Malaria Outbreak Detection Using Selected Deep Learning Algorithms

Abstract

Malaria remains a critical public health issue, predominantly affecting tropical and subtropical regions. Traditional diagnostic methods, such as microscopic analysis of blood smears, are time-intensive, require skilled personnel, and often lack sensitivity, particularly in detecting early-stage infections in low-resource settings. The study examines the application of deep-learning algorithms to enhance the prediction and diagnosis of malaria. Using visual data from blood smears, various neural network architectures, including Convolutional Neural Networks (CNNs), VGG-16, Inception, Xception, and DenseNet, were evaluated for their ability to detect malaria cases and analyze their correlation with mortality rates across regions. Results demonstrated that CNNs outperformed other models in accuracy and reliability, highlighting the potential of deep learning to improve early diagnosis and treatment. Integrating deep learning with image-based analysis offers a rapid, accessible, and precise diagnostic tool suitable for resource-constrained settings. These findings emphasize the transformative potential of advanced machine learning techniques in malaria control, enabling timely interventions and improved patient outcomes. The research advocates for incorporating deep learning models into public health systems. It encourages further studies to optimize these technologies and extend their use to other infectious diseases.