Publications Scientifiques

Permanent URI for this communityhttps://dspace.univ-boumerdes.dz/handle/123456789/10

Browse

Author: search.filters.author.Berrichi, AliHas files: No

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Vision Transformer Model for Gastrointestinal Tract Diseases Classification from WCE Images
    (Institute of Electrical and Electronics Engineers, 2024) Bella, Faiza; Berrichi, Ali; Moussaoui, Abdelouahab
    Accurate disease classification utilizing endoscopic images indeed poses a significant challenge within the field of gastroenterology. This research introduces a methodology for assisting medical diagnostic procedures and detecting gastrointestinal (GI) tract diseases by categorizing features extracted from endoscopic images using Vision Transformer (ViT) models. We propose three ViT-inspired models for classifying GI tract diseases using colon images acquired through wireless capsule endoscopy (WCE). The highest achieved accuracy among our models is 97.83%. We conducted a comparative analysis with three pre-trained CNN (Convolutional Neural Network) models namely, Xception, DenseNet121, and MobileNet, alongside recent research papers to validate our findings.
  • No Thumbnail Available
    Item
    Multi-Objective artificial bee colony algorithm for Parameter-Free Neighborhood-Based clustering
    (IGI Global, 2021) Boudane, Fatima; Berrichi, Ali
    Although various clustering algorithms have been proposed, most of them cannot handle arbitrarily shaped clusters with varying density and depend on the user-defined parameters which are hard to set. In this paper, to address these issues, the authors propose an automatic neighborhood-based clustering approach using an extended multi-objective artificial bee colony (NBC-MOABC) algorithm. In this approach, the ABC algorithm is used as a parameter tuning tool for the NBC algorithm. NBC-MOABC is parameter-free and uses a density-based solution encoding scheme. Furthermore, solution search equations of the standard ABC are modified in NBC-MOABC, and a mutation operator is used to better explore the search space. For evaluation, two objectives, based on density concepts, have been defined to replace the conventional validity indices, which may fail in the case of arbitrarily shaped clusters. Experimental results demonstrate the superiority of the proposed approach over seven clustering methods