Semi/Fully-Automated Segmentation of Gastric-Polyp Using Aquila-Optimization-Algorithm Enhanced Images

The incident rate of the Gastrointestinal-Disease (GD) in humans is gradually rising due to a variety reasons and Endoscopic/Colonoscopic-Image (EI/CI) supported evaluation GD an approved practice. Extraction suspicious section EI/CI essential diagnose disease its severity. proposed research aims implement joint thresholding segmentation framework extract Gastric-Polyp (GP) with better accuracy. GP detection system consist; (i) Enhancement region using Aquila-Optimization-Algorithm tri-level entropy (Fuzzy/Shannon/Kapur) between-class-variance (Otsu) technique, (ii) Automated (Watershed/Markov-Random-Field) semi-automated (Chan-Vese/Level-Set/Active-Contour) fragment, (iii) Performance validation scheme. experimental investigation was performed four benchmark EI dataset (CVC-ClinicDB, ETIS-Larib, EndoCV2020 Kvasir). similarity measures, such as Jaccard, Dice, accuracy, precision, sensitivity specificity are computed confirm clinical significance work. outcome this confirms that fuzzy-entropy combined Chan-Vese helps achieve measures compared alternative schemes considered research.