Evaluation of three convolutional neural network models applied to the detection of leukocytes in intravital microscopy video images.

• Grant number: 23/07612-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2023
• End date: January 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Ricardo José Ferrari
• Grantee: Leonardo Cavalcante da Silva
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

Resumo The in vivo study of cellular and molecular mechanisms underlying leukocyte-endothelium interaction in the microcirculation of various tissues and inflammatory conditions is of great importance for the development of new anti-inflammatory drugs. One commonly used model is Experimental Autoimmune Encephalomyelitis (EAE), which is widely employed in Multiple Sclerosis research. Microscopy Intravital (MI) imaging serves as the standard method for analysis, providing high temporal resolution and low spatial depth images. Currently, leukocyte-endothelium interactions in small animals are visually analyzed from MI image sequences. However, this manual procedure is time-consuming and can lead to observer fatigue, resulting in potential inaccuracies in statistics. In this context, this research proposal aims to evaluate the effectiveness of Faster R-CNN, YOLO, and SSD techniques for leukocyte detection in intravital microscopy images. The methodology involves applying these techniques alongside image preprocessing algorithms to improve detection quality and efficiency. Results will be quantitatively analyzed by comparing manual leukocyte annotations by an expert with the automated image processing results, using Precision-Recall curve methodology. This project will contribute to the development of automated tools for leukocyte detection, which will be highly valuable in the field of medical diagnostics.