Computational Intelligence Method for Detection of White Blood Cells Using Hybrid of Convolutional Deep Learning and...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• Infectious diseases pose significant global health and economic challenges.
• Microscopic analysis of blood cells is crucial for disease detection.
• Artificial intelligence (AI) and deep learning are increasingly vital for analyzing complex biological data.

Purpose of the Study:

• To develop and implement a reliable AI-based system for processing microscopic blood cell images.
• To accurately classify five distinct types of white blood cells using advanced computational techniques.
• To enhance the diagnostic capabilities for infectious diseases through automated image analysis.

Main Methods:

• Utilized a deep convolutional neural network for white blood cell classification.
• Combined Scale-Invariant Feature Transform (SIFT) with deep learning for feature extraction.
• Employed the Gram-Schmidt algorithm for image segmentation.
• Evaluated the system on the LISC and WBCis databases.

Main Results:

• Achieved high segmentation accuracy: 95.84% on the LISC dataset and 97.33% on the WBCis dataset.
• Demonstrated the effectiveness of the combined SIFT and deep learning approach for white blood cell classification.
• Validated the system's reliability and performance on standard benchmark datasets.

Conclusions:

• Deep learning models show significant promise for developing robust microscopic image processing systems.
• The proposed AI system offers a reliable tool for automated white blood cell analysis and disease diagnosis.
• This research contributes to advancing AI applications in medical diagnostics and infectious disease detection.