Related Experiment Video
Updated: Jul 18, 2026
09:34
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
3.9K
Heterogeneous virus classification using a functional deep learning model based on transmission electron microscopy
Niloy Sikder1,2, Md Al-Masrur Khan3, Anupam Kumar Bairagi4
1Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, The Netherlands.
Scientific Reports
|November 23, 2024
Summary
This study introduces a deep learning model for rapid virus identification using Transmission Electron Microscopy images. The model accurately classifies 14 virus types, offering a fast and reliable diagnostic tool.
Area of Science:
- Virology
- Computational Biology
- Microscopy
Background:
- Viruses are microscopic infectious agents with high adaptability and potential for severe complications.
- Accurate and timely virus identification is crucial for managing biological threats to plants and animals.
- Manual virus detection methods are often slow and lack precision.
Purpose of the Study:
- To develop a computer-based automatic diagnosis method for instant virus identification.
- To propose a deep learning-based classification model utilizing Transmission Electron Microscopy (TEM) images.
- To enhance the speed and accuracy of virus type determination.
Main Methods:
- Utilized a dataset of Transmission Electron Microscopy (TEM) images for virus classification.
- Implemented two image processing techniques for noise reduction in raw microscopy images.
- Developed and applied a Convolutional Neural Network (CNN) model for virus type classification.
Main Results:
- The proposed deep learning model achieved a maximum classification accuracy of 97.44%.
- The model demonstrated high reliability in differentiating among 14 distinct virus types.
- Achieved a maximum F1-score of 97.44%, indicating strong performance.
Conclusions:
- The deep learning-based approach offers an effective and reliable method for virus identification.
- This automated scheme provides a fast and dependable complement to existing diagnostic procedures.
- The model's high accuracy supports its implementation in clinical and research settings.
