PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
July 5, 2026DiagnosticsOpen Access

Image Processing and Deep Convolutional Neural Network Method for Automated Malaria Parasite Detection in Thin Blood Slide Images

View Full Paper
Ask AI
Bookmark
Share

Authors

KKKavita KumariTKTaruna KauraAMAbhishek Mewara

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates automated malaria detection in blood slide images, indicating accuracy improvements in screening methods.

Key Points

  • The study aimed to develop an automated malaria screening algorithm to identify malaria-infected red blood cells in blood slide images.
  • Employed digital image processing techniques for preprocessing, including watershed transform and connected component labeling.
  • Utilized convolutional neural networks to classify segmented red blood cells as normal or infected.
  • Evaluated five classification engines across a dataset of 2422 training images and 692 testing images.
  • VGG19 model achieved the highest accuracy of 99.57% in classifying malaria parasites from test images.
  • The proposed custom CNN model demonstrated competitive performance with an accuracy of 99.14%.
  • Transfer learning models consistently outperformed traditional methods in malaria parasite detection.

Cite This Study

Kumari et al. (2026) studied this question.

synapsesocial.com/papers/6a49f503f5d1d45b288002cehttps://doi.org/10.3390/diagnostics16132091
View Full Paper
Ask AI
Bookmark
Share