Efficient Abdominal Aortic Aneurysm Detection Using Optimized PNN Approach
R. Anuja, M. M. Gowthul Alam, K. Hariharan, P. Mukilan, M. Thanga Shalini, D. Karthikeyan · 2024
An Abdominal Aorta Aneursym (AAA) is generally an abnormal conditions that affects aorta, which causes severe damages to mankind including kidney damage and heart attack, thus proper diagnosis is essential to provide patients with adequate treatment. However, accurate detection of AAA faces certain difficulties such as inaccurate prediction which leads to reduced reliability, hence, this paper proposes image processing based AAA disease detection system. Firstly, segmentation utilizing spatial Fuzzy C-means Approach (FCM) is performed which segments the images for achieving higher detection rate with reduced incorrect prediction. Secondly, Feature Extraction process by Gray Level Co-occurrence Matrix (GLCM) is performed for extracting the most relevant features from segmented images and finally, Chaotic Dragonfly Optimized (CDO) Probabilistic Neural Network (PNN) classifier is deployed for acquiring enhanced classification process. Moreover, to validate performance efficacy of proposed system, Python simulation is implemented, which depicts increased accuracy (87%). Thereby, contributing positive impacts towards disease prediction, thus enabling to provide early and proper treatment.