Blood cell detection plays a vital role in the diagnosis of various hematological disorders and supports effective clinical decision-making. Conventional microscopic examination of blood smear images is time-consuming, labor-intensive, and dependent on skilled professionals, making it prone to human error. To address these challenges, this paper presents an automated blood cell detection and classification system using the YOLOv8 (You Only Look Once Version 8) deep learning model. The proposed system is designed to detect and classify three major blood cell types—Red Blood Cells (RBCs), White Blood Cells (WBCs), and Platelets—from microscopic blood smear images. A publicly available annotated dataset was used for training and validation after organizing the data in the YOLOv8 annotation format. The model was trained using optimized parameters to achieve accurate object localization and classification. YOLOv8 was selected because of its high detection speed, efficient feature extraction, and robust object detection capability. Experimental evaluation demonstrated that the proposed model achieved an overall detection accuracy (mAP) of 91.3%, indicating its effectiveness in identifying and classifying blood cells with high precision. The system accurately detects multiple blood cells simultaneously by generating bounding boxes and assigning the correct class labels. The proposed approach significantly reduces manual effort, diagnostic consistency, and minimizes human error during blood cell examination. It can serve as a reliable computer-aided diagnostic tool for hospitals, pathology laboratories, and research institutions by enabling faster and more accurate blood cell analysis. Furthermore, the system has the potential to support early disease diagnosis and can be extended in the future to detect abnormal blood cells and integrate with intelligent healthcare applications, thereby enhancing diagnostic accuracy and improving clinical decision-making.

Mersiha.S, Sri Swetha.V, Veeramani.B, Sanjeevi.C, Dr Pradeepa K,"YOLOv8-Based Blood Cell Detection and classification" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(4): Page(458-464) July - August 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.