Related Experiment Video
Updated: May 11, 2026
07:05
Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
Published on: February 15, 2022
Extracting, Recognizing, and Counting White Blood Cells from Microscopic Images by Using Complex-valued Neural
Hamid Akramifard1, Mohammad Firouzmand, Reza Askari Moghadam
1Faculty of Computer Engineering and IT University of Payame Nour, Tehran, IR Iran.
Journal of Medical Signals and Sensors
|May 30, 2013
Summary
This study presents a novel method for extracting and classifying white blood cells (WBCs) from microscopic images using complex-valued neural networks (CVNNs). The approach achieves high accuracy in WBC counting, offering a promising tool for medical diagnostics.
Area of Science:
- Medical image analysis
- Computational biology
- Artificial intelligence in healthcare
Background:
- Accurate counting of white blood cells (WBCs) is crucial for diagnosing various medical conditions.
- Traditional methods for WBC analysis can be time-consuming and prone to human error.
- Automated analysis of hematology images presents a significant challenge in medical diagnostics.
Purpose of the Study:
- To develop and evaluate a method for automated extraction, recognition, and counting of different types of WBCs from microscopic blood images.
- To leverage the capabilities of artificial neural networks (ANNs), specifically complex-valued neural networks (CVNNs), for enhanced classification accuracy and efficiency.
- To provide a reliable and accurate automated system for WBC differential counting to aid in medical diagnosis.
Main Methods:
- White blood cells (WBCs) are initially extracted from microscopic images using the RGB color system.
- Normalized feature vectors are generated based on the distinctive features and color schemes of each WBC type.
- Classification and counting of WBCs are performed using a complex-valued back-propagation neural network (CVNN).
Main Results:
- The proposed method demonstrates high accuracy in extracting and recognizing WBCs, even from low-quality images.
- Complex-valued neural networks (CVNNs) achieved significantly faster learning times compared to traditional real-valued neural networks.
- The system successfully quantifies the number of each type of white blood cell, providing quantitative diagnostic information.
Conclusions:
- The developed method offers an accurate and efficient approach for automated white blood cell (WBC) counting and classification.
- The use of complex-valued neural networks (CVNNs) proves advantageous for WBC image analysis due to improved speed and accuracy.
- This automated system holds potential for improving the efficiency and reliability of medical diagnostics in hematology.
