Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 8, 2026npj Precision OncologyOpen Access

Prognostic data extraction harnessing a privacy-preserving large language model: a clinician-AI collaborative retrospective evaluation in head and neck oncology

View Full Paper
Ask AI
Bookmark
Share

Authors

YZYujing ZouMcGill UniversityLSLaya Rafiee SevyeriMcGill UniversityFFFarhood FarahnakConcordia University

Discussion

Loading...

Member takes

Implication

Randomized trial evaluates prognostic variables in head and neck cancer, indicating effective clinician-AI collaboration.

Key Points

  • The study aims to assess the efficacy of large language models in extracting prognostic variables from head and neck cancer reports.
  • Evaluated Llama3.3-70B and seven other models on 1,360 reports (882 patients) for extracting prognostic data.
  • Conducted a majority-vote reference review by three radiation oncologists on a stratified 50-case subset.
  • Integrated extracted HPV status and comorbidity scores into a Cox Proportional Hazards model for survival analysis.
  • Llama3.3-70B achieved an F1 score of 98.6% with high clinician agreement.
  • Incorporation of LLM-extracted data significantly improved disease-free survival (p = 0.014, ΔC-index + 0.071).
  • Locoregional failure-free survival also improved (p = 0.026, ΔC-index + 0.108) with internal validation.

Cite This Study

Zou et al. (2026) studied this question.

synapsesocial.com/papers/6a265c42ad53cfb9357c588dhttps://doi.org/10.1038/s41698-026-01521-y
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Large language models enable prognostic stratification of cancer patients using real-world clinical notes2026
  2. 2An External Validation Study on Two Pre-Trained Large Language Models for Multimodal Prognostication in Laryngeal and Hypopharyngeal Cancer: Integrating Clinical, Treatment, and Radiomic Data to Predict Survival Outcomes with Interpretable Reasoning2025 · 2 citations
  3. 3Large Language Models Improve Cancer Survival Prediction Using Real-World Clinical Notes2025
  4. 4SCRIPT: Stratified clinical risk prediction from pathology reports using large language models2026
  5. 5A Scalable Method for Validated Data Extraction from Electronic Health Records with Large Language Models2026