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: Jul 15, 2026

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model
08:20

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model

Published on: October 27, 2023

FedPDM-Net: a federated prototype-guided disentangled deep learning framework for explainable malaria detection.

Komal Kumar Napa1, Sangeetha Murugan2, J Senthil Murugan3

  • 1Department of Artificial Intelligence and Data Science, Saveetha Engineering College, Chennai, India.

Scientific Reports
|July 13, 2026
PubMed
Summary

Related Concept Videos

Malaria01:29

Malaria

Malaria pathogenesis in humans reflects a delicate interplay between parasite biology and host response. Clinical illness reflects a host’s immune response to the parasite’s asexual replication cycle, which is often asymptomatic in individuals with partial immunity. From the parasite's perspective, transmission between mosquito and human with minimal host pathology is evolutionarily advantageous. Among the six Plasmodium species infecting humans, P. falciparum and P. vivax dominate in global...

You might also read

Related Articles

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

Sort by
Same author

Multimodal fusion of handwriting images and kinematic features for Parkinson disease detection.

Scientific reports·2026
Same author

Comment on "Efficacy analysis of small-incision in situ decompression under ultrasound combined with shear-wave elastography in the treatment of ulnar neuropathy at the elbow".

International orthopaedics·2026
Same author

Satellite-based oil spill detection using an explainable ViR-SC hybrid deep learning ensemble for improved accuracy and transparency.

Scientific reports·2026
Same author

BRAIN-META: A reproducible CNN-vision transformer meta-ensemble pipeline for explainable brain tumor classification.

MethodsX·2026
Same author

Efficient convolutional neural networks for acute lymphoblastic leukaemia prediction in computer vision.

Scientific reports·2025
Same author

Corrigendum to "Development of an explainable machine learning model for Alzheimer's disease prediction using clinical and behavioural features" [MethodsX 15 (2025) 103491].

MethodsX·2025

This study introduces FedPDM-Net, a privacy-preserving deep learning model for malaria detection using federated learning. The explainable AI approach ensures accurate and robust diagnosis, even in resource-limited settings.

Area of Science:

  • Medical Diagnostics
  • Artificial Intelligence
  • Public Health

Background:

  • Malaria diagnosis is critical, especially in resource-limited areas.
  • Current deep learning methods for malaria detection face privacy, scalability, and interpretability challenges.
  • Automated detection from blood smears is vital for timely intervention.

Purpose of the Study:

  • To develop FedPDM-Net, a federated, prototype-guided, disentangled deep learning framework for explainable malaria detection.
  • To address privacy, scalability, and interpretability issues in automated malaria diagnosis.
  • To provide a robust and reliable diagnostic tool for malaria.

Main Methods:

  • Federated learning for privacy-preserving collaborative training.
  • Prototype memory module for capturing infection patterns.
Keywords:
Disentangled learningExplainable AIFederated learningMalaria detectionPrototype learning

Related Experiment Videos

Last Updated: Jul 15, 2026

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model
08:20

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model

Published on: October 27, 2023

  • Disentangled feature learning to separate image characteristics.
  • Attention-based fusion and Grad-CAM for enhanced interpretability.

    • Main Results:

    • Achieved 97.3% accuracy (centralized) and 96.6% accuracy (federated) with minimal performance drop (0.7%).
    • Demonstrated robustness with >95% performance under perturbations and stable 5-fold cross-validation accuracy (96.83% ± 0.20%).
    • Showcased significant improvements over baselines (p < 0.05) and reliable confidence estimation (ECE = 0.032).

    Conclusions:

    • FedPDM-Net offers a robust, interpretable, and privacy-preserving solution for automated malaria diagnosis.
    • The framework is suitable for resource-limited settings, enhancing diagnostic capabilities.
    • Efficient federated communication (147 MB overhead over 5 rounds) supports practical implementation.