Effective Detection of Exomphalos in Fetal Ultrasound Images Using Canny edge detector

The detection of Exomphalos, a type of abdominal wall defect which consents the abdominal contents to obtrude outside as like sac structure using DSS-Decision Support System is proposed. Types of Exomphalos are major and minor exomphalos. In Exomphalos Major, the liver, intestines and occasionally other organs are confined in a sac whereas in Exomphalos Minor, the intestine alone rests in the sac. This proposed Computer aided decision support system takes mid-trimester ultrasound images of the abdominal wall of the fetus as input and processes the images using efficient algorithms at every phase to detect Exomphalos and further classify it. It aims at eradicating manual intervention and to accurate the diagnosis. This system has four stages: Pre-processing mid-trimester ultrasound images, Segmentation for region of interest, Feature Extraction and Feature Selection from the region of interest. In this paper, various filters and several edge detection algorithms are used for De-speckling and segmentation respectively. After the acquisition of results, their performances are compared and it has been foundthat Median filter is effectual for despeckling and Canny edge detection technique is efficient for segmentation.