Özet:

Abstract The idea of this project is to implement a Computer-Aided Detection system (CAD) for the early detection of pancreatic tumors based on the UNet++ architecture. Contrast Limited Adaptive Histogram Equalization (CLAHE) and Boosted Anisotropic Diffusion Filter (BADF) methods are used to enhance the MRI image. The pancreatic region associated with a lesion is precisely separated from the MRI image by segmentation. The best subset of texture characteristics is assessed by creating a classification system based on texture features that integrate HHO-based CNN and HHO-based Bag of visual terms. This will enhance classification accuracy. Transfer learning and Fine-tuning model using VGG 16 classifiers are utilised to create an automated system for classifying different grades of tumors in MRI Images. For various tumor classes, quantitative analysis is performed. The accuracy of the classification of the proposed classifier is validated and it is compared with the state-of-the-art approach.

Anahtar Kelimeler: