Customized Health Messaging Using RAG Context and Generative AI

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional exercise systems provide inflexible health recommendations that do not account for individual user lifestyle, attitude, or motivational styles, and exercise program libraries are cumbersome and inefficient for users to navigate.

Innovation Solution

A system utilizing a retrieval augmented generation (RAG) node to retrieve relevant historical health data, generate a contextual prompt based on user data, and provide it to a generative AI model (LLM) to create customized exercise and motivational messages, enhancing accuracy and efficiency by leveraging up-to-date information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If pre-determined health recommendations are provided to users, then the system can deliver standardized exercise and diet plans, but the recommendations become inflexible and inapplicable to individual user lifestyles and motivational styles

Engineering Contradiction:

Improveease of providing health recommendationsVSAvoidadaptability to user lifestyle and motivational style

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary actions by collecting and analyzing user data (lifestyle, attitude, motivational style) before generating health recommendations. This allows the system to pre-process individual user characteristics and use them to customize recommendations, resolving the contradiction between ease of provision and adaptability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system applies local quality by tailoring health recommendations to specific user characteristics rather than applying uniform pre-determined plans. Each user receives customized exercise and diet recommendations based on their individual profile, making the system both easy to operate and highly adaptable.

Inventive Principle:

Principle #3Local quality

2Quantity of substance

If exercise program libraries store massive amounts of exercise programs, then the system can provide comprehensive exercise options, but identifying and searching for relevant programs becomes difficult for users

Engineering Contradiction:

Improvequantity of exercise programsVSAvoidease of identifying and searching exercise programs

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The system uses feedback mechanisms by analyzing user data and automatically matching users with relevant exercise programs from the library. Instead of requiring users to manually search through massive program libraries, the system provides feedback-based recommendations tailored to user profiles, maintaining comprehensive program availability while improving ease of access.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system extracts relevant exercise programs from the massive library based on user characteristics and requirements. By selecting and presenting only the most relevant programs rather than displaying the entire library, the system maintains comprehensive options while making identification and selection easy for users.

Inventive Principle:

Principle #2Taking out (Extraction)

3Stability of the object's composition

If conventional systems use static training data for LLM models, then the model structure remains stable, but the recommendations become outdated and less accurate for current user needs

Engineering Contradiction:

Improvestability of model structureVSAvoidaccuracy of health recommendations

Core Design Contradiction:

Stability of the object's compositionVSMeasurement precision

Solution Approach 1:

The system implements dynamics by continuously updating user data and re-generating health recommendations based on current user status and historical data. While the LLM model structure remains stable, the input data and generated recommendations are dynamic and continuously refreshed, resolving the contradiction between structural stability and recommendation accuracy.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system applies parameter changes by updating user profile parameters (lifestyle, attitude, motivational style) and using these changed parameters to generate new recommendations. This allows the system to maintain a stable model structure while adapting recommendations to current user needs through parameter updates rather than model retraining.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250253026A1Devices, systems, and methods to create customized messages based on contextual information

Publication Date: 2025.08.07 IFIT INC

AI summary

A method for creating a customized message to a user is provided. The method includes receiving a request to create a customized message including user data. The method further includes retrieving, historical health data that correlate with the user data. The method further includes generating a contextual prompt for a model wherein the contextual prompt is based on the user data and the historical health data that correlates with the user data, and providing the contextual prompt to the model, wherein the contextual prompt causes the model to create the customized message.