating a customized insurance program including the adjusted coverage details and premium; and providing a premium discount option based on the removal of health risk factors or efforts to remove them.

[0035] In one embodiment, the insurance proposal unit may subdivide the insurance program by considering at least one additional piece of information among the pet's age, breed, history of past diseases, and living environment.

[0036] In one embodiment, the insurance proposal unit presents multiple insurance options considering the pet owner's preferences and budget, and may include at least one of the following as discount options based on efforts to eliminate health risk factors: completion of relevant education, change of feed, installation of safety devices, vaccination, administration of heartworm preventive medication, regular check-ups, self-physical examination, and improvement of the living environment.

[0037] A method for predicting the disease risk of a pet and providing insurance based thereon, performed by a pet disease risk prediction and insurance provision system according to another embodiment of the present application, may include: a step of collecting data regarding a pet by a data collection unit; a step of preprocessing the collected data by a data preprocessing unit; a step of learning a machine learning or deep learning-based disease risk prediction model using the preprocessed data by a model learning unit; a step of predicting the disease risk of a pet using the learned model by a risk prediction unit; a step of setting an insurance premium based on the predicted disease risk by an insurance premium calculation unit; a step of proposing an insurance product based on the calculated insurance premium by an insurance proposal unit; and a step of transmitting the predicted disease risk to an external server by a data transmission unit.

[0038] In one embodiment, the method may further include the step of calculating a grade based on the disease risk predicted by the grade calculation unit.

[0039] In one embodiment, the grade calculation step can calculate the disease risk level as a grade ranging from 1 to 100.

[0040] In one embodiment, the external server may include at least one of a pharmaceutical company server, an insurance company server, and a clinical trial center server.

[0041] In one embodiment, the step of proposing the insurance product may include: classifying the health condition of a pet based on a predicted disease risk; selecting a suitable insurance product according to the classified health condition; adjusting the coverage of the selected insurance product to suit the characteristics of the pet; calculating an insurance premium according to the adjusted coverage; and creating a customized insurance program including the calculated insurance premium and coverage.

[0042] In one embodiment, the step of proposing the insurance product may subdivide the insurance program by considering at least one additional piece of information among the pet's age, breed, history of past diseases, and living environment.

[0043] In one embodiment, the step of proposing the insurance product may present multiple insurance options by taking into account the pet owner's preferences and budget.

[0044] This application can predict the likelihood of disease occurrence in individual pets with high accuracy through a machine learning-based disease risk prediction model. This allows for the identification of the pet's health status in advance and the taking of appropriate preventive measures, thereby enabling the early detection and treatment of diseases.

[0045] This application can provide insurance services tailored to the health condition of individual pets by offering customized insurance products based on predicted disease risk. This can enhance equity in premium pricing and improve the satisfaction of policyholders.

[0046] This application encourages active health management by pet owners by specifically identifying health risk factors for companion animals and proposing improvement measures. This can lead to improved health and reduced medical costs for companion animals in the long term.

[0047] The insurance premium discount option based on health management efforts provided by this application has the effect of raising guardians' awareness of health management and encouraging preventive health management activities.

[0048] This application can provide suitable insurance products for elderly pets or pets with pre-existing conditions, thereby increasing accessibility to insurance services and reducing blind spots.

[0049] The dynamic and flexible insurance service provision method of the present application can appropriately reflect health changes according to the life cycle of pets, thereby enabling the provision of long-term and sustainable insurance services.

[0050] The easy-to-understand disease risk information provided by this application has an educational effect of raising guardians' awareness of health management and enhancing their interest in and understanding of pet health.

[0051] This application can contribute to medical research on companion animals through the accumulation and analysis of health data, which can lead to the qualitative improvement of medical services for companion animals in the long term.

[0052] This application can reduce the economic and emotional burden of raising pets by promoting their health and reducing medical expenses, and ultimately contribute to improving the quality of life for both pets and their owners.

[0053] The effects of the present application are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description in the claims.

[0054] FIG. 1 is an overall configuration diagram of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0055] FIG. 2 is a detailed configuration diagram of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0056] FIG. 3 is a diagram showing the types of data collected in a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0057] FIG. 4 is a diagram showing the detailed configuration of a data preprocessing unit and a risk prediction unit in a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0058] FIG. 5 shows an overall flowchart of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0059] FIG. 6 is a flowchart of a method for predicting the risk of pet diseases and providing insurance according to one embodiment of the present application.

[0060] FIG. 7 is a detailed flowchart of a method for predicting the risk of pet diseases and providing insurance according to one embodiment of the present application.

[0061] Hereinafter, some embodiments of the present application will be described in detail with reference to the exemplary drawings. In assigning reference numerals to the components of each drawing, the same components may have the same reference numeral as much as possible, even if they are shown in different drawings. Furthermore, in describing these embodiments, if it is determined that a detailed description of related known configurations or functions could obscure the essence of the technical concept, such detailed description may be omitted.

[0062] Definition of Terms

[0063] Where terms such as "includes," "has," "consists of," etc. are used in this specification, other parts may be added unless "only" is used. Where a component is expressed in the singular, it may include a plural unless there is a special explicit description.

[0064] Additionally, terms such as first, second, A, B, (a), (b), etc., may be used to describe the components of the present application. Unless otherwise specified, these terms are used merely to distinguish the components from other components, and the nature, order, sequence, or number of said components is not limited by such terms.

[0065] In this application, "Artificial Intelligence (AI)" refers to a technology that realizes human learning ability, reasoning ability, perceptual ability, natural language understanding ability, etc., as a computer program.

[0066] In this application, "learning" or "learning" is a term referring to performing machine learning through computing according to a procedure.

[0067] In this application, "Machine Learning" refers to a field of artificial intelligence in which a computer learns based on data, recognizes patterns, and performs decision-making.

[0068] In this application, "companion animal" refers to an animal living with humans, and primarily refers to dogs and cats, but is not limited thereto.

[0069] In this application, "disease risk" refers to a numerical representation of the likelihood of contracting a specific disease, and in this application, it refers to a value predicted through a machine learning model.

[0070] In this application, "insurance premium" refers to the amount paid by an insured to an insurance company in accordance with an insurance contract.

[0071] In this application, "health risk factors" refers to factors that may have a negative impact on the health of companion animals, and include genetic factors, environmental factors, lifestyle habits, etc.

[0072] In this application, "preprocessing" refers to the process of converting raw data into a form suitable for analysis, and includes operations such as data cleaning, transformation, and normalization.

[0073] In this application, "LSTM (Long Short-Term Memory)" refers to a type of recurrent neural network (RNN) and is a deep learning model effective for processing time series data.

[0074] In this application, "CNN (Convolutional Neural Network)" refers to a deep learning model specialized for image processing, which is a neural network capable of simultaneously performing feature extraction and classification.

[0075] In this application, "Pet-tech" is a general term for the application of advanced technology to the pet-related industry, and in this application, it primarily refers to devices and services for monitoring the health status of pets.

[0076] In this application, terms such as "unit," "module," "device," or "system" are intended to refer to a combination of hardware as well as software driven by said hardware. For example, hardware may be a data processing device including a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), or other processors. Additionally, software may refer to a running process, object, executable, thread of execution, program, etc.

[0077]

[0078] System structure

[0079] FIG. 1 is an overall configuration diagram of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0080] Referring to FIG. 1, the pet disease risk prediction and insurance provision system of the present application includes a server (10), a user terminal (20), and an external server (30).

[0081] The server (10) is a key component that performs the core functions of the present application and includes a data collection unit, a data preprocessing unit, a model training unit, a risk prediction unit, an insurance premium calculation unit, an insurance proposal unit, etc. The server (10) processes collected pet data, predicts disease risk through a machine learning model, and proposes a customized insurance product based on this.

[0082] The user terminal (20) serves as an interface through which the pet owner interacts with the system. Through this, the owner can input information about the pet and check the predicted disease risk and proposed insurance product information. The user terminal (20) can be a device of various forms, such as a smartphone, tablet, or PC.

[0083] External servers (30) may include pharmaceutical company servers, insurance company servers, clinical trial center servers, EMR servers, etc. These external servers may provide additional data or receive predicted disease risk information and utilize it for their respective purposes.

[0084] The server (10), the user terminal (20), and the external server (30) are connected to enable bidirectional communication. Through this, real-time exchange and updating of data are performed, and the efficiency and accuracy of the system can be increased.

[0085] Through this configuration, the present application can comprehensively analyze health data of pets and provide individualized disease risk prediction and customized insurance services.

[0086] FIG. 2 is a detailed configuration diagram of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0087] Referring to FIG. 2, the pet disease risk prediction and insurance provision system of the present application includes a personal terminal (20), a data preprocessing unit (2), a model learning unit (3), a risk prediction unit (4), an insurance premium calculation unit (5), an insurance proposal unit (6), a report generation unit (7), and an external server (30).

[0088] The personal terminal (20) serves as an interface through which the pet owner interacts with the system. Various data regarding the pet is collected through the personal terminal (20). This data includes pet owner input information (1a), medical record reception information (1b), pet tech product information (1c), life log (1d), etc.

[0089] The pet owner input information (1) may include at least one of the following: the pet's registration number, breed, front and side photos, neutering status, environment, weight, obesity level, medical history, walking information, health checkup results, disease name, surgery history, prescription history, medication history, hospital receipt, food information, eye photos, walking video and insurance product information.

[0090] The medical record receiving information (1) may include at least one of health check-up, treatment, prescription, diagnosis name, and genomic information. The prescription includes the drug name, dosage, duration of administration, method of administration, precautions, drug interactions, history of side effects, treatment responsiveness, medication adherence, etc.

[0091] Pet tech product information (1) may include at least one of the following: frequency of urination, amount of urine, frequency of defecation, amount of defecation, amount of exercise, behavioral pattern, amount of food intake, and CCTV information.

[0092] Lifelog (1) may include at least one of activity amount, visited places, blog, app log and search information.

[0093] The data preprocessing unit (2) preprocesses data collected through a personal terminal (20). The data to be preprocessed may include at least one of text data including electronic health records, image data including X-ray, MRI, ultrasound images, eye photos and images, tooth photos and images, image data, voice data, biosignal data including ECG, blood pressure, heart rate and respiratory rate, and data collected from a wearable device.

[0094] The data preprocessing unit (2) can extract text features using a natural language processing model including an LSTM or a Transformer for text data. For image data, it can extract image features using an image processing model including a CNN or a ResNet. For biosignal data, it can extract biosignal features using a time series data processing model including a 1D-CNN or an LSTM.

[0095] The data preprocessing unit (2) can be processed by various machine learning or deep learning algorithms including but not limited to Logistic Regression, Random Forest, Clustering, SVM, KNN, etc., and can improve prediction accuracy through ensemble learning techniques such as Bagging, Boosting, and Stacking.

[0096] The model training unit (3) trains a machine learning-based disease risk prediction model using preprocessed data. In this process, about 20 indicators are selected and used. The selected indicators include blood tests (creatinine, BUN, SDMA, electrolytes, etc.), urine tests (urine specific gravity, UPC, sugar, pH, etc.), medical history data (weight change, water intake, urine output, etc.), environmental data (diet, feeding method, living environment, activity level, etc.), and basic information (disease name, breed, gender, age, weight, obesity level, etc.).

[0097] The risk prediction unit (4) predicts the disease risk of the pet using a trained model. This model can predict the risk of at least one of heart disease, kidney disease, joint disease, skin disease, cancer, neurological disease, and endocrine disease.

[0098] The premium calculation unit (5) determines the premium based on the current health condition of the pet based on the predicted disease risk. In this process, a weight (regression coefficient) is multiplied by each characteristic (risk factor) that affects the disease risk to calculate a final score based on the individual value. This is stratified into percentiles, and the formula is optimized through model verification.

[0099] The insurance proposal department (6) proposes insurance products based on the calculated premiums and provides premium discounts based on the removal of identified health risk factors or efforts to remove them. In the insurance proposal process, the insurance program can be subdivided by considering at least one additional piece of information among the pet's age, breed, history of past diseases, and living environment. Additionally, multiple insurance options can be presented by considering the pet owner's preferences and budget.

[0100] The report generation unit (7) generates a report based on the predicted disease risk. This report may include disease risk grading, the likelihood of occurrence within a specific period, the risk relative to the average, actions to increase the risk, and methods to lower the risk.

[0101] The external server (30) may include at least one of a pharmaceutical company server, an insurance company server, a clinical trial center server, and an EMR server. The predicted disease risk may be transmitted to these external servers (30).

[0102] In addition, the system of the present application may further include a data transmission unit and a grade calculation unit, although not shown in FIG. 2.

[0103] The data transmission unit performs the role of transmitting the disease risk predicted by the risk prediction unit (4) to an external server (30). By safely and efficiently transmitting the predicted disease risk information to the external server (30), the data transmission unit enables pharmaceutical companies, insurance companies, clinical trial centers, etc., to utilize this information. Through this, the data can be utilized in various fields such as the development of new treatments, improvement of insurance products, and clinical research.

[0104] The grade calculation unit performs the role of calculating a grade based on the disease risk predicted by the risk prediction unit (4). The grade calculation unit converts the predicted disease risk into a grade ranging from 1 to 10. The grade calculated in this way allows for a more intuitive understanding and comparison of the pet's health status. This grade information can be utilized in the premium calculation unit (5) and the insurance proposal unit (6) to make more accurate and detailed insurance product proposals.

[0105] FIG. 3 is a diagram showing the types of data collected in a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0106] Referring to FIG. 3, the data collected in the pet disease risk prediction and insurance provision system of the present application can be broadly classified into four categories: pet owner input information (1a), medical record reception information (1b), pet tech product information (1c), and life log (1d).

[0107] The pet owner input information (1a) is information entered directly by the owner and may include the following items:

[0108] Registration Number: Official registration number of the pet

[0109] Species: Species of pet (e.g., dog, cat, etc.)

[0110] Front and side photos: Photos showing the pet's appearance

[0111] Neutering Status: Whether the pet has been neutered

[0112] Environment: Information on the environment where pets live

[0113] Weight: Pet's current weight

[0114] Obesity level: The degree of obesity in pets

[0115] Other medical records: Health status of the pet observed by the guardian, etc.

[0116] Walking Information: Walking frequency, time, distance, etc.

[0117] Health checkup results: Regular health checkup results

[0118] Disease Name: Information on past or present diseases

[0119] Surgical History: Past surgical experience and types

[0120] Prescription History: Received Prescription Information

[0121] Medication History: Information on medications currently being taken or taken in the past

[0122] Hospital Receipt: Details of Medical Expenses

[0123] Feed Information: Type and amount of feed to be given, etc.

[0124] Eye photograph: A photograph used to check the condition of eye health.

[0125] Walking video: A video that allows you to check your pet's walking condition

[0126] Insurance Product Information: Information on currently subscribed insurance products (if applicable)

[0127] Medical record receiving information (1b) is information received directly from an animal hospital or medical institution and may include the following items:

[0128] Health Checkup: Results of regular or irregular health checkups

[0129] Treatment: Type and details of treatment received

[0130] Prescription: Prescribed medications and treatments

[0131] Diagnosis: Official diagnosis given by medical staff or diagnosis entered by the guardian

[0132] Genomic Information: Genetic test results (if applicable)

[0133] Pet tech product information (1c) is information collected from various smart devices that monitor the health status of pets, and may include the following items:

[0134] Urinary frequency and volume: Monitoring urination patterns

[0135] Bowel movement frequency and volume: Monitoring bowel movement patterns

[0136] Exercise volume: Daily activity level, calorie expenditure, etc.

[0137] Behavioral patterns: Routine behavioral patterns such as sleep time and rest time.

[0138] Feed intake: Amount of feed consumed per day

[0139] CCTV Information: Pet behavior observed through cameras installed in the home

[0140] Lifelog (1d) is information automatically collected from the daily life of the pet and the guardian, and may include the following items:

[0141] Activity level: Daily step count, distance traveled, etc.

[0142] Places visited: Places frequently visited (e.g., parks, veterinary clinics, etc.)

[0143] Blog: Pet-related blog posts written by guardians

[0144] App Log: Pet-related App Usage History

[0145] Search: Online search history related to pet health

[0146] These various data are input into the system of the present application and processed in the data preprocessing unit (2). The preprocessed data is used in the model training unit (3) to train a machine learning or deep learning-based disease risk prediction model.

[0147] By comprehensively analyzing such extensive and diverse data, this application enables a more accurate and comprehensive prediction of disease risk compared to existing methods that rely solely on medical information. In particular, by considering the pet's daily life patterns, environment, and dietary habits, it offers the advantage of early detection and prevention of chronic disease risks.

[0148] In addition, based on this data, about 20 key indicators can be selected and utilized for further analysis. For example, these include blood test results (creatinine, BUN, SDMA, electrolytes, etc.), urine test results (urinary specific gravity, UPC, glucose, pH, etc.), medical history data (weight change, water intake, urination volume, etc.), environmental data (diet, feeding method, living environment, activity level, etc.), and basic information (disease name, breed, gender, age, weight, obesity level, etc.).

[0149] The data collected and analyzed in this way is ultimately used to predict the disease risk of pets and propose customized insurance products based on this. This enables the simultaneous achievement of improved pet health and enhanced quality of insurance services.

[0150] FIG. 4 is a diagram showing the detailed configuration of a data preprocessing unit and a risk prediction unit in a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0151] Referring to FIG. 4, the pet disease risk prediction and insurance provision system of the present application includes a data preprocessing unit (2) and a risk prediction unit (4).

[0152] The data preprocessing unit (2) performs the role of processing various forms of collected data and converting them into a form suitable for analysis. The data preprocessing unit (2) processes the following data types:

[0153] Text data: Electronic health records, diagnoses, prescription history, etc.

[0154] Image data: X-ray, MRI, ultrasound images and images, ocular photographs, dental photographs, etc.

[0155] Video / Audio Data: Walking footage, heart sounds, etc.

[0156] Biosignal data: ECG, blood pressure, heart rate, respiratory rate, etc.

[0157] Lifelog data: activity level, food intake, sleep patterns, etc.

[0158] Other (Genomic) Data: Genetic test results, etc.

[0159] An appropriate preprocessing model is applied for each data type:

[0160] Heart disease model: Heart-related data processing

[0161] Kidney Disease Model: Data Processing Related to Kidney Function

[0162] Joint Disease Model: Processing Joint and Movement-Related Data

[0163] Skin disease model: Data processing related to skin condition

[0164] Cancer Model: Tumor-Related Data Processing

[0165] Neurological Disorder Model: Nervous System-Related Data Processing

[0166] Endocrine Disease Models: Hormone and Metabolism Data Processing

[0167] These preprocessing steps are performed using deep learning techniques such as LSTM, Transformer, CNN, ResNet, and 1D-CNN. For example, important features are extracted from text data using LSTM or Transformer, and image data is processed through CNN or ResNet.

[0168] The risk prediction unit (4) predicts the risk level for each disease based on the preprocessed data. In this process, detailed risk levels such as heart disease risk, kidney disease risk, joint disease risk, skin disease risk, cancer risk, neurological disease risk, and endocrine disease risk are calculated.

[0169] The risk prediction unit (4) calculates the risk using about 20 major indicators. These indicators include blood test results (creatinine, BUN, SDMA, electrolytes, etc.), urine test results (urine specific gravity, UPC, sugar, pH, etc.), medical history data (weight change, water intake, urine output, etc.), environmental data (diet, feeding method, living environment, activity level, etc.), and basic information (disease name, breed, gender, age, weight, obesity level, etc.).

[0170] In the risk calculation process, a final score based on individual values ​​is calculated by multiplying each characteristic (risk factor) affecting disease risk by a weight (regression coefficient). This score is calculated using the following formula:

[0171] Risk Score = w1 x Feature1 + w2 x Feature2 + ... + wn x Featuren

[0172] Here, wn is the weight of each feature, and Featuren is the feature value for each individual.

[0173] The calculated risk scores are stratified into percentiles and presented for intuitive understanding. During this process, the formula is continuously optimized through model validation.

[0174] The disease-specific risk calculated in this way is transmitted to the premium calculation unit (5) and used as basic data for determining customized premiums for individual pets. In addition, through the report generation unit (7), it can be provided to the pet owner in an easy-to-understand health report format.

[0175] FIG. 5 shows an overall flowchart of a pet disease risk prediction and insurance provision system according to one embodiment of the present application.

[0176] Referring to FIG. 5, the pet disease risk prediction and insurance provision system of the present application is largely composed of the steps of data collection, data processing, risk prediction, and result utilization.

[0177] First, in the data collection stage, health data of pets is collected through the animal hospital (100) and the company's pet health checkup service, PETING (200). General medical data is collected from the animal hospital (100), and more detailed and regular health checkup data is collected through PETING (200). The data collected in this way is integrated in the form of checkup results (dedicated DB linkage information) (300).

[0178] The collected data is stored in a set DB (400), which serves as the basis of the system. The set DB (400) contains about 20 major indicators. These include blood test results (creatinine, BUN, SDMA, electrolytes, etc.), urine test results (urine specific gravity, UPC, sugar, pH, etc.), medical history data (weight change, water intake, urine output, etc.), environmental data (diet, feeding method, living environment, activity level, etc.), and basic information (disease name, breed, gender, age, weight, obesity level, etc.).

[0179] Next, in the AI ​​model development (500) stage, a risk prediction AI model is developed and trained based on data from the set DB (400). This process can be performed by the model training unit (3). The AI ​​model is trained to predict the disease risk of pets using machine learning techniques.

[0180] The risk prediction AI (600) predicts the disease risk of individual pets based on a trained model. This can be performed by the risk prediction unit (4). The predicted result is output in the form of a health checkup report (700), which includes individual risk information.

[0181] The health checkup report (700) provides information such as risk scores, risk factors, and preventive measures. This corresponds to information generated by the report generation unit (7). This information can be provided directly to the pet owner and used for the health management of the pet.

[0182] Finally, the output of the system can be utilized in various ways. For example, a customized insurance product is developed and provided based on the predicted risk level through the rational linkage insurance product (800). This reflects the functions of the premium calculation unit (5) and the insurance proposal unit (6).

[0183] Individualized health management programs are provided according to predicted risks through disease-specific customized insurance products (900).

[0184] Through insurance underwriting and loss ratio management (1000), the insurance company can utilize the results of this system to perform more accurate insurance underwriting and loss ratio management.

[0185] FIG. 6 is a flowchart of a method for predicting the risk of pet diseases and providing insurance according to one embodiment of the present application.

[0186] Referring to FIG. 6, the method for predicting the risk of pet diseases and providing insurance according to the present application consists of the following steps. This method may be operated by a processor, and each step may be performed by each component of the system described in FIG. 2.

[0187] First, in the step (S610) of collecting data about a pet by the data collection unit (1), health-related data of the pet is collected from various sources. This data may include pet owner input information, medical record reception information, pet tech product information, life logs, etc.

[0188] Next, in the step (S620) of preprocessing the collected data by the data preprocessing unit (2), the collected raw data is converted into a form suitable for analysis. In this process, various types of data, such as text data, image data, and biosignal data, are processed in a manner suitable for each.

[0189] In the step (S630) of training a machine learning-based disease risk prediction model using data preprocessed by the model training unit (3), an AI model is trained based on the preprocessed data. In this process, a risk prediction model for various diseases such as heart disease, kidney disease, and joint disease is developed.

[0190] In the step (S640) of predicting the disease risk of a pet using a model learned by the risk prediction unit (4), data of individual pets is input into the learned model to calculate the risk for each disease.

[0191] In the step (S650) of determining the insurance premium based on the current health condition of the pet based on the disease risk predicted by the insurance premium calculation unit (5), an individualized insurance premium is calculated based on the predicted risk.

[0192] In the step (S660) of proposing an insurance product based on the premium calculated by the insurance proposal department (6) and providing a premium discount based on the removal or removal efforts of identified health risk factors, a customized insurance product is proposed and an incentive is provided based on health improvement efforts.

[0193] In the step (S670) of transmitting the predicted disease risk level by the data transmission unit to an external server, the analysis results are shared with relevant institutions.

[0194] Finally, in the step (S680) of generating a report based on the disease risk predicted by the report generation unit, a comprehensive health report to be provided to the pet owner is prepared.

[0195] Each of these steps may proceed sequentially, but depending on the system configuration or user requirements, some steps may be performed simultaneously or in a different order.

[0196] FIG. 7 is a detailed flowchart of a method for predicting the risk of pet diseases and providing insurance according to one embodiment of the present application.

[0197] Referring to FIG. 7, the method for predicting the risk of pet diseases and providing insurance according to the present application consists of the following detailed steps. This method can be operated by a processor, and each step can be performed by each component of the system described in FIG. 2.

[0198] First, in the step (S710) of classifying the health status of the pet based on the predicted disease risk, the overall health status of the pet is evaluated based on the disease risk calculated by the risk prediction unit (4). In this process, the risk of various diseases, such as heart disease, kidney disease, and joint disease, is comprehensively considered.

[0199] Next, in the step (S720) of calculating the premium based on the current health status according to the classified health status, the premium calculation unit (5) calculates a basic premium suitable for each individual pet based on the previously classified health status. In this step, the pet's age, breed, and past disease history are also taken into consideration.

[0200] In the step of adjusting coverage details based on the calculated premium (S730), the insurance proposal unit (6) designs optimal coverage details suitable for each pet based on the calculated premium. In this process, coverage for specific diseases can be strengthened or adjusted according to the predicted disease risk.

[0201] In the step of creating a customized insurance program including adjusted coverage details and premiums (S740), the previously adjusted coverage details and calculated premiums are combined to form a customized insurance product suitable for the individual pet. In this step, the pet owner's preferences and budget may also be considered.

[0202] Finally, in the step (S750) of providing premium discount options based on the removal of health risk factors or efforts to remove them, the insurance proposal unit (6) proposes an incentive system based on efforts to improve the health of the pet. This may include premium discount options for various forms of health management activities such as completing relevant education, changing food, installing safety devices, vaccination, regular checkups, walking, and exercise.

[0203] While each of these steps may proceed sequentially, some steps may be performed simultaneously or in a different order depending on system settings or user requirements. Through this method, customized insurance products that accurately reflect the current health status of pets can be provided, while simultaneously encouraging active participation from pet owners to promote their pets' health. This can ultimately contribute to achieving two goals at once: improving the quality of life for pets and reducing the financial burden on pet owners.

[0204] Such a system and method for predicting the risk of pet diseases and providing insurance may be implemented as an application or in the form of program instructions that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., either individually or in combination.

[0205] The program instructions recorded on the above-mentioned computer-readable recording medium are those specifically designed and configured for the present application, but may also be those known and available to those skilled in the art of computer software.

[0206] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0207] Examples of program instructions include machine codes, such as those generated by a compiler, as well as high-level language codes that can be executed by a computer using an interpreter, etc. The hardware device may be configured to operate as one or more software modules to perform processing according to the present application, and vice versa.

[0208] Although the foregoing has been described with reference to embodiments, those skilled in the art will understand that various modifications and changes can be made to the present application without departing from the spirit and scope of the application as set forth in the following claims.

[0209] This application relates to a system and method for predicting disease risk based on a pet's health data and living environment information, and providing customized insurance products accordingly. It can be utilized in various industrial fields, such as the pet insurance industry, veterinary medical services, and pet healthcare platforms. In particular, this invention contributes to reducing medical costs, promoting preventive medical care, and enhancing the equity and efficiency of insurance services by continuously monitoring a pet's health status to identify risk factors and suggest improvement measures. Furthermore, it enables the creation of new markets by establishing indicators to provide appropriate insurance products even to pets that previously had difficulty obtaining insurance. Additionally, by providing a discount structure that reflects the pet owner's efforts in health management in the insurance premium, it is expected to revitalize the pet health management ecosystem. Therefore, this invention can be practically implemented and applied across the entire pet insurance and healthcare industry.

Claims

1. In a system that predicts the disease risk of companion animals and provides insurance based thereon, A data collection unit that collects data regarding pets; A data preprocessing unit that preprocesses collected data; A model training unit that trains a machine learning or deep learning-based disease risk prediction model using preprocessed data; Risk prediction unit that predicts the disease risk of pets using a trained model; An insurance premium calculation unit that determines an insurance premium suitable for the pet's current health condition based on predicted disease risk; and A pet disease risk prediction and insurance provision system characterized by including an insurance proposal unit that proposes insurance products based on calculated premiums and provides premium discounts based on the removal of identified health risk factors or efforts to remove them.

2. In Paragraph 1, A pet disease risk prediction and insurance provision system characterized by further including a grade calculation unit that calculates a grade based on the predicted disease risk.

3. In Paragraph 2, A pet disease risk prediction and insurance provision system characterized by the above-mentioned grade calculation unit calculating the disease risk level into grades ranging from 1 to 100.

4. In Paragraph 1, A pet disease risk prediction and insurance provision system comprising an external server and a personal terminal, wherein the data collection unit collects data through the personal terminal.

5. In Paragraph 4, A pet disease risk prediction and insurance provision system characterized in that the above-mentioned external server includes at least one of a pharmaceutical company server, an insurance company server, a clinical trial center server, and an EMR server.

6. In Paragraph 4 or 5, A pet disease risk prediction and insurance provision system characterized by further including a data transmission unit that transmits the predicted disease risk to the external server.

7. In Paragraph 1, A pet disease risk prediction and insurance provision system characterized by the above-mentioned data collection unit collecting data including at least one of pet guardian input information, medical record reception information, pet tech product information, and life logs.

8. In Paragraph 7, A system for predicting disease risk and providing insurance for pets, characterized in that the above-mentioned pet guardian input information includes at least one of the following: pet registration number, breed, front and side photos, neutering status, environment, weight, obesity level, medical history, walking information, health checkup results, disease name, surgery history, prescription history, medication history, hospital receipt, food information, eye photos, walking video, and insurance product information.

9. In Paragraph 7, A pet disease risk prediction and insurance provision system characterized by the above-mentioned medical record reception information including at least one of health checkup, treatment, prescription, diagnosis name, and genomic information.

10. In Paragraph 7, A pet disease risk prediction and insurance provision system characterized by including at least one of the above pet tech product information, such as urination frequency, urine volume, defecation frequency, defecation volume, exercise volume, behavioral pattern, food intake, and CCTV information.

11. In Paragraph 7, A pet disease risk prediction and insurance provision system characterized by the above-mentioned lifelog including at least one of activity volume, visited locations, blogs, app logs, and search information.

12. In Paragraph 1, The above data preprocessing unit preprocesses the data collected by the above data collection unit, and The above data A pet disease risk prediction and insurance provision system characterized by preprocessing at least one of text data including electronic health records, image data including X-ray, MRI, ultrasound images, ocular photographs and images, dental photographs and images, image data, voice data, biosignal data including ECG, blood pressure, heart rate and respiratory rate, and data collected from wearable devices.

13. In Paragraph 12, A pet disease risk prediction and insurance provision system characterized by the above data preprocessing unit extracting text features using a natural language processing model including an LSTM or a Transformer for the above text data.

14. In Paragraph 12, A pet disease risk prediction and insurance provision system characterized by the above-mentioned data preprocessing unit extracting image features using an image processing model including CNN or ResNet for the image data.

15. In Paragraph 12, A pet disease risk prediction and insurance provision system characterized by the above data preprocessing unit extracting features of the biosignal using a time series data processing model including 1D-CNN or LSTM for the biosignal data.

16. In Paragraph 1, A pet disease risk prediction and insurance provision system characterized by the above-mentioned disease risk prediction model predicting the risk of at least one of heart disease, kidney disease, joint disease, skin disease, cancer, neurological disease, and endocrine disease.

17. In Paragraph 1, It further includes a report generation unit that generates a report based on predicted disease risk, and The above report is a pet disease risk prediction and insurance provision system characterized by including at least one of disease risk grading, the probability of occurrence within a specific period, risk relative to the average, actions to increase risk, and methods to lower risk.

18. In Paragraph 1, The aforementioned insurance proposal department, A step of classifying the health status of pets based on predicted disease risk; A step of calculating current health status-based insurance premiums according to classified health status; A step of adjusting coverage details based on the calculated premium; Steps for creating a customized insurance program including adjusted coverage details and premiums; and A pet disease risk prediction and insurance provision system characterized by performing the step of providing an insurance premium discount option based on the removal of health risk factors or efforts to remove them.

19. In Paragraph 18, The above-mentioned insurance proposal unit is characterized by subdividing the insurance program by considering at least one additional piece of information among the pet's age, breed, past disease history, and living environment.

20. In Paragraph 18 or 19, The above-mentioned insurance proposal section presents multiple insurance options considering the preferences and budget of the pet owner, and is characterized by including at least one of the following as discount options based on efforts to eliminate health risk factors: completion of relevant education, change of food, installation of safety devices, vaccination, administration of heartworm preventive medication, regular check-ups, self-physical examination, improvement of living environment, walking, and exercise.

21. A method for predicting the disease risk of a pet and providing insurance based thereon, performed by a pet disease risk prediction and insurance provision system, wherein A step of collecting data regarding pets by a data collection unit; A step of preprocessing data collected by a data preprocessing unit; A step of training a machine learning or deep learning-based disease risk prediction model using data preprocessed by a model training unit; A step of predicting the disease risk of a pet using a model trained by a risk prediction unit; A step of determining the insurance premium based on the disease risk predicted by the insurance premium calculation unit; A step of proposing an insurance product based on the premium calculated by the insurance proposal department; and A method for predicting disease risk in pets and providing insurance, characterized by including the step of transmitting the predicted disease risk level to an external server by a data transmission unit.

22. In Paragraph 21, A method for predicting disease risk in pets and providing insurance, characterized by further including a step of calculating a grade based on the disease risk predicted by a grade calculation unit.

23. In Paragraph 22, A method for predicting disease risk and providing insurance for pets, characterized by the step of calculating the above grade, which calculates the disease risk level into a grade ranging from 1 to 100.

24. In Paragraph 21, A method for predicting disease risk in companion animals and providing insurance, characterized in that the above-mentioned external server includes at least one of a pharmaceutical company server, an insurance company server, a clinical trial center server, and an EMR server.

25. In Paragraph 21, The step of proposing the above insurance product is, A step of classifying the health status of pets based on predicted disease risk; Step of selecting a suitable insurance product based on classified health conditions; A step of adjusting the coverage of the selected insurance product to suit the characteristics of the pet; A step of calculating the premium based on the adjusted coverage details; and A method for predicting pet disease risk and providing insurance, characterized by including the step of generating a customized insurance program including calculated premiums and coverage details.

26. In Paragraph 25, A method for predicting disease risk and providing insurance for pets, characterized by the step of proposing the insurance product above subdividing the insurance program by considering at least one additional piece of information among the pet's age, breed, past disease history, and living environment.

27. In Paragraph 25 or 26, A method for predicting pet disease risk and providing insurance, characterized by the step of proposing the insurance product above presenting multiple insurance options while considering the pet owner's preferences and budget.

Images