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 Video

Updated: Oct 26, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.4K

MMiDaS-AE: Multi-modal Missing Data aware Stacked Autoencoder for Biomedical Abstract Screening.

Eric W Lee1, Byron C Wallace2, Karla I Galaviz1

  • 1Emory University.

Proceedings of the ACM Conference on Health, Inference, and Learning
|July 26, 2021
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Models for Integrated HIV-Hypertension Care: Comparative Analysis of Six Integration Approaches from Implementation Research Projects Across Africa.

Research square·2026
Same author

Does Domain-Specific Retrieval Augmented Generation Help LLMs Answer Consumer Health Questions?

Proceedings of machine learning research·2026
Same author

Impact of Imaging Protocols on Thermal Detection of Pressure Injuries: Threshold versus Deep Learning Across Skin Tones.

medRxiv : the preprint server for health sciences·2026
Same author

Correction: Impact of skin tone, environmental, and technical factors on thermal imaging.

PloS one·2026
Same author

Comparing machine and deep learning models for pediatric anxiety classification using structured EHRs and area-based measures of health data.

PloS one·2026
Same author

GDA-AM: ON THE EFFECTIVENESS OF SOLVING MIN-IMAX OPTIMIZATION VIA ANDERSON MIXING.

... International Conference on Learning Representations·2026
Same journal

Variational Learning of Individual Survival Distributions.

Proceedings of the ACM Conference on Health, Inference, and Learning·2022
Same journal

Deidentification of free-text medical records using pre-trained bidirectional transformers.

Proceedings of the ACM Conference on Health, Inference, and Learning·2021
Same journal

Multiple Instance Learning for Predicting Necrotizing Enterocolitis in Premature Infants Using Microbiome Data.

Proceedings of the ACM Conference on Health, Inference, and Learning·2021
Same journal

CaliForest: Calibrated Random Forest for Health Data.

Proceedings of the ACM Conference on Health, Inference, and Learning·2021
Same journal

TASTE: Temporal and Static Tensor Factorization for Phenotyping Electronic Health Records.

Proceedings of the ACM Conference on Health, Inference, and Learning·2021
Same journal

Adverse Drug Reaction Discovery from Electronic Health Records with Deep Neural Networks.

Proceedings of the ACM Conference on Health, Inference, and Learning·2020
See all related articles

We developed a novel Multi-modal Missing Data aware Stacked Autoencoder (MMiDaS-AE) to semi-automate article screening for systematic reviews (SRs). This AI model significantly improves efficiency and accuracy in identifying relevant studies for health research.

Area of Science:

  • Information Science
  • Biomedical Informatics
  • Artificial Intelligence

Background:

  • Systematic reviews (SRs) are crucial for synthesizing health research but are labor-intensive.
  • Identifying relevant studies for SRs involves manually screening thousands of articles, a time-consuming process.
  • Current methods for study identification in SRs lack efficiency and scalability.

Purpose of the Study:

  • To propose and evaluate MMiDaS-AE, a novel AI model for semi-automating the screening process in systematic reviews.
  • To leverage multi-modal data (document, topic, citation networks) for improved article relevance prediction.
  • To address missing data challenges in multi-modal representations for enhanced model robustness.

Main Methods:

  • Developed a Multi-modal Missing Data aware Stacked Autoencoder (MMiDaS-AE).
Keywords:
Applied computing → Health informaticsInformation systems → Clustering and classificationMissing Data ImputationMulti-modal Stacked AutoencoderSystematic Review

Related Experiment Videos

Last Updated: Oct 26, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.4K
  • Utilized three representations: document embeddings, topic modeling, and citation networks.
  • Implemented a stacked autoencoder to learn a compressed shared representation from multi-modal inputs.
  • Incorporated a missing data imputation mechanism for incomplete modalities.

    • Main Results:

    • The MMiDaS-AE model demonstrated significant performance improvements in semi-automating SR screening.
    • The multi-modal approach effectively captured article relevancy by integrating diverse data sources.
    • The imputation strategy enhanced the model's ability to handle missing citation or topic information.

    Conclusions:

    • MMiDaS-AE offers a promising solution for semi-automating systematic review screening, reducing manual effort.
    • The model's ability to handle missing data makes it practical for real-world SR applications.
    • This approach has the potential to accelerate the synthesis of evidence in health research.