DETECTION OF HOOKWORM USING DEEP LEARNING

Abstract

Hookworms are intestinal, blood-feeding, parasitic roundworms that cause types of infection known as helminthiases. It has tubular structure with grayish white or pinkish semi-transparent body. The most serious effects of hookworm infection are blood loss leading to anemia, in addition to protein loss. Wireless Capsule Endoscopy (WCE) has become a widely used diagnostic technique to examine inflammatory diseases and disorders. To detect the hookworms two CNN networks, namely edge extraction network and hookworm classification network are used in the existing system. The proposed system has Median filter to remove noise from the dataset. Canny Edge detection is used to detect the edges of hookworm. This technique has low SNR (Signal to Noise Ratio) compared to other edge detection techniques. The architecture of CNN (Convolutional Neural Network) is modified to use it as a classifier. It classifies the hookworm and normal Wireless Capsule Endoscopy (WCE) dataset.