Deep Learning Solutions for WBC Classification in Juvenile Visayan Warty Pigs

Precise monitoring of the health status of visayan warty pigs, particularly through the analysis of blood samples, provides conservationists and researchers with valuable insights into their immune responses and overall health. The manual identification of white blood cells (WBCs) in blood smears is a laborious task and automation can be costly. To address these challenges, this research employs deep learning techniques for recognition of WBCs. Deep learning method used in this study is Convolutional Neural Networks (CNNs) which utilize convolutional layers to extract essential features from images. These extracted features are then fed into neural networks for categorization, effectively distinguishing between different classes of images. The intrinsic capacity of CNNs to autonomously learn intricate image attributes empowers them to make accurate predictions on new data, making them especially well-suited for tasks centered around image classification. Various deep learning algorithms were employed and compared, with the best-performing algorithm subsequently fine-tuned to enhance its robustness and achieve higher levels of accuracy. The most proficient deep learning models achieved an impressive accuracy of 98.86% in classifying WBCs from juvenile visayan warty pigs. This accomplishment underscores the remarkable accuracy of the top-performing model.