XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI

Automated segmentation of gastrointestinal polyps is a critical step in the early detection and prevention of colorectal cancer (CRC), which is one of the most common causes of cancer-related deaths worldwide. This article presents a U-Net-based model enhanced with Attention Mechanisms and Atrous Sp...

Full description

Saved in:
Bibliographic Details
Main Authors: Arjun Kumar Bose Arnob, Muhammad Mostafa Monowar, Md. Abdul Hamid, M. F. Mridha
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Open Journal of the Computer Society
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11095343/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849762087176241152
author Arjun Kumar Bose Arnob
Muhammad Mostafa Monowar
Md. Abdul Hamid
M. F. Mridha
author_facet Arjun Kumar Bose Arnob
Muhammad Mostafa Monowar
Md. Abdul Hamid
M. F. Mridha
author_sort Arjun Kumar Bose Arnob
collection DOAJ
description Automated segmentation of gastrointestinal polyps is a critical step in the early detection and prevention of colorectal cancer (CRC), which is one of the most common causes of cancer-related deaths worldwide. This article presents a U-Net-based model enhanced with Attention Mechanisms and Atrous Spatial Pyramid Pooling (ASPP) for accurate polyp segmentation. To address the challenges of varying polyp sizes, indistinct boundaries, and complex textures, the model used a combined loss function (Binary Cross-Entropy and Dice Loss). Additionally, Gradient-Weighted Class Activation Mapping (Grad-CAM) was integrated to provide visual explanations of the model’s decisions to increase trust and interpretability by clinical practitioners. The presented model was evaluated on five benchmark datasets, achieving a Dice Coefficient of 0.8378 and a Mean Intersection over Union (mIoU) of 0.8427. The comparative analysis highlighted its superiority when compared to state-of-the-art contemporary approaches, with a precision and accuracy of 97%. Qualitative analyses also underline the ability to accurately delineate polyps, even in difficult situations. Although the model exhibited satisfactory performance, it still faced challenges regarding boundary misclassification and reduced efficacy in datasets with high variability. The next steps of this research will focus on domain adaptation and integration of additional modalities to enhance generalizability. This study provides a step toward automated polyp detection and demonstrates the potential of explainable artificial intelligence (XAI) to change the accuracy of diagnosis and healthcare for patients.
format Article
id doaj-art-0f884bbc443a410ab7cdb16cd9c9f0f8
institution DOAJ
issn 2644-1268
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Open Journal of the Computer Society
spelling doaj-art-0f884bbc443a410ab7cdb16cd9c9f0f82025-08-20T03:05:50ZengIEEEIEEE Open Journal of the Computer Society2644-12682025-01-0161283129310.1109/OJCS.2025.359220411095343XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AIArjun Kumar Bose Arnob0https://orcid.org/0009-0003-2244-2328Muhammad Mostafa Monowar1https://orcid.org/0000-0003-2822-2572Md. Abdul Hamid2https://orcid.org/0000-0001-9698-4726M. F. Mridha3https://orcid.org/0000-0001-5738-1631Department of Computer Science, American International University-Bangladesh, Dhaka, BangladeshDepartment of Information Technology, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi ArabiaDepartment of Information Technology, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi ArabiaDepartment of Computer Science, American International University-Bangladesh, Dhaka, BangladeshAutomated segmentation of gastrointestinal polyps is a critical step in the early detection and prevention of colorectal cancer (CRC), which is one of the most common causes of cancer-related deaths worldwide. This article presents a U-Net-based model enhanced with Attention Mechanisms and Atrous Spatial Pyramid Pooling (ASPP) for accurate polyp segmentation. To address the challenges of varying polyp sizes, indistinct boundaries, and complex textures, the model used a combined loss function (Binary Cross-Entropy and Dice Loss). Additionally, Gradient-Weighted Class Activation Mapping (Grad-CAM) was integrated to provide visual explanations of the model’s decisions to increase trust and interpretability by clinical practitioners. The presented model was evaluated on five benchmark datasets, achieving a Dice Coefficient of 0.8378 and a Mean Intersection over Union (mIoU) of 0.8427. The comparative analysis highlighted its superiority when compared to state-of-the-art contemporary approaches, with a precision and accuracy of 97%. Qualitative analyses also underline the ability to accurately delineate polyps, even in difficult situations. Although the model exhibited satisfactory performance, it still faced challenges regarding boundary misclassification and reduced efficacy in datasets with high variability. The next steps of this research will focus on domain adaptation and integration of additional modalities to enhance generalizability. This study provides a step toward automated polyp detection and demonstrates the potential of explainable artificial intelligence (XAI) to change the accuracy of diagnosis and healthcare for patients.https://ieeexplore.ieee.org/document/11095343/Endoscopic image analysisgrad-CAMgastrointestinal polypsemantic segmentation
spellingShingle Arjun Kumar Bose Arnob
Muhammad Mostafa Monowar
Md. Abdul Hamid
M. F. Mridha
XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
IEEE Open Journal of the Computer Society
Endoscopic image analysis
grad-CAM
gastrointestinal polyp
semantic segmentation
title XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
title_full XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
title_fullStr XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
title_full_unstemmed XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
title_short XPolypNet: A U-Net-Based Model for Semantic Segmentation of Gastrointestinal Polyps With Explainable AI
title_sort xpolypnet a u net based model for semantic segmentation of gastrointestinal polyps with explainable ai
topic Endoscopic image analysis
grad-CAM
gastrointestinal polyp
semantic segmentation
url https://ieeexplore.ieee.org/document/11095343/
work_keys_str_mv AT arjunkumarbosearnob xpolypnetaunetbasedmodelforsemanticsegmentationofgastrointestinalpolypswithexplainableai
AT muhammadmostafamonowar xpolypnetaunetbasedmodelforsemanticsegmentationofgastrointestinalpolypswithexplainableai
AT mdabdulhamid xpolypnetaunetbasedmodelforsemanticsegmentationofgastrointestinalpolypswithexplainableai
AT mfmridha xpolypnetaunetbasedmodelforsemanticsegmentationofgastrointestinalpolypswithexplainableai