Digital Therapeutic Message Screening With Risk Feedback Loops

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital therapeutic content generation systems face challenges with accuracy, relevancy, and specificity due to manual creation difficulties, resource-intensive processes, and inaccuracies in AI-generated content, leading to ineffective message provision that can worsen user adherence and condition degradation.

Innovation Solution

An interconnected architecture using multiple risk models and generative AI models to evaluate, validate, and iteratively train content, ensuring relevance and accuracy by incorporating feedback from administrators and end users, thereby optimizing resource consumption and improving user interaction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual content creation is used for digital therapeutics, then content accuracy and specificity can be maintained, but the process becomes tedious, time-consuming, and resource-intensive

Engineering Contradiction:

Improvecontent accuracyVSAvoidcontent generation efficiency

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The content creation process is divided into distinct modules: AI-generated draft content, risk evaluation by specialized models, administrator review, and iterative refinement. This segmentation allows automated efficiency at the generation stage while maintaining human oversight for accuracy-critical evaluation and review stages.

Inventive Principle:

Principle #1Segmentation

2Productivity

If AI techniques are used for automated content generation, then productivity and scalability are improved, but accuracy, relevancy, and specificity deteriorate due to hallucinations and inaccurate subject matter

Engineering Contradiction:

Improvecontent generation scalabilityVSAvoidcontent accuracy

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The system implements a feedback loop where risk evaluation results from specialized models are fed back to administrators for review, and administrator corrections are used to iteratively refine both the generated content and the underlying AI models. This continuous feedback mechanism improves accuracy while maintaining automated scalability.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

Specialized risk evaluation models act as intermediaries between the AI content generation system and the final content delivery to users. These intermediary models assess content for accuracy, relevancy, and safety before it reaches the end user, filtering out harmful or inaccurate information while preserving the scalability of automated generation.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If multiple AI models are used to generate content items, then content diversity and coverage are improved, but the complexity of parsing and validating content increases significantly

Engineering Contradiction:

Improvecontent coverageVSAvoidcontent validation complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The validation system is segmented into multiple specialized risk evaluation models, each designed to assess specific aspects of content from different sources. This segmentation allows the system to handle diverse content types efficiently by applying appropriate evaluation criteria to each content item regardless of its source.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The risk evaluation framework is designed as a universal system that can assess content from multiple AI models and human creators using the same set of risk models. This multi-functional approach simplifies validation complexity by providing a unified evaluation mechanism that works across diverse content sources.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Loss of energy

If content is provided without proper regulation, then resource consumption is reduced, but user adherence deteriorates and condition degradation occurs

Engineering Contradiction:

Improvecomputing resource consumptionVSAvoiduser adherence

Core Design Contradiction:

Loss of energyVSReliability

Solution Approach 1:

The system performs preliminary risk evaluation and content regulation before delivering content to users. By pre-assessing content accuracy and safety using specialized risk models, the system ensures that only appropriate content is provided, maintaining user adherence and preventing condition degradation while avoiding the need for extensive post-delivery monitoring.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12537083B2Systems and methods for regulating provision of messages with content from disparate sources based on risk and feedback data

Publication Date: 2026.01.27 CLICK THERAPEUTICS INC

AI summary

Aspects of the present disclosure are directed to systems, methods, and computer readable media for configuring generation of digital therapeutic content for provision. A computing system may identify content to be provided via a network. The computing system may apply the content to a machine learning (ML) model to generate an output. The computing system may determine, from the output, a compliance status of the content. The computing system may identify, based on applying the ML model, at least a subsection of the content to be modified. The computing system may identify the content to be modified responsive to determining the compliance status of the content.