Objective:

To develop and pilot-validate an AI-assisted pedigree mapping system based on a gynecologic oncology specialty large language model, intended to support genetic counseling by extracting personal and family history and generating standardized pedigree charts.

Approach:

• Participatory development: Gynecologic oncologists, genetic counselors, clinical research nurses, patient representatives, AI engineers, and medical education specialists contributed to iterative system design.
• Dialogue and pedigree generation: The system used guided dialogue and natural language processing to identify kinship, disease, and age information, then generated and updated visual pedigrees according to genetic standards.
• Interface and privacy: A mobile application and microservices architecture supported communication between users and system components. Sensitive information was encrypted during transmission, and conversation data were anonymized with access restricted to authorized administrators.

Key Findings:

• The system was designed to extract structured personal and family history from multiturn dialogue and generate visual pedigree charts in real time.
• The provided context describes the system’s development and intended functions but does not report pilot-validation results or quantitative performance measures.

Interpretation:

The article presents a participatory, specialty-focused AI system intended to assist pedigree collection and charting during genetic counseling for hereditary gynecologic tumors.

Limitations:

• The supplied article context ends during the methods section and does not include the pilot-validation procedures, results, or discussion.
• No sample size, accuracy estimates, usability outcomes, or comparison with standard pedigree mapping are provided in the supplied context.