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

Bi-level multi-source learning for heterogeneous block-wise missing data.

Shuo Xiang1, Lei Yuan1, Wei Fan2

  • 1School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, AZ, USA; Center for Evolutionary Medicine and Informatics, The Biodesign Institute, Arizona State University, Tempe, AZ, USA.

Neuroimage
|August 31, 2013
PubMed
Summary

Related Concept Videos

Associative Learning01:27

Associative Learning

Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...

You might also read

Related Articles

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

Sort by
Same author

Author Correction: UKB-MDRMF: a multi-disease risk and multimorbidity framework based on UK biobank data.

Nature communications·2026
Same author

MyESL: A Software for Evolutionary Sparse Learning in Molecular Phylogenetics and Genomics.

Molecular biology and evolution·2025
Same author

Rhizosphere microbial diversity and functional roles in tea cultivars: insights from high-throughput sequencing and functional isolates.

Plant signaling & behavior·2025
Same author

LLaFS++: Few-Shot Image Segmentation With Large Language Models.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

CATI: A medical context-enhanced framework for diagnosis code assignment in the UK Biobank study.

Artificial intelligence in medicine·2025
Same author

UKB-MDRMF: a multi-disease risk and multimorbidity framework based on UK biobank data.

Nature communications·2025

This study introduces a novel bi-level learning model to integrate diverse Alzheimer's Disease data, even with missing information. The method effectively combines multiple data sources for improved analysis without data imputation.

Area of Science:

  • Biomedical data science
  • Computational neuroscience
  • Genomics and proteomics

Background:

  • Bio-imaging generates high-dimensional, heterogeneous data crucial for biomedical applications like Alzheimer's Disease (AD) research.
  • Integrating multiple data sources (neuroimaging, genetics, proteomics) enhances predictive power but requires effective feature selection and handling of missing data blocks.
  • The Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset exemplifies block-wise missingness, with varying availability of MRI, FDG-PET, CSF, and proteomic data.

Purpose of the Study:

  • To develop a unified bi-level learning model for integrating multiple heterogeneous data sources in the presence of block-wise missing data.
  • To address the challenges of feature pruning and data source selection for interpretable models in high-dimensional biomedical data.
  • To present a novel approach that avoids data imputation for improved performance and generalizability.
Keywords:
Alzheimer's diseaseBlock-wise missing dataMulti-modal fusionMulti-sourceOptimization

Related Experiment Videos

Main Methods:

  • A unified bi-level learning framework is proposed, unifying feature-level and source-level analysis.
  • The model is extended to handle block-wise missing data without imputation, generalizing to various applications.
  • Efficient optimization algorithms are developed for both complete and incomplete multi-source data modeling.

Main Results:

  • The proposed models demonstrate superior performance compared to existing approaches on incomplete multi-source data.
  • The unified model incorporates various feature learning methods as special cases.
  • Comprehensive evaluation using ADNI data, including MRI, FDG-PET, CSF, and proteomics, validates the model's effectiveness.

Conclusions:

  • The developed bi-level learning model effectively integrates heterogeneous data sources, even with block-wise missing entries, for Alzheimer's Disease research.
  • The imputation-free approach for incomplete data offers significant advantages in performance and applicability.
  • The proposed methods provide a robust and efficient framework for multi-source data analysis in complex biomedical studies.