PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
May 7, 2026PLoS ONEOpen Access

Interpretable miRNA-based prediction model for early detection of pancreatic cancer: Development and cross-platform validation

View Full Paper
Ask AI
Bookmark
Share

Authors

ZYZhu YLZLinglin ZhuYLYumei Liu

Discussion

Loading...

Member takes

Overview

Diagnostic model shows strong validation in pancreatic cancer, indicating potential for clinical application.

Key Points

  • To develop and validate an interpretable diagnostic model for early detection of pancreatic cancer using a 20-miRNA signature.
  • Developed a predictive model based on a 20-miRNA signature using public datasets.
  • Included 801 samples for training and validation, with 767 used in model development.
  • Applied Random Forest classifier and assessed interpretability using SHAP analysis.
  • Evaluated diagnostic performance through cross-validation and independent external validation.
  • Achieved a cross-validation AUC of 0.87 with sensitivity of 84.7% and specificity of 83.1%.
  • External validation showed AUC values ranging from 0.78 to 0.83 across independent datasets.
  • Model performance was consistent across various sample types and platforms.

Cite This Study

Y et al. (2026) studied this question.

synapsesocial.com/papers/69fbe2b3164b5133a91a225bhttps://doi.org/10.1371/journal.pone.0348699
View Full Paper
Ask AI
Bookmark
Share