IVD Kit Proposal Generation Using Language Models for Literature Overload

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge of generating research and development proposals for in-vitro diagnostic kits is exacerbated by the vast number of biological papers published annually, making it impractical for companies to review all relevant documents manually, especially with the rapid pace of technological updates.

Innovation Solution

A server-based method using a pre-trained language model, such as one trained by RLHF, to automatically generate a research and development proposal by integrating intended use information with technical, resource, design, regulatory, and performance specifications, leveraging documents like papers, patents, and marketability surveys.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If manual review of biological papers is performed to generate R&D proposals, then comprehensive information can be obtained, but the time and resources required become prohibitively large

Engineering Contradiction:
Improvecomprehensive informationVSAvoidtime required
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

A language model serves as an intermediary between the vast corpus of biological papers and the R&D proposal generation process. The model automatically processes and synthesizes information from multiple papers, extracting relevant technical specifications, resource requirements, design parameters, regulatory considerations, and performance metrics without requiring manual review of each document.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The manual mechanical process of reviewing papers is replaced with an automated computational system. The language model processes textual information from numerous papers simultaneously, generating comprehensive R&D proposals in a fraction of the time required for manual review while maintaining or improving information completeness.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If keyword-based paper selection is used to reduce review workload, then fewer papers need to be reviewed, but the selected papers may not reflect the newest technology

Engineering Contradiction:
Improvereview efficiencyVSAvoidtechnology currency
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The language model is pre-trained on extensive biological literature, enabling it to recognize and prioritize the newest and most relevant technologies when generating R&D proposals. This preliminary training ensures that even when processing a limited set of papers, the model can identify cutting-edge developments and incorporate them appropriately.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback mechanisms where the language model continuously learns from new publications and updates its knowledge base. This allows the model to maintain awareness of the latest technological developments and reflect them in generated proposals, ensuring technology currency even with reduced paper review volumes.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If detailed manual review of all papers is conducted, then accurate and complete proposals can be generated, but the process becomes too complex and resource-intensive

Engineering Contradiction:
Improveproposal accuracyVSAvoidprocess complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The language model performs self-service by automatically gathering, analyzing, and synthesizing information from multiple biological papers to generate complete R&D proposals. The model independently identifies technical specifications, resource requirements, design parameters, regulatory considerations, and performance metrics without requiring complex manual review processes or extensive human intervention.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The language model serves multiple functions simultaneously: it acts as a literature reviewer, data extractor, analyst, and proposal writer. This multi-functionality consolidates what would otherwise require multiple separate processes and experts into a single automated system, reducing overall process complexity while maintaining comprehensive coverage.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4700783A1Method for automatically generating proposal for development for in vitro diagnostic kit, and server therefor
Publication Date: 2026.02.25 SEEGENE INC
  • EP4700783A1 patent drawingFigure 1
  • EP4700783A1 patent drawingFigure 2
  • EP4700783A1 patent drawingFigure 3

AI summary

A method for automatically generating a research and development proposal for an in vitro diagnostic kit, performed by a server for automatically generating the research and development proposal, includes: acquiring an initial proposal including intended use information for the kit; providing the acquired initial proposal to a pre-trained language model; acquiring a specification (spec) for the kit from the language model receiving the initial proposal, wherein the acquired specification for the kit includes at least one of a technical specification to be used for the research and development of the kit, a resource specification for the research and development, a design specification for the kit, a regulatory approval specification that is needed to meet for approval, and a performance specification for the kit; and controlling to generate the research and development proposal for the kit including the acquired specification and the intended use information for the kit.