Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI)
Mohammad Anas
Jamia Hamdard (Government Aided Deemed to be University)
Computer Science & IT Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Intestinal disorders categorization in endoscopic images using deep learning architectures Esha Saxena; Suraiya Parveen; Mohd. Abdul Ahad; Meenakshi Yadav; Mohammad Anas
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 15, No 3: June 2026
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v15.i3.pp2347-2356

Abstract

Gastroenterology is revolutionized by advancements in artificial intelligence (AI). As the gastrointestinal (GI) tract is consulted, globally 40% of the world's and 18% of the Indian population are affected. AI is a reliable sword for diagnosing issues related to the GI tract. The learning capabilities of deep learning (DL) techniques make it widely helpful in medical investigations. The variety of data available in the medical sector generates the need for an appropriate model for every problem domain. The purpose of this research is to explore the significance of medical image pre-processing and the implementation of pre-trained DL models on endoscopic images for the diagnosis of disease. Convolutional neural network (CNN)-based architectures have robust diagnostic potential for medical images. It can assist physicians as a tool for disease analysis, screening and help in investigating further needs. The paper also provides a comparative performance framework showing CNN architectures and preprocessing techniques for endoscopic images to highlight the key points important for investigating GI tract related diseases. The endoscopic images were trained over VGG-16, ResNet-50 and DenseNet-121, DL models. The result suggests that VGG-16 and ResNet-50 gave promising results with an accuracy maximum of 87.50%.
Co-Authors Esha Saxena Meenakshi Yadav Mohd. Abdul Ahad Suraiya Parveen