Digital support system for personalized nutrition based on genetic, medical, and lifestyle data

Abstract

A digital system for personalized nutrition, consisting of: • A data processing module (101) that integrates genetic reports, electronic health records (EHRs) and lifestyle data; • in a natural language processing (NLP) system (102) to extract and structure data from unstructured genetic documents; • A machine learning engine that uses Latent Dirichlet Allocation (LDA)(103) to analyze genetic patterns and categorize individuals; • A predictive analysis module (104) that correlates genetic markers with lifestyle and health risks to refine dietary recommendations; • A web-based interface (105) that enables healthcare professionals to upload genetic reports, analyze categorized risk profiles, and create individualized nutrition plans; • A continuously updated nutritional database(106) that matches genetic markers with food components to provide scientifically sound recommendations; • A secure, encrypted data storage(107) and processing framework that ensures compliance with GDPR and HIPAA regulations.

Description

Field of the invention:

[0001] The present invention relates to the field of personalized nutrition, nutrigenomics, and artificial intelligence (AI)-driven health recommendations. More specifically, it is a digital platform that integrates genetic data, electronic health records, and lifestyle information to generate individualized nutritional recommendations using advanced machine learning techniques. Background of the invention:

[0002] The field of personalized nutrition and nutrigenomics has evolved significantly in recent years due to advances in genetic research, machine learning, and health technology. Nutrigenomics, a branch of science that studies the relationship between a person's genetic makeup and their response to specific nutrients, has enabled a deeper understanding of how diet influences health at the molecular level. Despite their potential, however, existing personalized nutrition technologies have several limitations.

[0003] Traditional dietary recommendations have long been based on general guidelines that apply to broad populations rather than to individuals. While these approaches are useful for general public health, they fail to account for the considerable variability in individual metabolic responses, food intolerances, and genetic predispositions. For example, a single dietary guideline may not be effective for individuals with specific genetic mutations affecting lactose intolerance, caffeine metabolism, or fat processing.

[0004] Recent advances in direct-to-consumer genetic testing have increased access to genetic insights, allowing users to understand their predisposition to diseases such as obesity, diabetes, and cardiovascular disease. Companies like 23andMe and AncestryDNA offer genetic reports that reveal basic dietary effects. However, these solutions are often static, providing a snapshot of genetic predisposition without accounting for lifestyle changes, environmental factors, or ongoing medical conditions in real time.

[0005] Machine learning and artificial intelligence have been used in digital nutrition platforms such as MyFitnessPal and Noom, which offer diet tracking and calorie counting features. However, these applications primarily focus on macronutrient and calorie intake rather than a holistic integration of genetic, medical, and lifestyle data. AI-driven recommendation systems such as IBM Watson Health have shown promise for healthcare decision support, but their application in personalized nutrition remains underdeveloped due to the challenges of integrating unstructured genetic and medical data.

[0006] Furthermore, clinical decision support systems (CDSS) in hospitals often include dietary guidelines for patients, but require manual input and lack automation in the creation of customized diet plans. These systems are typically designed for clinicians rather than consumers, limiting accessibility and usability.

[0007] The limitations of current technologies can be summarized as follows: 1. Statistical recommendations: Most genetically based nutrition platforms offer fixed recommendations that do not evolve based on new health data or lifestyle changes. 2. Limited data integration: Existing AI nutrition platforms do not integrate data from multiple sources such as genetic reports, electronic health records (EHRs), and lifestyle inputs. 3. Lack of predictive analytics: Rule-based dietary guidelines do not use predictive models to adapt in real time to a person's changing health status. 4. User accessibility issues: Many clinical nutrition systems are designed for healthcare professionals rather than consumers, making them complex and difficult to navigate. 5. Data security concerns: Direct-to-consumer genetic platforms often raise privacy concerns because sensitive health data is not always managed securely.

[0008] To address these gaps, the present invention introduces an AI-powered digital support system for personalized nutrition designed to integrate genetic, medical, and lifestyle data into an adaptive, real-time diet recommendation engine. By leveraging latent Dirichlet allocation (LDA) for data clustering, predictive modeling for health outcomes, and secure cloud-based storage, this invention overcomes the challenges of existing personalized nutrition technologies. Subject of the invention:

[0009] The primary goal of this invention is to develop a digital support system for personalized nutrition that provides precise and adaptive nutritional recommendations based on a person's genetic predisposition, medical history, and lifestyle factors. The system aims to bridge the gap between nutrigenomics research and the practical implementation of dietary guidelines by integrating diverse data sources and leveraging advanced AI-driven analytics to generate dynamic, scientifically sound nutritional guidance.

[0010] Another goal is the efficient processing of unstructured genetic and medical data using natural language processing (NLP) and machine learning models to enable structured extraction and analysis of key genetic indicators. By leveraging latent Dirichlet allocation (LDA) and predictive analytics, the invention aims to identify hidden genetic patterns and their relationships with nutritional and metabolic functions to provide a more comprehensive and precise approach to personalized nutrition.

[0011] The invention also aims to facilitate secure and user-friendly access to personalized nutrition plans via a web-based interface that allows healthcare professionals to upload genetic reports, review risk profiles, and create customized nutrition plans. By integrating real-time lifestyle inputs, including physical activity and dietary habits, the system continuously refines recommendations, ensuring that nutritional advice remains relevant and adaptable to changing health conditions.

[0012] Additionally, the invention focuses on the development of a continuously growing nutrition database that matches genetic markers with specific nutritional components to ensure users receive evidence-based meal suggestions tailored to their unique metabolic and health profiles. The system prioritizes privacy and security by implementing encryption protocols and ensuring compliance with GDPR and HIPAA regulations, thus protecting sensitive genetic and medical information while enabling effective, personalized nutrition decision-making.

[0013] By achieving these goals, the invention provides a scalable, customizable, and clinically relevant platform that enhances personalized nutrition, preventive healthcare, and clinical decision support, ultimately improving health outcomes and disease prevention through precise nutritional interventions. Summary of the invention:

[0014] The invention introduces an AI-powered digital platform for personalized nutrition that overcomes these limitations by: 1. Integration of genetic, health, and lifestyle data: The system consolidates genetic markers, electronic health records (EHRs), and real-time lifestyle data, including dietary intake, activity levels, and microbiome data. 2. AI and machine learning-powered adaptive recommendations: The platform uses Latent Dirichlet Allocation (LDA) and predictive analytics to continuously refine dietary suggestions based on evolving health data. 3. Automated data collection and processing: The system uses natural language processing (NLP) techniques to extract structured data from unstructured genetic reports in PDF and DOCX formats. 4. Personalized nutrition recommendations: AI-driven algorithms analyze genetic predispositions, metabolic pathways, and disease risks to create customized nutrition plans. 5. Secure and privacy-oriented data management: The platform ensures encrypted storage, GDPR / HIPAA-compliant data processing, and secure web-based access for healthcare professionals.

[0015] The platform integrates multiple data sources, including genetic reports (oncology, cardiology, and nutrigenomics), electronic health records, and lifestyle data. The system processes unstructured genetic data by converting PDF and DOCX formats into structured datasets. The processed genetic data is then subjected to LDA, which identifies hidden patterns in the genetic and lifestyle data to categorize patients into subgroups.

[0016] Using this categorized data, the system creates personalized nutritional recommendations. The AI ​​model dynamically updates the nutritional suggestions as new data is integrated, ensuring real-time adaptation. The system's web-based interface allows healthcare providers to securely upload genetic reports, review risk profiles, and adjust meal plans based on scientific evidence.

[0017] Additionally, a proprietary nutrition database maps genetic markers to nutritional needs. Patients with specific genetic predispositions, such as lactose intolerance or cardiovascular risk, receive tailored nutritional advice. The system's learning mechanism continuously refines recommendations based on patient results. Short description of the drawing Fig. . shows a block diagram of the system according to the invention. Detailed description of the invention

[0018] The digital platform described in this invention is designed to bridge the gap between nutrigenomics and practical nutritional interventions through advanced data integration, AI-driven insights, and user-centric applications. This is achieved by collecting, analyzing, and processing data from multiple sources, including genetic profiles, lifestyle habits, and electronic health records (EHRs). Unlike traditional nutrition systems based on generalized recommendations, this platform dynamically updates its nutritional suggestions based on real-time user data, providing highly customized nutritional advice tailored to each individual's genetic predispositions and health conditions.

[0019] One of the key features of this invention is its ability to extract and structure unstructured genetic data. Genetic reports from various fields, such as oncology, cardiology, and nutrigenomics, are often presented in various formats, including PDFs and DOCX documents. The platform leverages advanced natural language processing (NLP) and machine learning methods to extract relevant data points such as gene names, RSIDs, and associated health implications, and then structure them into a format suitable for further analysis. The extracted data is processed using Latent Dirichlet Allocation (LDA), which identifies hidden patterns within the genetic dataset and classifies individuals into subgroups based on shared biological markers.

[0020] This structured genetic data is then combined with user-reported lifestyle factors, including dietary habits, physical activity levels, and sleep patterns. Integration with electronic health records (EHRs) further improves the accuracy of recommendations by incorporating medical histories, lab results, and physician insights. The system applies predictive analytics to identify correlations between genetic markers, lifestyle behaviors, and potential health risks, enabling the creation of personalized nutrition plans aimed at mitigating disease risk and optimizing overall well-being.

[0021] A key advantage of this platform is its AI-driven recommendation engine, which continuously refines dietary suggestions based on new data inputs. Unlike traditional genetic testing services that provide static recommendations, this system adapts dynamically, ensuring users receive the most up-to-date and relevant nutritional advice. For example, if a user's current health data indicates a change in blood sugar or cholesterol levels, the system will adjust dietary recommendations accordingly to effectively manage these parameters.

[0022] To facilitate user interaction and accessibility, the platform offers a secure web-based interface. This interface allows healthcare professionals, including nutritionists and physicians, to upload genetic reports, review categorized risk profiles, and create customized nutrition plans based on scientific research. Patients can also access their personalized nutrition plans via a mobile-friendly dashboard, where they receive nutritional recommendations, meal plan suggestions, and notifications about necessary dietary adjustments based on ongoing health monitoring.

[0023] The platform's food database serves as a critical component for creating accurate and scientifically sound nutrition recommendations. The database contains a comprehensive list of foods, each labeled with its corresponding nutrient composition and health effects. Foods are matched to a user's genetic profile to ensure that suggested dietary choices align with their unique metabolic needs. For example, individuals with a genetic predisposition to lactose intolerance receive dairy-free alternatives, while those with variations in their lipid metabolism genes receive dietary suggestions that optimize fat intake for heart health.

[0024] Security and data protection are paramount in the design of this invention. Given the sensitivity of genetic and health data, the platform integrates advanced encryption techniques, secure data storage solutions, and access control measures that comply with international regulations such as GDPR and HIPAA. Data anonymization techniques are also employed to ensure that personal information remains protected while enabling meaningful data analysis and insight generation.

[0025] The digital system for personalized nutrition works by integrating genetic reports, electronic health records (EHRs), and lifestyle data into a unified, AI-driven platform. At the core of the system is a data processing module that collects and organizes information from various sources to ensure a comprehensive understanding of a person's health profile. This module works by extracting relevant details from genetic reports, including gene names, RSIDs, and associated health markers, and structuring them into a format suitable for analysis.

[0026] To process unstructured genetic and medical documents, the system uses a natural language processing (NLP) engine. This component scans PDFs and DOCX files, identifies key genetic indicators and their associated medical implications. The NLP engine improves the efficiency and accuracy of data capture by detecting specific genetic markers and mapping them to known health conditions and nutritional needs.

[0027] Once the data is extracted and structured, a machine learning engine using Latent Dirichlet Allocation (LDA) analyzes genetic patterns and categorizes individuals into relevant subgroups. By applying LDA, the system identifies latent genetic associations and groups users based on their genetic predispositions. These clusters provide insights into shared metabolic traits, disease susceptibilities, and nutritional needs, forming the basis for personalized nutrition recommendations.

[0028] Following genetic classification, a predictive analytics module correlates genetic markers with lifestyle behaviors and health risks. This module evaluates real-time data such as physical activity levels, dietary habits, and recent medical test results to dynamically refine recommendations. By continuously updating based on new data, the system ensures that nutritional advice remains relevant and adaptable to the user's evolving health status.

[0029] A secure web-based interface serves as the user-facing component of the platform. Healthcare professionals, including dietitians and physicians, can use this interface to upload genetic reports, analyze categorized risk profiles, and create individualized nutrition plans. The system offers interactive visualizations that highlight risk factors and dietary recommendations tailored to each patient's genetic predisposition and lifestyle.

[0030] To support scientifically sound nutritional advice, the system includes a continuously updated nutrient database. This database stores detailed information on foods linked to genetic markers and metabolic pathways. The system uses this database to generate meal suggestions that match the user's genetic needs, thus ensuring a precise and personalized approach to nutrition.

[0031] A critical component of the system is its secure data storage and processing framework, which ensures compliance with GDPR and HIPAA regulations. Encryption protocols protect sensitive genetic and medical data, while anonymization techniques preserve privacy without compromising analytical accuracy. The system's robust security measures protect patient data while enabling effective data-driven decision-making in personalized nutrition.

[0032] By integrating these components, the invention provides a holistic and adaptive digital platform for personalized nutrition. It goes beyond static genetic reports and traditional dietary recommendations by dynamically updating nutrition plans based on real-time health data, insights from machine learning, and predictive modeling. Combining AI-driven analytics, secure data management, and a user-friendly clinical interface, this system sets a new standard in precision nutrition and preventative healthcare.

[0033] In summary, this invention offers a transformative approach to personalized nutrition by integrating data from multiple sources, applying advanced AI-driven analytics, and providing dynamic, user-centered nutritional recommendations. It goes beyond traditional, static nutritional advice and creates a continuously evolving, evidence-based nutrition system tailored to individual genetic and health profiles. By combining AI-driven personalization, secure data management, and a clinically validated nutrition database, this invention sets a new standard in precision nutrition and preventative healthcare. List of reference symbols 100 systems 101 Data processing module 102 Natural Language Processing (NLP) System 103 Latent Dirichlet Allocation (LDA) 104 Predictive Analytics Module 105 web-based interface 106 Nutrition database 107 encrypted data storage

Claims

[1] A digital system for personalized nutrition, consisting of: • A data processing module (101) that integrates genetic reports, electronic health records (EHRs) and lifestyle data; • in Natural Language Processing (NLP) systems(102) to extract and structure data from unstructured genetic documents; • A machine learning engine that uses Latent Dirichlet Allocation (LDA)(103) to analyze genetic patterns and categorize individuals; • A predictive analysis module (104) that correlates genetic markers with lifestyle and health risks to refine dietary recommendations; • A web-based interface (105) that allows health professionals to upload genetic reports, analyze categorized risk profiles, and create personalized nutrition plans; • A continuously updated nutritional database(106) that matches genetic markers with dietary components to provide scientifically sound recommendations; • A secure, encrypted data storage(107) and processing framework that ensures compliance with GDPR and HIPAA regulations. [2] The system of claim 1, wherein the NLP module extracts genetic data, including gene names, RSIDs, and associated health implications, from PDF and DOCX formats. [3] The system of claim 1, wherein the machine learning engine dynamically updates the nutritional recommendations based on evolving user health data and predictive risk modeling. [4] The system of claim 1, wherein the web-based interface provides interactive visualizations of genetic risks, nutritional recommendations, and predictive health outcomes. [5] The system of claim 1, wherein the nutrient database is continuously expanded based on clinical research, user feedback, and machine learning insights. [6] The system of claim 1, wherein a privacy-preserving architecture ensures anonymization of genetic and lifestyle data while maintaining analytical accuracy.

Images