Related Experiment Video
Updated: Jun 29, 2025
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
4.0K
DLBCNet: A Deep Learning Network for Classifying Blood Cells
Ziquan Zhu1, Zeyu Ren1, Siyuan Lu1
1School of Computing and Mathematical Sciences, University of Leicester, Leicester LE1 7RH, UK.
Big Data and Cognitive Computing
|April 1, 2024
Summary
This study introduces DLBCNet, a novel deep learning network for multi-class blood cell classification. The model achieves high accuracy, demonstrating improved blood cell analysis performance.
Area of Science:
- Hematology
- Medical Imaging
- Artificial Intelligence
Background:
- Blood analysis provides crucial health insights.
- Deep learning (DL) models are increasingly used for automated blood cell analysis.
- Existing DL models for blood cell analysis have limitations.
Purpose of the Study:
- To propose a novel deep learning network, DLBCNet, for multi-classification of blood cells.
- To enhance the accuracy and performance of automated blood cell diagnosis.
Main Methods:
- Developed DLBCNet, incorporating a blood cell generative adversarial network (BCGAN) for synthetic image generation.
- Utilized a pre-trained ResNet50 as a backbone for feature extraction.
- Employed an enhanced transformer recurrent network (ETRN) for improved classification.
Main Results:
- Achieved an average accuracy of 95.05%.
- Reported average sensitivity, precision, specificity, and F1-score of 93.25%, 97.75%, 93.72%, and 95.38%, respectively.
- Demonstrated superior performance compared to existing state-of-the-art methods.
Conclusions:
- DLBCNet significantly improves multi-classification performance for blood cells.
- The proposed model offers a promising advancement in automated hematological analysis.
- The results indicate the potential of DLBCNet for clinical applications.
