Ancient blood cell classification on explication using convolutional neural networks

Ancient blood cell classification on explication using convolutional neural networks

연구 분야: Artificial Intelligence

학회: Multimedia Tools and Applications

In medical diagnostics, blood cell classification is an essential responsibility. Traditional procedures, which take time and are prone to human error, manually examine blood smears. The emphasis is on ancient blood cell samples, which have special difficulties due to differences in staining methods and acquisition technologies. Conventional techniques struggle to identify significant patterns in these tiny organisms, which limits our understanding of prehistoric health, ways of life, and evolutionary processes. During the classification of ancient blood cells, the damaged and morphologically altered condition presents a significant issue in archaeology. In recent years, CNNs have shown promise as tools for image recognition and classification, potentially resolving these issues. This work shows the effectiveness of CNNs in unravelling the riddles of ancient blood via a review of current developments and case examples, offering fresh perspectives on the lives and legacies of our ancestors. This research investigates the employment of support vector machines (SVM) to automate blood cell classification and convolutional neural networks (CNNs) for feature extraction. The use of CNNs and SVM for the classification of ancient blood cells outlines the fundamentals of CNNs and their adaptation to archaeological settings. This work also successfully completes preprocessing and segmentation. Many parameters ultimately evaluate the work’s performance.