Automated Mosquito Vector Identification and Malaria Parasite Detection using Deep Learning Approach

Malaria, an infectious disease predominantly transmitted through the bites of infected mosquitoes, continues to pose a significant worldwide health burden. To effectively tackle this matter, it is imperative to not only accurately identify the specific mosquito vectors involved but also to precisely detect the presence of malaria parasites inside blood samples. This research presents two case studies utilizing deep learning approach for malaria parasite and its vector identification. Through the deep learning object detection model, this study develops a neural network-based algorithm in identifying mosquito vectors. The automated object detection model has shown promising results in identifying Aedes, Culex and Anopheles mosquitoes. In addition, this research presents the detection of Plasmodium parasites in microscopic blood smear images. Four deep learning models were developed and compared to classify four human malaria species. The DenseNet-121 achieved the best accuracy of 99.5%. The automated technique discussed herein has far-reaching ramifications that encompass the enhancement of public health endeavors, the facilitation of targeted interventions, and ultimately the advancement of the global campaign against malaria.