Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A Bilayer Feature Fusion Framework for Pan-Cancer Survival Prediction Based on Multihead Attention and Adaptive

Yun Chen1, Zhifang Deng1, Lili Wang1

  • 1School of Medical Information and Engineering, Xuzhou Medical University, Xuzhou, China.

JMIR Medical Informatics
|March 30, 2026
PubMed
Summary

Related Concept Videos

Cancer Survival Analysis01:21

Cancer Survival Analysis

Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Secure High-Order Gene Interaction Detecting Method for Infectious Diseases.

Computational and mathematical methods in medicine·2022
See all related articles

This study introduces a novel framework for pan-cancer survival prediction, effectively balancing high accuracy with robust privacy protection for sensitive medical data. The method enhances individualized oncology by integrating multimodal data while safeguarding patient information.

Area of Science:

  • Precision Medicine
  • Computational Oncology
  • Bioinformatics

Background:

  • Pan-cancer survival prediction is vital for personalized cancer treatment.
  • Multimodal data fusion improves prediction accuracy but often neglects data sensitivity and privacy.
  • Existing methods lack a balance between precise feature extraction and secure data handling.

Purpose of the Study:

  • To propose a bilayer feature fusion framework utilizing multi-head attention and adaptive differential privacy.
  • To achieve a balance between accurate feature extraction and sensitive medical data protection.
  • To enhance pan-cancer survival prediction while ensuring patient privacy.

Main Methods:

  • Integrated multi-head attention mechanism for bilayer feature extraction and fusion.
Keywords:
adaptive differential privacymultihead attentionpan-cancer survival prediction

Related Experiment Videos

  • Employed layer-wise relevance analysis to guide adaptive Laplacian noise injection for differential privacy.
  • Validated performance using concordance index (C-index) and 5-fold cross-validation against state-of-the-art methods.

    • Main Results:

    • Achieved superior pan-cancer and single-cancer survival prediction performance.
    • The trimodal combination (clinical, mRNA, microRNA) yielded the highest pan-cancer C-index (0.799).
    • Privacy protection reduced accuracy minimally (0.01-0.03) while maintaining ϵ-differential privacy; outperformed deep learning models in 18/20 cancer types.

    Conclusions:

    • The proposed framework offers a viable solution for accurate and privacy-preserving pan-cancer survival prediction.
    • Establishes a foundation for secure utilization of medical data in oncology.
    • Future work can incorporate pathological images and proteomics for expanded applications.