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: Dec 11, 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

1.5K

Using autoencoders as a weight initialization method on deep neural networks for disease detection.

Mafalda Falcão Ferreira1,2, Rui Camacho1,2, Luís F Teixeira1,2

  • 1Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias, s/n, Porto, 4200-465, Portugal.

BMC Medical Informatics and Decision Making
|August 22, 2020
PubMed
Summary

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

CBVLM: Training-free explainable concept-based Large Vision Language Models for medical image classification.

Computers in biology and medicine·2025
Same author

A two-step concept-based approach for enhanced interpretability and trust in skin lesion diagnosis.

Computational and structural biotechnology journal·2025
Same author

Anatomical Concept-based Pseudo-labels for Increased Generalizability in Breast Cancer Multi-center Data.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Fill in the blank for fashion complementary outfit product Retrieval: VISUM summer school competition.

Machine vision and applications·2023
Same author

First insight into oral microbiome diversity in Papua New Guineans reveals a specific regional signature.

Molecular ecology·2022
Same author

Machine learning methods to predict attrition in a population-based cohort of very preterm infants.

Scientific reports·2022

Machine learning models effectively detect various cancers using gene expression data. Fine-tuning autoencoder weights significantly improved classification accuracy for RNA-Seq and image datasets, outperforming existing methods.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning in Oncology

Background:

  • Cancer remains a leading cause of mortality, driving research into AI-driven diagnostic tools.
  • Gene expression analysis offers a promising avenue for automated cancer sample classification.

Purpose of the Study:

  • To classify five cancer types (thyroid, skin, stomach, breast, lung) using RNA-Seq data.
  • To evaluate autoencoders (AEs) as a deep neural network weight initialization technique for cancer classification.
  • To assess the generalizability of the methodology on image-based datasets (malaria, breast masses).

Main Methods:

  • Utilized RNA-Seq datasets for five cancer types and image-extracted features.
  • Employed three distinct autoencoders (AEs) for deep neural network weight initialization.
Keywords:
AutoencodersCancerClassificationDeep learningGene expression analysis

More Related Videos

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

2.2K

Related Experiment Videos

Last Updated: Dec 11, 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

1.5K
Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

2.2K
  • Compared weight fixing versus fine-tuning strategies and different AE embedding methods.
  • Investigated the impact of AE architecture, latent vector dimension, and data imputation on performance.
  • Validated the pipeline on malaria and breast cancer image datasets, using held-out test sets to prevent overfitting.

    • Main Results:

    • The methodology achieved high performance across RNA-Seq and image datasets.
    • Outperformed established baselines with an average F1 score of 99.03% (thyroid cancer), 89.95% (malaria), and 98.84% (breast cancer).
    • Achieved high Matthews Correlation Coefficient (MCC) scores: 0.99, 0.84, and 0.98 for the respective datasets.

    Conclusions:

    • Fine-tuning autoencoder weights in the top layers consistently yielded superior results across all datasets.
    • The proposed machine learning pipeline demonstrates robust performance and generalizability in cancer detection.
    • This approach surpasses previous reported results, offering a powerful tool for accelerating cancer diagnosis.