Colorectal Polyps Shape Classification Based On CNN Models

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Author: Alawnah, Al-mo'men bellah
Format: Recurso digital
Published: Zenodo 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866902077656006656
author Alawnah, Al-mo'men bellah
author_facet Alawnah, Al-mo'men bellah
contents <p>Abstract: Colorectal polyps are precursors to colorectal cancer, making their early detection<br>and classification crucial for effective treatment. This study focuses on developing<br>Convolutional Neural Networks (CNNs) to classify the shapes of colorectal polyps in<br>endoscopic images. Various CNN architectures, including ResNet101, VGG16, and a custom<br>CNN, were trained and evaluated on the Kvasir dataset, which consists of annotated endoscopic<br>images classified into flat, sessile, and pedunculated polyps.<br>The dataset was divided into 75% for training and 25% for testing, with preprocessing<br>techniques such as resizing, normalization, and augmentation applied to enhance model<br>performance. Transfer learning was leveraged to enhance feature extraction, and the models<br>were evaluated using metrics such as accuracy, recall, precision, and F1-score.<br>Experimental results showed that ResNet101 achieved the highest accuracy of 95.15%,<br>followed by VGG16 (94.85%) and the custom CNN (94.24%). Models such as InceptionV3 and<br>Xception exhibited comparatively lower performance. These findings suggest that CNNs,<br>particularly ResNet101, can effectively classify colorectal polyp shapes, aiding in early<br>diagnosis and treatment. Future work will focus on hyperparameter optimization, ensemble<br>learning, and advanced deep learning techniques to further enhance classification accuracy</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15128796
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Colorectal Polyps Shape Classification Based On CNN Models
Alawnah, Al-mo'men bellah
Colorectal polyps shape, Deep learning, CNN models, PARIS classification.
<p>Abstract: Colorectal polyps are precursors to colorectal cancer, making their early detection<br>and classification crucial for effective treatment. This study focuses on developing<br>Convolutional Neural Networks (CNNs) to classify the shapes of colorectal polyps in<br>endoscopic images. Various CNN architectures, including ResNet101, VGG16, and a custom<br>CNN, were trained and evaluated on the Kvasir dataset, which consists of annotated endoscopic<br>images classified into flat, sessile, and pedunculated polyps.<br>The dataset was divided into 75% for training and 25% for testing, with preprocessing<br>techniques such as resizing, normalization, and augmentation applied to enhance model<br>performance. Transfer learning was leveraged to enhance feature extraction, and the models<br>were evaluated using metrics such as accuracy, recall, precision, and F1-score.<br>Experimental results showed that ResNet101 achieved the highest accuracy of 95.15%,<br>followed by VGG16 (94.85%) and the custom CNN (94.24%). Models such as InceptionV3 and<br>Xception exhibited comparatively lower performance. These findings suggest that CNNs,<br>particularly ResNet101, can effectively classify colorectal polyp shapes, aiding in early<br>diagnosis and treatment. Future work will focus on hyperparameter optimization, ensemble<br>learning, and advanced deep learning techniques to further enhance classification accuracy</p>
title Colorectal Polyps Shape Classification Based On CNN Models
topic Colorectal polyps shape, Deep learning, CNN models, PARIS classification.
url https://doi.org/10.5281/zenodo.15128796