CLASSIFICATION OF BLOOD SMEAR IMAGES FOR DETECTION OF MALARIA USING MACHINE-LEARNING ALGORITHMS

: Malaria are serious diseases that require quick and accurate diagnosis for effective treatment. The traditional method of detecting these diseases involves examining blood smear images under a microscope, which is time-consuming, requires skilled professionals, and is prone to human error. To address these issues, this study aims to build an automated classification system for blood smear images by leveraging machine learning and deep learning approaches. The system incorporates Convolutional Neural Networks (CNNs), utilizes transfer learning with models like ResNet50, MobileNet, DenseNet, VGG16, VGG19, and Inception V3, and employs YOLO (You Only Look Once) for object detection tasks. The images are first pre-processed to enhance their quality, followed by feature extraction to identify key patterns. The model then classifies the images into infected and non-infected categories with high accuracy. The findings indicate that this method enables quicker, more accurate, and economical disease detection, minimizing the need for manual examination.This system can be highly beneficial, especially in areas with limited access to medical facilities and expert pathologists.