System and method for providing pet home care guide and guardian-customized information based on emr analysis

Abstract

The present invention relates to a system and method for providing a pet home care guide and customized information to a guardian based on medical history. The system according to the present invention comprises: a data interface unit that collects electronic medical record (EMR) data of a pet by linking with an animal hospital server; a guardian profile analysis unit that determines a report generation mode (public / simple mode) and a template type (detailed description / comparative analysis type) by analyzing the guardian's number of hospital visits and the main symptom (CC) disclosure option value; a challenge mapping unit that extracts home care missions optimized for categories such as observation, medication, and environmental management by analyzing diagnosis names; an adaptive report generation unit that generates an intermediate report including customized medical information and inner thoughts messages from the pet's perspective based on the determined template and mission; a compliance filtering unit that generates a final report by identifying words indicating a negative prognosis or expressions with potential legal risks within the text of the intermediate report and automatically replacing them with predefined safe expressions; and a report provision unit that transmits the final report to a guardian terminal. Accordingly, this can provide contextually relevant and easy-to-understand information to guardians lacking medical knowledge, prevent medical and legal risks in advance, and strengthen adherence to home treatment and emotional bonding with pets.

Description

Technology Field

[0001] The present invention relates to a pet home care and customized information provision system, and more specifically, to a technology that analyzes electronic medical record (EMR) data from an animal hospital to generate an adaptive report optimized for the guardian's visit history and the characteristics of the pet, and provides a home care challenge mission to improve treatment compliance and compliance filtering to prevent medical and legal risks. Background Technology

[0003] Amidst demographic changes such as the rapid increase in single-person households and population aging, the "pet family" trend—where companion animals are recognized not merely as pets but as family members sharing their lives—is becoming widespread in modern society. Consequently, societal expectations regarding the health management and disease treatment of companion animals have evolved beyond the passive treatment of the past to a stage demanding precise diagnosis and systematic follow-up care comparable to that of human physicians. In line with this trend, veterinary clinics are mandatorily adopting and operating Electronic Medical Record (EMR) systems to digitize and manage medical records, prescriptions, and examination images, thereby enhancing the efficiency of medical care and systematically accumulating medical history data.

[0004] However, conventional EMR systems are inherently "expert-centered systems" designed to enable veterinarians, as medical professionals, to perform clinical duties efficiently and comply with legal retention obligations. Consequently, there is a fundamental limitation in that the vast amount of vital sign data, various test results, and specialized medical terms listed as abbreviations recorded in EMRs are nearly impossible for general pet owners without a medical background to intuitively decipher. Even if a veterinarian provides verbal explanations during the consultation, it is extremely difficult for a psychologically intimidated owner to fully remember and implement the correct medication timing, storage methods, or emergency response procedures after returning home. This "asymmetry of medical information" and "volatility of transmission" lead to reduced treatment compliance at home, which diminishes treatment effectiveness or causes owners to rely on unverified folk remedies found online, thereby becoming a major cause of the worsening condition of their pets.

[0005] Although there have been various technological attempts to address these problems and assist caregivers in child-rearing, existing technologies each exhibit distinct limitations in their respective domains.

[0006] First, a technology that provides educational information according to the life cycle of pets (public patent 10-2022-0077415, etc.) has been proposed. This technology is useful in that it provides guidance to novice pet owners by offering information such as basic training, vaccination timing, and pet etiquette required according to the pet's age. However, this is merely a static information system that provides textbook-like 'general principles,' and it has a fatal limitation in that it cannot provide 'individualized medical guidance' in real time, such as dietary restrictions, exercise restrictions, and medication management, which are immediately needed for pets diagnosed with specific diseases like 'chronic renal failure' or 'patellar luxation' at the hospital today.

[0007] Second, numerous technologies that analyze a pet's health status and recommend customized feed or nutritional supplements (e.g., Registered Patents 10-2271991, 10-2527364) have been commercialized. However, these technologies are primarily aimed at 'dietary therapy' that replenishes deficient nutrients based on surveys or genetic test results, or at linking them to the commerce of related products. Although the treatment of diseases requires not only nutritional supply but also complex nursing actions such as accurate drug administration, wound disinfection, and monitoring for abnormal signs, conventional technologies fail to encompass these aspects and remain at the level of food intake guidelines, making them difficult to utilize as practical aids for treatment.

[0008] Third, with the recent advancement of IoT (Internet of Things) technology, techniques that measure a pet's activity level or calorie expenditure using wearable devices such as smart pendants (e.g., Registered Patent 10-2022893) are gaining attention. However, such sensor-based technologies merely collect physical data regarding the "frequency of movement" through accelerometers or gyroscopes; they fail to identify medical causal relationships regarding whether the decrease in movement is due to "joint pain," "weakness caused by heart disease," or simply "aging." In other words, sensor data not combined with hospital EMR data is merely a monitoring metric and lacks the depth of information to serve as a basis for judging disease prognosis or providing specific behavioral guidelines.

[0009] Above all, the aforementioned conventional technologies are limited to simply listing data as is or mechanically matching and displaying it, failing to consider the "psychological state of the guardian" or "information reception capacity" at all. For example, as the number of hospital visits by guardians increases, information regarding specific numerical changes or prognosis management is needed rather than basic information already known; however, existing systems repeatedly provide only uniform information, causing information fatigue.

[0010] A more serious problem is that there are no safeguards against the 'medical and legal risks' that may arise during the process of automatically generating medical reports using AI technology. If sensitive text such as 'potential for death,' 'consideration of euthanasia,' or 'inherent aggression' included in the original medical chart, or definitive diagnostic expressions made before a diagnosis is confirmed, are exposed to guardians without any filtering process, guardians may suffer severe psychological trauma or develop distrust regarding the hospital's medical practices, which is highly likely to escalate into a medical dispute.

[0011] Therefore, there is an urgent need to develop a human-centered intelligent home care guide system that is based on EMR, which is core data in actual medical settings, but actively adjusts the level of detail of information by recognizing contexts such as the guardian's visit history, detects expressions that may cause medical or legal issues in advance and softens them into safe language, and supports emotional bonding with pets beyond simple information delivery. Prior art literature

[0013] Republic of Korea Published Patent No. 10-2022-0077415 Republic of Korea Registered Patent No. 10-2271991 Republic of Korea Registered Patent No. 10-2527364 Republic of Korea Registered Patent No. 10-2022893 The problem to be solved

[0014] The present invention aims to solve the problems of the prior art described above, and its primary objective is to support guardians with limited medical knowledge in intuitively understanding complex Electronic Medical Record (EMR) data. To this end, the invention seeks to establish an adaptive information provision system that automatically recognizes context, such as the guardian's visit history and the characteristics of the pet, and selectively provides 'detailed explanatory' content guiding new guardians on disease definitions and basic management, and 'comparative analysis' content showing trends in medical data changes to existing guardians.

[0015] Furthermore, the present invention aims to fundamentally prevent medical and legal risks that may arise during the process of generating AI-automated reports. Specifically, it seeks to provide an intelligent compliance filtering technology that identifies in advance negative prognostic terms, such as 'death' or 'euthanasia,' which can cause psychological trauma to guardians, or definitive expressions that may lead to veterinary disputes, and automatically converts them into comforting phrases or softened expressions.

[0016] Furthermore, the present invention aims to go beyond simply notifying treatment results and improve actual adherence to treatment at home. To this end, it aims to induce caregivers to perform continuous and systematic follow-up care by automatically recommending specific 'home care challenge missions' linked to the diagnosis and visualizing the caregiver's mission performance results through date stamps and a calendar UI.

[0017] Ultimately, the purpose of this invention is to provide a healthcare platform that acts as an emotional bridge between pets and their guardians, rather than merely a conveyor of medical information, by comprehensively analyzing the guardian's app usage patterns and mission performance data to calculate the 'temperature of bonding' with the pet, and dynamically changing the emotional tone of reports or the composition of content according to this index. means of solving the problem

[0019] To solve the aforementioned problems, the present invention proposes a treatment history-based pet home care guide and a system for providing customized information to pet owners, which generates adaptive reports by recognizing the owner's context based on EMR data from an animal hospital, filters medical and legal risks, and encourages continuous home care.

[0020] The system of the present invention is equipped with a data interface unit that collects EMR data of companion animals in real time by linking with an animal hospital server. This data interface unit detects the time of completion of treatment and automatically loads prescription details, test results, etc., along with a medical record in SOAP format consisting of subjective findings, objective findings, evaluations, and plans for the treatment.

[0021] The collected EMR data is transmitted to the guardian profile analysis unit, which extracts profile information including the guardian's cumulative number of hospital visits and the type of pet (dog / cat). In particular, as a key feature of this invention, the report generation mode is determined by verifying the Chief Complaint (CC) disclosure option value pre-set by the guardian. If the guardian wishes to disclose the Chief Complaint, a report is generated in 'Full Disclosure Mode,' which includes the diagnosis, detailed description, and home care mission; if the guardian desires only simple information to protect privacy, a report is generated in 'Simple Mode,' which contains only a personal message. This dual-mode system ensures flexibility in providing medical information while respecting the guardian's willingness to accept the information.

[0022] Furthermore, the guardian profile analysis department classifies the guardian as a new member when the number of visits is less than 2 and generates a first template identifier, and classifies the guardian as an existing member when the number of visits is 2 or more and generates a second template identifier, thereby laying the foundation for providing customized information based on the guardian's understanding of the disease.

[0023] The adaptive report generation unit receives the template identifier and generation mode to generate an intermediate report. In the case of the public mode, for new guardians who have received the first template identifier, it generates education-oriented content that provides detailed guidance on basic information about the disease, such as the definition of the diagnosis, causes of onset, symptoms, and general management methods. Conversely, for existing guardians who have received the second template identifier, it omits basic explanations and generates trend analysis-oriented content focusing on test results changed compared to previous medical records, medication changes, and whether the prognosis has improved. To this end, the adaptive report generation unit includes a historical data extraction module that extracts values ​​of key test items from past N medical records, a numerical change calculation module that calculates the rate of increase / decrease and the direction of change by comparing the extracted past and current values, a drug tracking module that tracks changes in the type, dosage, and frequency of administration of prescribed medications, and a visualization module that generates a time-series graph visualizing the rate of increase / decrease or change summary information summarizing numerical changes in text.

[0024] Meanwhile, the Challenge Mapping Unit analyzes the diagnosis and extracts customized missions suitable for the disease from the database, choosing from four categories: 'Observation,' 'Daily Care,' 'Medication,' and 'Environmental Management.' The challenge database contains pre-stored multiple diagnosis codes, lists of associated challenge missions, and priority weights for each diagnosis; the Challenge Mapping Unit retrieves these related missions by querying the database using the diagnosis code as a key. If multiple diagnoses exist within a single medical treatment, priority weights set according to the severity or urgency of each diagnosis are applied to select the top N missions. By combining these selected missions in a balanced manner across categories, the unit provides a home care guide that is feasible for caregivers to implement.

[0025] The adaptive report generation unit also generates 'inner thoughts messages' written from the perspective of the pet. In this process, it independently applies a title determination logic that selects a title for the guardian based on predefined title determination rules derived from pet type information, and a logic that adjusts the emotional tone according to the number of visits. Specifically, for new guardians with fewer than two visits, it applies a first emotional tone emphasizing gratitude and hope, such as "Thank you for treating me so diligently even though we just met today," and for existing guardians with two or more visits, it applies a second emotional tone emphasizing intimacy and affection, such as "Nice to see you again, Butler! Please take good care of me this time too." By generating natural language text in this manner, it provides content that forms an emotional bond rather than a mechanical message.

[0026] The compliance filtering unit, which is a key differentiating element of the present invention, analyzes the text of the intermediate report generated by the adaptive report generation unit using natural language processing technology to identify expressions that may pose legal risks. The compliance filtering unit is equipped with a substitution rule table in which target keywords are mapped to alternative expressions. By referring to this table, it deletes words with negative prognoses, such as "death" or "euthanasia," or replaces them with comforting phrases such as "Let's get through this difficult time together." Furthermore, if a specific drug name is detected, it replaces it with a sentence explaining the efficacy or precautions of the drug to help the guardian understand, going beyond a simple listing of drug names. Additionally, if expressions indicating aggression in the pet ("attempting to bite," "fierceness," etc.) are detected, they are converted into softened expressions such as "defensive behavior due to pain response," thereby ensuring that the guardian does not develop a negative perception of the pet. Once the final report is completed after undergoing this compliance filtering, the report provider transmits it to the guardian's terminal.

[0027] The present invention goes beyond merely providing one-time information and includes a feedback mechanism for managing the caregiver's continuous treatment adherence. The mission management unit receives a signal indicating the completion of a home care challenge mission from the caregiver terminal, and calculates a daily health management score by referring to category attributes in the challenge database to assign a high first weight to missions in the 'medication' and 'observation' categories and a relatively low second weight to missions in the 'daily care' and 'environment management' categories. The calculated score and mission execution history are converted into date-specific stamp images and mapped to a calendar-type UI, allowing the caregiver to intuitively check their care history, and the report provider provides this as visualized treatment adherence history information.

[0028] Furthermore, the present invention further includes a bonding index calculation unit for measuring and strengthening emotional bonding between a guardian and a pet. This calculation unit calculates a "bonding temperature index" in real time, representing the bond between the guardian and the pet, by utilizing complex variables such as the guardian's report viewing time, mission completion frequency, and hospital visit cycle. When this bonding temperature index exceeds a preset threshold, the adaptive report generation unit increases the expression intensity of the emotional tone control logic to generate text with a more intimate tone, or dynamically changes the UI configuration of the report by activating a "hidden content" area in the form of a diary recording the pet's daily life in addition to the basic report, thereby providing a reward for the guardian's dedicated care and inducing continuous participation.

[0029] Through the above configuration, the present invention realizes an integrated healthcare platform that respects the guardian's visit history and privacy settings while providing accurate and safe medical information based on EMR data, improves treatment compliance through actionable home care missions and visualized feedback, and ultimately strengthens the emotional bond between the guardian and the pet. Effects of the invention

[0031] According to the present invention, by automatically converting and providing Electronic Medical Record (EMR) data from an animal hospital into 'detailed descriptive' or 'comparative analysis' content tailored to the guardian's visit history and information processing capacity, even guardians with limited medical knowledge can intuitively understand their pet's condition. This effectively relieves vague anxiety for new guardians and prevents information fatigue for existing guardians by reducing unnecessary redundant explanations. At the same time, from the perspective of the animal hospital, it significantly improves operational efficiency by automating repetitive post-treatment explanation tasks and ensures the delivery of accurate medical information free from human error.

[0032] Furthermore, the present invention provides an intelligent compliance filtering function that fundamentally blocks medical and legal risks that may arise from AI-generated reports. By identifying negative prognostic words such as "death" or "euthanasia" or definitive diagnostic expressions in advance and automatically replacing them with comforting phrases or safe language, it minimizes the psychological impact on the guardian. In particular, through delicate language adjustments—such as softening descriptions of aggressive pet behavior into pain responses—it ensures that guardians do not develop negative perceptions of their pets. Consequently, this significantly reduces the possibility of unnecessary misunderstandings or legal disputes between the hospital and the guardian.

[0033] Furthermore, the present invention provides a 'home care challenge mission' linked to a diagnosis, and by visualizing weighted scores based on the importance of the mission and performance history in a calendar UI, it dramatically improves adherence to treatment at home. Guardians can feel a sense of accomplishment by confirming through objective indicators how much their care activities have contributed to the health management of their pets, which serves as a powerful motivation to actively participate in the treatment process beyond simple information consumption.

[0034] Finally, the present invention provides an interactive experience that dynamically changes the emotional tone or content composition of reports based on a 'Communion Temperature Index' calculated by comprehensively analyzing the guardian's app usage patterns and mission performance data. By activating 'hidden content (such as a secret diary)' to provide rewards to guardians who achieve a high communion temperature, or by delivering heartfelt messages in a more intimate tone, the invention offers value as an integrated healthcare platform that brings the joy of parenting to guardians and strengthens the emotional bond with their pets. Brief explanation of the drawing

[0036] FIG. 1 is a block diagram schematically illustrating the overall configuration of a pet home care guide and guardian-customized information provision system based on medical history according to one embodiment of the present invention. FIG. 2 is a flowchart illustrating the overall flow of an information provision method according to one embodiment of the present invention in chronological order. FIG. 3 is a flowchart illustrating in detail the process of determining a report template and generating content according to the number of guardian visits and the main symptom (CC) disclosure option according to one embodiment of the present invention. FIG. 4 is a conceptual diagram to explain the operating principle of the compliance filtering unit, which is a core technology of the present invention, and the structure of the substitution rule table. Figure 5 is a data flow diagram illustrating the process in which a home care challenge mission according to a diagnosis name is mapped and the results of the performance are fed back to a calendar UI in the present invention. FIG. 6 is a diagram illustrating the screen configuration of a final report provided to a guardian terminal according to the present invention. Specific details for implementing the invention

[0037] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings.

[0038] However, the present invention is not limited to the embodiments disclosed below but will be implemented in various different forms.

[0039] The embodiments described in this specification are provided to ensure that the disclosure of the invention is complete and to fully inform those skilled in the art of the scope of the invention.

[0040] And the present invention is defined only by the scope of the claims.

[0041] Accordingly, in some embodiments, well-known components, well-known operations, and well-known techniques are not specifically described to avoid the present invention being interpreted ambiguously.

[0042] Additionally, throughout the specification, the same reference numerals refer to the same components, and the terms used (mentioned) in this specification are for describing embodiments and are not intended to limit the invention.

[0043] In this specification, the singular form includes the plural form unless specifically stated otherwise in the text, and components and operations referred to as 'comprising (or comprising)' do not exclude the presence or addition of one or more other components and operations.

[0044] Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning that is commonly understood by those skilled in the art to which the present invention belongs.

[0045] Also, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless otherwise defined.

[0046] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. However, the present invention is not limited to the following embodiments, and it is obvious to those skilled in the art that various modifications are possible within the scope of the technical spirit of the present invention.

[0047] 1. Overall System Configuration

[0048] FIG. 1 is a block diagram schematically illustrating the overall configuration of a pet home care guide and guardian-customized information provision system (100) based on medical history according to one embodiment of the present invention.

[0049] Referring to FIG. 1, the information providing system (100) according to the present invention may be configured to include a data interface unit (110), a guardian profile analysis unit (120), a challenge mapping unit (130), an adaptive report generation unit (140), a compliance filtering unit (150), a report providing unit (160), a bonding index calculation unit (170), and a mission management unit (180).

[0051] 1.1 Data Interface Section

[0052] The data interface unit (110) is connected via a network with an EMR server (300) built inside the animal hospital and performs the function of collecting pet medical record data in real time at the time when treatment is completed. The data interface unit (110) can support medical data standard protocols such as REST API, HL7, and FHIR, and can adaptively select a connection method depending on the type of EMR system of the animal hospital (e.g., IntoVetGE).

[0053] The collected data may include basic information about the pet (breed, age, gender, neutering status, weight, etc.), guardian information (name, contact number, unique identifier), today's diagnosis (Diagnosis Code), prescription history (drug name, dosage, frequency of administration, duration of administration), test results (lab data such as blood tests, imaging tests, and urine tests), and medical chart data in SOAP (Subjective, Objective, Assessment, Plan) format prepared by a veterinarian.

[0054] In particular, the data interface unit (110) loads together the 'Chief Complaint (CC) disclosure option value' that the guardian has previously set through a mobile application or a hospital kiosk. This option value may be data in the form of a Boolean or a flag indicating whether the guardian wants to receive both the diagnosis name and detailed medical information (full disclosure) or wants to receive only brief information for privacy protection (simple mode).

[0056] 1.2 Guardian Profile Analysis Department

[0057] The guardian profile analysis unit (120) plays the role of establishing a report generation strategy by identifying the guardian's context based on the collected data. The guardian profile analysis unit (120) performs two main judgment logics.

[0058] First, the guardian's hospital visit history is analyzed. Specifically, the guardian's cumulative number of visits is calculated by querying past medical records stored in the relevant animal hospital's database. If the number of visits is less than a pre-set threshold (e.g., 2 times), the guardian is classified as a 'New Member' and a first template identifier (Template_ID = "NEW") is generated; if it is above the threshold, the guardian is classified as an 'Existing Member' and a second template identifier (Template_ID = "REPEAT") is generated. At this time, the number of visits may count only visits for the same pet, or the total number of visits for all pets raised by the guardian may be counted. Preferably, while the number of visits for the same pet is used as the basis, the re-visit status for the same disease is also determined, allowing for further subdivision into categories such as 'Revisit for New Disease'.

[0059] Second, the report generation mode is determined by checking the main symptom disclosure option value. If the option value is 'Public (True)', it is set to 'Full Mode', which includes the diagnosis, detailed description, and home care mission; if it is 'Private (False)', it is set to 'Lite Mode', which excludes sensitive medical information and includes only the pet's inner thoughts message. This dual-mode system contributes to ensuring flexibility in providing medical information while respecting the guardian's willingness to accept information and privacy preferences.

[0060] The guardian profile analysis unit (120) combines the two judgment results above to generate a parameter set (template identifier, generation mode, number of visits, type of pet, etc.) to be finally transmitted to the adaptive report generation unit (140).

[0062] 1.3 Challenge Mapping Section

[0063] The challenge mapping unit (130) performs the function of automatically extracting home care challenge missions optimized for the diagnosis name of the pet by referring to the challenge database (200).

[0064] The challenge database (200) is a relational or NoSQL database that stores a list of home care missions corresponding to each disease, using a disease-specific diagnosis code (e.g., a standardized disease identification code such as the Korean Standard Classification of Animal Diseases and Causes of Death (KSKCD)) as a key. Each mission is classified into four categories, such as 'Observation', 'Daily Care', 'Medication', and 'Environment', and each mission is assigned a priority weight indicating importance.

[0065] For example, if the diagnosis is 'acute gastroenteritis', the challenge database (200) may have the following mission mapped to it:

[0066]

[0067] The challenge mapping unit (130) queries the challenge database (200) using a diagnosis code extracted from EMR data as a key and primarily extracts all missions associated with the diagnosis. If multiple diagnosis names exist in a single medical treatment (e.g., gastroenteritis + dermatitis), all mission lists corresponding to each diagnosis name are extracted, and then priority weights set according to the severity or emergency level of each diagnosis name are applied to finally select the top N missions (e.g., 8 to 12). At this time, a balancing algorithm may be applied so that at least one mission is included for each of the four categories.

[0068] The selected mission list is transmitted to the adaptive report generation unit (140) along with attributes such as mission ID, mission name, category, description text, and weight.

[0070] 1.4 Adaptive Report Generation Section

[0071] The adaptive report generation unit (140) is a core component of the present invention and generates an 'intermediate report' by combining the template identifier and generation mode determined by the guardian profile analysis unit (120) and the mission list extracted by the challenge mapping unit (130).

[0072] The adaptive report generation unit (140) differentiates the content generation strategy according to the template identifier when in full public mode. When the first template identifier (new member) is received, it generates 'education-centered content (detailed descriptive text)' that provides detailed basic information about the disease, such as the definition of the disease, causes of onset, major symptoms, general treatment methods, and prognosis. This is intended to relieve the anxiety of guardians encountering the disease for the first time and to help them accurately understand the disease.

[0074] On the other hand, when a second template identifier (existing member) is received, the basic explanation is omitted, and 'trend analysis-centered content (comparative analysis type text)' is generated focusing on test values ​​changed compared to previous medical records, drug change history, and whether symptoms have improved. To this end, the adaptive report generation unit (140) internally has the following four modules:

[0075] - History data extraction module (141): Extracts values ​​of major test items (e.g., BUN, Creatinine, ALT, weight, etc.) from past N medical records (e.g., 3 to 5 times) in the form of a time series array.

[0076] - Value change calculation module (142): Compares the extracted past value with today's value to calculate the rate of increase / decrease (%) and the direction of change (increase / decrease / maintain). For example, if the previous BUN value was 30 mg / dL and today's is 25 mg / dL, it is calculated as "-16.7% decrease".

[0077] - Drug tracking module (143): Compares the previous prescription with today's prescription to track changes in the type, dosage, and frequency of administration of the drug. It identifies newly added drugs, discontinued drugs, and drugs with adjusted dosages and clearly informs the guardian.

[0078] - Visualization module (144): Generates a time series graph (line chart, etc.) visualizing the above-mentioned rate of increase / decrease data or a summary of changes in text summarizing the numerical changes. Example: "The BUN level decreased by 16.7% compared to last week and is close to the normal range."

[0080] Meanwhile, the adaptive report generation unit (140) generates a 'heart message' written from the pet's perspective, regardless of the generation mode. This is intended to provide emotional comfort to the guardian and to allow the treatment process to be perceived as an interaction activity with the pet. When generating the heart message, the following two logics are applied independently:

[0081] - Logic for determining the title: The type of pet (dog / cat) and breed information is retrieved from the title-breed mapping table (220) to select the title for the guardian. For example, titles such as "Mom" or "Dad" can be used for a dog, and "Butler" for a cat.

[0082] - Emotional Tone Adjustment Logic: For new guardians who have visited less than 2 times, a first emotional tone emphasizing gratitude and hope is applied, such as "We met for the first time today, but thank you for treating me so diligently. I want to get better quickly so we can go for a walk together!"; for existing guardians who have visited 2 times or more, a second emotional tone emphasizing intimacy and cuteness is applied, such as "Mom~ Nice to see you again! I think I've gotten much better because you took good care of my medicine!"

[0083] The adaptive report generation unit (140) generates an intermediate report containing a detailed description of the diagnosis name, information on changes in test values, prescription history, a list of home care challenge missions, inner thoughts messages, etc. through the above process and transmits it to the compliance filtering unit (150).

[0085] 1.5 Compliance Filtering Section

[0086] The compliance filtering unit (150) is a key differentiating element of the present invention and performs the function of analyzing the text of the generated intermediate report using natural language processing (NLP) technology to identify expressions with legal or medical risks and automatically replace them with safe expressions.

[0087] The compliance filtering unit (150) has a substitution rule table (210). The substitution rule table (210) is a data structure in which a target keyword (or regular expression pattern) and a corresponding replacement expression are mapped. For example:

[0088]

[0089] The compliance filtering unit (150) iterates through all text fields of the intermediate report and checks for the existence of the target keyword through regular expression matching or a keyword search algorithm. When a keyword is detected, it looks up a corresponding replacement expression in the replacement rule table (210) and replaces the original text.

[0090] In particular, when a specific drug name is detected, instead of simply displaying the name, it is replaced with a sentence explaining the drug's efficacy or mechanism of action to help guardians understand the purpose of the medication and improve medication adherence. Additionally, if expressions indicating aggression in the pet are detected, they are softened to "temporary behavior due to a pain response" to prevent guardians from developing negative perceptions of their pets.

[0091] The text, which has been legally and medically refined through processing by the compliance filtering unit (150), is completed as a ‘final report’ and delivered to the report providing unit (160).

[0093] 1.6 Report Provision Department

[0094] The report providing unit (160) performs the function of sending the final filtered report to the guardian's terminal (smartphone, tablet, PC, etc.). The report providing unit (160) can support various transmission channels such as push notifications, SMS, email, and in-app messages, and delivers the report through a preferred channel that the guardian has set in advance.

[0095] The transmitted report can be viewed through the report detail screen of the mobile application or a web browser, and guardians can expand or collapse each section within the report (diagnosis name description, test results, challenge mission, inner thoughts message, etc.) using an accordion UI, etc.

[0096] After transmission is completed, the report providing unit (160) stores the transmission history (transmission date and time, receipt confirmation status, viewing start time, viewing end time, viewing section, etc.) in the log database (230) of the server (referring to the server of the system (100). This transmission history data is used as analysis input data for the communication index calculation unit (170) to be described later.

[0098] 1.7 Mission Management Department

[0099] The mission management unit (180) plays a role in tracking and evaluating the guardian's home care activities after the report is transmitted. When the guardian indicates that a specific challenge mission has been completed on the mobile application through a checkbox or the like, the mission management unit (180) receives a signal that the mission has been completed.

[0100] The mission management department (180) refers to the category attributes of the challenge database (200) to determine which category a completed mission belongs to, and calculates a daily health management score by applying a weighting factor for each category. For example, missions in the 'medication' and 'observation' categories may be given a high first weight (e.g., 1.0) as they have a direct impact on treatment effects, and missions in the 'daily care' and 'environment management' categories may be given a relatively low second weight (e.g., 0.7).

[0101] The daily health management score can be calculated using the following formula:

[0102]

[0103] The calculated scores and mission completion history are converted into date-specific stamp images (e.g., stars, hearts, paw prints, etc.) and mapped to a calendar-style UI. Through the calendar, guardians can intuitively check how many days they have faithfully performed home care and on which dates they missed missions.

[0104] The mission management unit (180) provides this visualized treatment compliance history information to the guardian terminal (400) through the report providing unit (160), thereby inducing the guardian to objectively evaluate their care activities and feel a sense of accomplishment. In addition, the mission performance data is also used as an input variable for the empathy index calculation unit (170).

[0106] 1.8 Empathy Index Calculation Section

[0107] The emotional bond index calculation unit (170) performs the function of calculating a ‘emotional bond temperature index’ in real time to quantify the bond between the guardian and the pet. The emotional bond temperature index is calculated by comprehensively analyzing the following three variables:

[0108] (1) Report viewing time: An indicator of how thoroughly the guardian reads the report, measuring the time spent from the start time of viewing the report to the end time.

[0109] (2) Mission performance frequency: Measure the ratio (%) of missions actually completed compared to the assigned challenge missions or the number of consecutive days of performance.

[0110] (3) Hospital Visit Frequency: Evaluate the regularity of hospital visits for regular follow-up appointments or vaccinations. Higher scores are awarded for adherence to the recommended follow-up schedule.

[0111] The sympathetic temperature index can be calculated on a 0 to 100-point scale by normalizing the three aforementioned variables and taking a weighted average. For example:

[0112]

[0113] The calculated interaction temperature index is fed back to the adaptive report generation unit (140) and used to dynamically change the content configuration when generating the next report. Specifically, if the interaction temperature index exceeds a preset threshold (e.g., 80 points), the adaptive report generation unit (140) can perform the following actions:

[0114] - Action 1: Increases the expression intensity of the emotional tone control logic to generate inner feelings messages with a more intimate and affectionate tone. Example: "Mom, you're the best! Thanks to you, I'm completely healthy now!"

[0115] - Second Action: In addition to the basic report, a 'hidden content' area in the form of a diary recording the pet's daily life is activated to dynamically change the report's UI configuration. The hidden content may include storytelling-style content such as "I met another dog friend on the way to the vet today" or "It hurt a little when I got the shot, but it was okay because Mom petted me."

[0116] This reward mechanism provides guardians with the joy of raising and encourages them to maintain motivation for continuous care, ultimately strengthening the emotional bond between the pet and the guardian.

[0118] 2. Overall flow of information provision methods

[0119] FIG. 2 is a flowchart illustrating the overall flow of an information provision method according to an embodiment of the present invention in chronological order. Hereinafter, with reference to FIG. 2, a pet home care guide and a method for providing information tailored to a guardian according to the present invention will be described in detail step by step.

[0121] 2.1 EMR Data and Option Loading Step (S100)

[0122] In step S100, when the information providing system (100) detects a treatment completion signal from the animal hospital server (referring to the EMR server (300)), it loads the EMR data and main symptom (CC) disclosure option value for the corresponding treatment.

[0123] A treatment completion signal may be automatically generated when a veterinarian completes writing a treatment chart on the EMR system and clicks the 'Treatment Completed' button, or when a billing officer completes the billing process. The data interface unit (110) of the information provision system (100) detects these treatment completion signals in real time through an event listener or polling method.

[0124] When a signal indicating the completion of treatment is detected, the data interface unit (110) sends a data lookup request to the EMR server (300) using the unique identifier (treatment ID) of the corresponding treatment case. The EMR server (300) receives the request and returns the basic information of the pet, guardian information, SOAP chart, diagnosis name, prescription history, test results, etc., of the corresponding treatment in a data format such as JSON or XML.

[0125] At the same time, the data interface unit (110) accesses the guardian account database and retrieves the 'main symptom disclosure option value' that the guardian has previously set. This option value is the result of the guardian toggling the "disclosure of diagnosis name and detailed information" item to ON / OFF in the settings menu of the mobile application, and may be stored in the form of a Boolean value (True / False) or an integer flag (1 / 0).

[0126] The loaded EMR data and option values ​​are bundled into a single data package and transmitted to the guardian profile analysis unit (120).

[0128] 2.2 Context Awareness and Template Decision Step (S200)

[0129] In step S200, the guardian profile analysis unit (120) determines whether the guardian's cumulative number of visits is less than 2 times, determines one of the templates of the first mode or the second mode, and determines the level of detail of the report according to the disclosure option value.

[0130] Specifically, the guardian profile analysis unit (120) first executes a query on the hospital's medical history database using a guardian unique identifier (e.g., member ID) included in the EMR data. The query may be in the following form:

[0131]

[0132] This query calculates the number of times the guardian has previously visited with the same pet. By comparing the calculated number of visits with a pre-set threshold (e.g., 2 times):

[0133] - Visit count < 2 → Select Mode 1 (New Guardian Mode), Create 1 Template Identifier

[0134] - Visit count ≥ 2 times → Select Second Mode (Existing Guardian Mode), Generate Second Template Identifier

[0135] Next, check the main symptom disclosure option value:

[0136] - Option value = True(Public) → Set to Public mode

[0137] - Option value = False(Private) → Simple Mode setting

[0138] The guardian profile analysis unit (120) combines the two above decision results to finally generate the following set of parameters:

[0139]

[0140] This parameter set is passed to the adaptive report generation unit (140) and the challenge mapping unit (130).

[0142] 2.3 Content Creation and Report Assembly Step (S300)

[0143] In step S300, the adaptive report generation unit (140) generates differentiated content according to the determined mode, combines home care missions extracted by the challenge mapping unit (130), and assembles a draft report by applying a persona based on the type of pet and the number of visits.

[0144] First, we will explain the processing for the case where the public mode is set to "FULL".

[0145] If determined as Mode 1 (template_id = "NEW"):

[0146] - The adaptive report generation unit (140) looks up a diagnosis code from a disease information database and generates a detailed descriptive text including the definition, cause of onset, major symptoms, general treatment methods, prognosis, etc. of the disease.

[0147] - For example, if the diagnosis is "acute gastroenteritis":

[0148]

[0149] If determined to be mode 2 (template_id = "REPEAT"):

[0150] - The adaptive report generation unit (140) omits basic explanations and focuses on comparative analysis with previous medical records.

[0151] - The history data extraction module (141) retrieves past 3 to 5 medical records.

[0152] - The numeric change calculation module (142) calculates the rate of increase or decrease of the main inspection item:

[0153]

[0154] - The drug tracking module (143) identifies prescription changes:

[0155]

[0156] - The visualization module (144) generates a graph of the numerical change or provides a text summary.

[0157] Meanwhile, the challenge mapping unit (130) queries the challenge database (200) in parallel based on the diagnosis name code to extract home care missions of four categories. For example:

[0158]

[0159] Next, the adaptive report generation unit (140) generates an inner thoughts message:

[0160] - The naming logic selects a name based on the type of pet (dog / cat).

[0161] - The emotional tone adjustment logic adjusts the tone based on the number of visits:

[0162] - Number of visits < 2 (1st emotional tone):

[0163]

[0164] - Number of visits ≥ 2 times (Second emotional tone):

[0165]

[0166] In Simple Mode (mode = "LITE"):

[0167] - Generate a brief report containing only inner thoughts messages, without generating detailed diagnosis descriptions or home care missions.

[0168] The adaptive report generation unit (140) assembles all the above content elements (detailed description or comparative analysis, challenge mission, inner thoughts message) into one structured document to complete the intermediate report.

[0170] 2.4 Medical Compliance Filtering Step (S400)

[0171] In step S400, the compliance filtering unit (150) performs natural language processing on the text within the draft report to detect assertive expressions or negative prognostic words that may be legally liable, and automatically converts them into predefined safe expressions.

[0172] The processing process of the compliance filtering unit (150) is as follows:

[0173] (1) Text parsing: Extract all text fields from the intermediate report and break them down into sentences.

[0174] (2) Keyword detection: Regular expression or keyword matching is performed for each sentence. It is checked whether there is a match by comparing it with the detection target pattern defined in the substitution rule table (210).

[0175] (3) Application of substitution rules: When a keyword is detected, it is replaced with an alternative expression mapped to that keyword.

[0176] - Detected text: "The prognosis is poor and continuous observation is required."

[0177] - Text after replacement: "It requires continuous observation and management."

[0178] (4) Drug name substitution: When a specific drug name is detected, the drug information database is consulted and replaced with the description of the drug's efficacy.

[0179] - Detected text: "Administering prednisolone."

[0180] - Text after replacement: "We proceed with treatment to alleviate the inflammatory response using immunomodulatory drugs."

[0181] (5) Refinement of aggressive expressions: Refine negative expressions in descriptions of pet behavior.

[0182] - Detected text: "Exhibited aggressive behavior during treatment."

[0183] - Text after replacement: "Displayed defensive behaviors due to pain or anxiety during the examination."

[0184] (6) Final verification: Reassemble the replaced text to verify if it is natural in context. Adjust sentence connectors if necessary.

[0185] A final report is completed after processing by the compliance filtering unit (150) to ensure legal and medical safety.

[0187] 2.5 Final report transmission and storage step (S500)

[0188] In step S500, the report providing unit (160) transmits the filtered final report to the guardian terminal (400) and stores the transmission history in the server (referring to the server of the system (100)).

[0189] The report provider (160) first checks the guardian's notification settings to identify the preferred transmission channel. Transmission channels are provided in various forms, such as push notifications, SMS, email, and in-app messages.

[0190] Send a report link along with a notification message such as "Today's medical report for [Pet Name] has arrived through the selected channel!"

[0191] When a guardian clicks the notification, a mobile application or web browser launches and the report details screen loads. The report screen can be composed of the following sections:

[0192]

[0193] Each section is implemented with an accordion UI, allowing guardians to selectively expand only the information they want.

[0194] The report providing unit (160) stores the following information in the log database (230) after transmission is complete:

[0195] - Report ID, Guardian ID, Transmission Date and Time, Transmission Channel, Receipt Date and Time (Time of Notification Click), Viewing Start Date and Time, Viewing End Date and Time, List of Viewed Sections, Time Spent

[0196] These transmission history data are used as analysis input data for the communication index calculation unit (170).

[0198] 2.6 Treatment Adherence Management and Calendar Update Step (S600)

[0199] Step S600 is an optional step specified in claim 10 and can be performed after step S500. In this step, when a signal indicating the completion of the home care mission is received from the guardian terminal (400), a daily health score is calculated by reflecting the importance weight of the mission, and the daily health score and the mission execution history are updated by mapping them as visual objects to the corresponding date on the calendar UI.

[0200] The specific processing steps are as follows:

[0201] 1. Reception of mission completion signal: When a guardian selects a checkbox for a specific mission in the challenge section of the mobile app, the app sends a mission completion signal to the server (referring to the server of the system (100)).

[0202] 2. Calculation of weighted score: The mission management department (180) checks the category of the mission and applies weights for each category:

[0203] - Medication (T03), Observation (T01): Weight 1.0

[0204] - Daily Care (T02), Environmental Management (T04): Weight 0.7

[0205] Daily Health Management Score Calculation Example:

[0206]

[0207] 3. Stamp Image Generation: Convert the calculated score into a grade and select a stamp image corresponding to the grade:

[0208] - 90 points or more: Gold star, 70–89 points: Silver star, 50–69 points: Bronze star, Less than 50 points: Gray sole

[0209] 4. Calendar UI Mapping: The Mission Management Department (180) maps a stamp image and a score to the calendar cell for the corresponding date (2024.12.04):

[0210] 5. Compliance History Provided: When a guardian clicks on a specific date on the calendar, they can view the detailed mission completion history for that date and notes written by the guardian.

[0211]

[0212] 6. Calculation of consecutive performance days: The mission management department (180) can calculate the number of consecutive days the guardian has performed the mission and provide additional rewards:

[0213]

[0214] This visualized feedback encourages caregivers to objectively evaluate their care activities and feel a sense of accomplishment, and goes beyond providing one-time information to effectively manage their ongoing treatment adherence.

[0215] Additionally, the mission performance data is transmitted to the communion index calculation unit (170) and is used as a variable to determine whether to adjust the emotional tone or activate hidden content when generating the next report.

[0217] 3. Generation of adaptive reports based on guardian context

[0218] FIG. 3 is a flowchart illustrating in detail the process of determining a report template and generating content according to the number of guardian visits and the main symptom (CC) disclosure option according to one embodiment of the present invention.

[0219] Referring to FIG. 3, the guardian profile analysis unit (120) branches into four paths by combining the number of visits and CC disclosure options. For new guardians with fewer than two visits, it generates education-oriented content that provides detailed guidance on the definition, cause, and treatment methods of the disease.

[0220] New Guardian Content Example (Chronic Renal Failure Case):

[0221]

[0222] For existing guardians who have visited two or more times, basic explanations are omitted, and comparative analysis content is generated focusing on changes in test results and prescription changes compared to previous treatment.

[0223] Existing guardian content example (same pet, 3rd visit):

[0224]

[0225] If the CC disclosure option is set to private, the system operates in a simplified mode that omits diagnosis and medical information, providing only the inner thoughts message. This enables the provision of information optimized for the guardian's context and privacy preferences.

[0227] 4. Detailed Compliance Filtering Process

[0228] FIG. 4 is a conceptual diagram for explaining the operating principle of the compliance filtering unit (150), which is the core technology of the present invention, and the structure of the substitution rule table (210).

[0229] Referring to FIG. 4, the replacement rule table (210) is a data structure in which detection target keywords and replacement expressions are mapped, and includes negative prognosis words such as 'death / euthanasia', definitive expressions such as 'poor prognosis', specific drug names, expressions such as 'aggressive'. The compliance filtering unit (150) detects the text of the intermediate report through regular expression matching and automatically replaces it with a safe expression.

[0230] For example, "The prognosis is poor and may lead to death" is replaced with "Continuous observation and close consultation with a veterinarian are required," "Administering prednisolone" is replaced with "Proceeding with treatment to alleviate the inflammatory response using immunomodulatory drugs," and "aggressive behavior" is softened to "defensive behavior due to pain or anxiety." This fundamentally eliminates medical and legal risks associated with AI-generated reports.

[0232] 5. Diagnosis-based Challenge Mapping and Compliance Management

[0233] Figure 5 is a data flow diagram illustrating the process in which a home care challenge mission according to a diagnosis name is mapped and the results of the performance are fed back to a calendar UI in the present invention.

[0234] Referring to FIG. 5, the challenge mapping unit (130) queries the challenge database (200) using a diagnosis code as a key to extract missions in four categories (observation / daily care / medication / environmental management). Each mission is assigned a priority weight indicating importance, and if there are multiple diagnosis names, the top 8 to 12 missions are selected by considering the severity.

[0235] When a guardian completes a mission in the mobile app, the mission management unit (180) calculates a daily health management score by applying differential weights for each category. A weight of 1.0 is applied to the medication and observation category, and a weight of 0.7 is applied to the daily care and environment management category. For example, if 2.7 points are achieved out of a total weight of 4.8 points, the daily score becomes 56.3 points, which is converted into a gold / silver / bronze star or a paw print stamp according to the score range and mapped to the corresponding date in the calendar UI.

[0236] When a guardian clicks on a specific date on the calendar, they can view the details of the mission performed on that day and their notes; by calculating the number of consecutive days of performance and providing encouraging messages such as "3 consecutive days achieved!", it maintains motivation for continuous care.

[0238] 6. Final Report Screen Layout and Dynamic UI Based on Interaction Temperature

[0239] FIG. 6 is a diagram illustrating the screen configuration of a final report provided to a guardian terminal according to the present invention.

[0240] Referring to FIG. 6, the report screen is composed of sections such as a header (410) (pet profile), condition summary (430), test result change (431), medication reminder (432), home care challenge (440), and inner thoughts message (450), and each section is implemented as an accordion UI so that the guardian can selectively view only the information they want.

[0241] In particular, the emotional temperature index is displayed in the form of a gauge (420) at the top of the screen, and when the emotional temperature is 80 points or higher, the 'hidden content' area is activated. The hidden content (460) includes storytelling content in the form of a diary that records the pet's daily life from the pet's perspective ("I met another dog friend on the way to the vet today," "It hurt a little when I got the injection, but it was okay because Mom petted me," etc.) and provides a reward to the guardian.

[0242] The UI changes dynamically based on the bonding temperature index: encouraging messages are added between 41 and 69 points, an intimate tone and animation effects are used between 70 and 79 points, and a maximum intimate tone, hidden content is activated, and a special badge is displayed between 80 and 100 points. These dynamic changes provide immediate feedback on the caregiver's efforts and maintain long-term motivation for engagement through gamification and emotional connection.

[0244] 7. Other Examples and Variations

[0245] The above-described embodiments are merely preferred examples of the present invention, and various modifications are possible within the scope of the technical concept of the present invention.

[0246] For example, the guardian classification criteria can be adjusted from 2 visits to 3 or 5 visits, or a 3-level classification of new / intermediate / advanced can be applied instead of the dichotomous classification of new / existing. Additionally, the main symptom disclosure option can be expanded from a simple ON / OFF to multi-level options such as "full disclosure / partial disclosure / minimal disclosure."

[0247] The categories of the challenge missions are not limited to four (observation / daily care / medication / environmental management) and may include additional categories such as "exercise / rehabilitation," "nutrition / diet," and "emotional care." The variables for calculating the sympathetic temperature index may also reflect additional factors such as "number of report shares," "Q&A participation," and "feedback provision," in addition to report viewing time, mission performance frequency, and visit cycle.

[0248] Furthermore, the present invention can be applied to the management of not only companion animals but also livestock or wild animals, and can be applied in a similar manner to providing home care guides for patients after discharge at human medical institutions other than animal hospitals.

[0250] The present invention is characterized by resolving information asymmetry between medical and non-medical professionals by reprocessing complex electronic medical records (EMR) to suit the perspective and circumstances of guardians. Specifically, the system maximizes the utility of information by independently determining a guardian's visit history and selectively providing detailed educational information to novice guardians and analysis information focused on numerical changes to experienced guardians. In this process, AI-based compliance filtering technology is applied to automatically soften sensitive expressions regarding prognosis or words with potential for dispute into safe language, thereby preventing psychological trauma for guardians and preemptively blocking legal risks for the hospital.

[0251] Furthermore, going beyond simple information retrieval, the present invention induces practical treatment behaviors within the home through customized challenge missions linked to diagnoses. The guardian's mission performance history is visualized via a calendar UI, acting as a motivator to increase treatment compliance; moreover, these care activities are quantified as an 'emotional bonding index,' which dynamically changes the emotional tone and content composition of the report. Through this, the present invention goes beyond being a mere conveyor of medical information to provide value as an integrated healthcare platform that fosters a sense of accomplishment in parenting and strengthens the emotional bond with pets.

[0252] Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art to which the present invention pertains should understand that the present invention can be implemented in other specific forms without changing its technical concept or essential features, and therefore the embodiments described above should be understood as illustrative in all respects and not restrictive.

[0253] Furthermore, the scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0255] 100: A system providing pet home care guides and customized information for guardians based on medical history 110: Data Interface Section 120: Guardian Profile Analysis Department 130: Challenge Mapping Department 140: Adaptive Report Generation Unit 141: History Data Extraction Module 142: Numerical Change Calculation Module 143: Drug Tracking Module 144: Visualization Module 150: Compliance Filtering Section 160: Report Provider 170: Empathy Index Calculation Unit 180: Mission Management Department 200: Challenge Database 210: Substitution Rule Table 220: Designation-Breed Mapping Table 230: Log database 300: EMR Server 400: Guardian device S100: EMR Data and Option Loading Step S200: Context Awareness and Template Decision Step S300: Content creation and report assembly steps S400: Medical Compliance Filtering Step S500: Final report transmission and saving step S600: Treatment adherence management and calendar update steps

Claims

Claim 1 A data interface unit that collects Electronic Medical Record (EMR) data of a pet by linking with an animal hospital server; a guardian profile analysis unit that extracts profile information including the number of times the guardian visited the hospital and the type of pet based on the collected data, and determines the report generation mode as either a public mode or a simplified mode according to the set Chief Complaint (CC) disclosure option value; a challenge mapping unit that analyzes the pet's diagnosis name and extracts customized missions from a database corresponding to at least one category among observation, daily care, medication, and environmental management; an adaptive report generation unit that generates an intermediate report including a detailed description based on the diagnosis name, the customized mission, and an inner thoughts message from the pet's perspective in the case of the public mode, and generates an intermediate report including only the inner thoughts message excluding the detailed description and mission in the case of the simplified mode; and a compliance filtering unit that analyzes the text of the intermediate report to identify expressions with potential legal risks and replaces them with predefined safe expressions to generate a final report. A pet home care guide and guardian-customized information provision system based on medical history, characterized by including a report providing unit that transmits the final report to a guardian terminal. Claim 2 A medical history-based pet home care guide and guardian-customized information provision system, characterized in that, in claim 1, the guardian profile analysis unit classifies the user as a new member and generates a first template identifier when the number of visits is less than 2, and classifies the user as an existing member and generates a second template identifier when the number of visits is 2 or more, and the adaptive report generation unit generates education-centered content that provides basic information on the disease and management methods when the first template identifier is received, and generates trend analysis-centered content that provides information on numerical changes compared to previous medical records and details of drug changes when the second template identifier is received. Claim 3 In paragraph 2, the adaptive report generation unit comprises: a historical data extraction module that extracts numerical values ​​of major test items from past N medical records; a numerical change calculation module that calculates the rate of increase / decrease and the direction of change by comparing the extracted past values ​​with current values; a drug tracking module that tracks changes in the type, dosage, and number of administrations of prescribed drugs; and a visualization module that generates a time-series graph visualizing the rate of increase / decrease or change amount summary information summarizing the numerical changes in text; characterized in that it is a medical history-based pet home care guide and guardian-customized information provision system. Claim 4 In claim 1, the adaptive report generation unit is characterized by generating natural language text by independently applying: a naming determination logic that selects a guardian's name according to a predefined naming determination rule based on the type information of the pet when generating the inner thoughts message; and an emotional tone adjustment logic that selects a first emotional tone emphasizing gratitude and hope when the number of visits is less than 2, and selects a second emotional tone emphasizing intimacy and cuteness when the number of visits is 2 or more. Claim 5 A system for providing a pet home care guide and customized information for a guardian based on medical history, further comprising, in claim 4, a 'sympathy temperature index' that calculates in real time a 'sympathy temperature index' representing the bond between the guardian and the pet using the guardian's report viewing time, mission performance frequency, and hospital visit cycle as variables, wherein the adaptive report generation unit performs at least one of the following operations: a first operation in which, when the sympathy temperature index exceeds a preset threshold, the expression intensity of the emotional tone adjustment logic is increased to generate text with a more intimate tone; and a second operation in which, in addition to the basic report, a 'hidden content' area in the form of a pet's diary is activated to dynamically change the UI configuration of the report. Claim 6 In claim 1, the challenge mapping unit comprises a challenge database storing a plurality of diagnosis codes, a list of challenge missions associated with each diagnosis, and priority weights, and queries the database using the diagnosis code as a key to extract at least one mission mapped to the diagnosis among the categories of 'observation', 'daily care', 'medication', and 'environmental management', and if multiple diagnosis names exist, selects the top N missions by applying priority weights set according to the severity or urgency of each diagnosis, and combines the selected missions by category, characterized by a medical history-based pet home care guide and guardian-customized information provision system. Claim 7 In claim 6, the system further comprises a mission management unit that receives a signal of completion of a specific challenge mission among the challenge mission list from the guardian terminal, wherein the mission management unit calculates a daily health management score by referring to the category attributes of the challenge database, assigning a first weight to missions in the 'medication' and 'observation' categories and assigning a second weight lower than the first weight to missions in the 'daily care' and 'environment management' categories, converting the calculated score and mission performance history into date-based stamp images and mapping them to a calendar-type UI, and providing this as visualized treatment compliance history information through the report providing unit. Claim 8 A system for providing a pet home care guide and customized information for guardians based on medical history, wherein the compliance filtering unit comprises a substitution rule table in which a target keyword for detection is mapped to an alternative expression, and by referring to the substitution rule table, it deletes or replaces negative prognostic words such as 'death' or 'euthanasia' with comforting phrases, and when a specific drug name is detected, it searches the table for a sentence describing the efficacy or precautions of the corresponding drug and replaces it, and when an expression indicating aggression in a pet is detected, it converts it into a softened expression describing behavior caused by a pain response. Claim 9 A method for providing pet home care guides and customized information to guardians, performed by a computer system, comprising: (a) a step of loading EMR data and a main symptom (CC) disclosure option value for the corresponding treatment when a treatment completion signal is detected from an animal hospital server; (b) a context recognition and template determination step of determining whether the guardian's cumulative number of visits is less than 2 times to determine one of a templates, either a first mode or a second mode, and determining the level of detail of the report according to the disclosure option value; (c) a content generation step of assembling a draft report by generating detailed descriptive text including a definition of the disease and management method when the first mode is determined, generating comparative analysis text including data comparing numerical changes with previous treatment records when the second mode is determined, combining home care missions mapped to the diagnosis name, and applying a persona based on the type of pet and the number of visits; (d) a medical compliance filtering step of performing natural language processing on the text within the draft report to detect definitive expressions or words indicating a negative prognosis that may carry legal liability, and automatically converting them into predefined safe expressions. and (e) a step of transmitting a filtered final report to a guardian terminal and storing the transmission history on a server; characterized by a method for providing a pet home care guide and guardian-customized information based on medical history. Claim 10 In claim 9, the method for providing a pet home care guide and customized information for a guardian based on medical history is characterized by managing the guardian's continuous treatment compliance beyond one-time information provision, by further including: a step of calculating a daily health score by reflecting the importance weight of the mission when a signal of completion of the home care mission is received from the guardian terminal after step (e); and a step of updating the daily health score and mission performance history by mapping them as visual objects on the corresponding date on the calendar UI.

Images