Custom GPT models for complex rheumatology systematic reviews: A two-part evaluation of data extraction and prognosis appraisal - Takeaways - MDSpire

Custom GPT models for complex rheumatology systematic reviews: A two-part evaluation of data extraction and prognosis appraisal

  • By

  • Pamela Munguía-Realpozo

  • Edith Ramírez-Lara

  • Claudia Mendoza-Pinto

  • Ivet Etchegaray-Morales

  • Juan Carlos Solis-Poblano

  • Marco Alejandro Trinidad-González

  • Jorge Ayón-Aguilar

  • Máximo Alejandro García-Flores

  • Álvaro José Montiel-Jarquín

  • July 9, 2026

Share

  • 1

    Systematic reviews are essential for evidence-based decision-making in rheumatology but are resource-intensive, particularly in data extraction and risk-of-bias appraisal.

  • 2

    Large language models (LLMs) like customized GPTs can streamline systematic review workflows by processing full-text articles and generating structured outputs.

  • 3

    This study evaluates the performance of customized GPT models against human reviewers in data extraction and QUIPS-based risk-of-bias assessment in rheumatology.

  • 4

    The evaluation involved a retrospective analysis of studies from two PROSPERO-registered systematic reviews focusing on metabolomics in systemic lupus erythematosus.

  • 5

    Customized GPTs were developed using the OpenAI ChatGPT platform and tailored with task-specific instructions for systematic review tasks.

Original Source(s)

Related Content