JMIR Publications

ABSTRACT

Background:

Pedigree mapping serves as the first step in genetic counseling. Traditional methods relying on manual pedigree drawing and information collection are inefficient and prone to errors.

Objective:

This study aims to design an intelligent pedigree mapping system based on a specialized large-scale model, intended to automate family history collection and standardized pedigree map generation through conversational interaction, thereby enhancing clinical efficiency of genetic counseling.

Methods:

This study adopted a participatory design and iterative development approach. We formed an interdisciplinary team comprising gynecologic oncologists, genetic counselors, AI engineers, and patient representatives. The system is built around a proprietary, specialized large language model for gynecologic oncology, featuring a micro-services architecture that includes pedigree data collection. Through four iterative cycles, we integrated specialized knowledge bases using Retrieval-Augmented Generation (RAG) technology and applied a novel "Graph-Transformer" algorithm to handle complex kinship relationships. The system's performance was comprehensively evaluated based on the accuracy of pedigree map generation and user feedback on usability.

Results:

We have significantly improved system performance by optimizing dialogue strategies, expanding the knowledge base, and applying self-developed algorithms to improve data extraction accuracy. The initial database of the system defines 100 core information nodes and matches about 800 user expression examples, while the final pilot version of the database is expanded to 245 core information nodes and 2256 user expression examples. Although the AI function is limited in the early stage, the final version of the system has an accuracy rate of more than 95% in generating genealogy nodes and relationships. Test users generally gave positive reviews of the system's innovation, ease of use, and intuitiveness of information presentation, believing it to have great potential in clinical practice.

Conclusions:

The developed system is an accurate, efficient, and clinically well-received auxiliary tool. After sufficient clinical validation, it holds promise for integration into routine clinical workflows to optimize genetic counseling services. Clinical Trial: Not applicable.