Retracted: Performance Analysis of Hookworm Detection using Deep Convolutional Neural Network

Abstract

Hookworm is a disease that is caused by an intestinal parasite called roundworm. Owing to poor hygiene in the budding countries, hookworm disease is a main source of maternal and toddler morbidity. The previous work focuses on bleeding area recognition and frame localization is unravelled with Support Vector Machine (SVM) and KNN classifier. However the hookworms exhibit dissimilar profile, widths and curve orientations. These difficulties represent an incredible trouble for programmed hookworm recognition. The proposed system consists of two Convolutional Neural Networks, specifically edge withdrawal system along with hookworm categorization system, which maintain a strategic distance from the edge quality reserving and accelerate the order. To combine the tubular areas got from the edge withdrawal system and the attribute map got from the hookworm taxonomy system, two edges pooling layers are proposed. In this proposed method, the accuracy and the area under curve analysis in detecting the hookworm shows its prospective for clinical application. The tool used for this process is MATLAB Software.