Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification

Skin cancer, with its rising global prevalence, remains a crucial healthcare challenge, necessitating efficient and early detection for better patient outcomes. While deep convolutional neural networks have advanced image classification, current models struggle with diverse lesion types, variable im...

Full description

Saved in:
Bibliographic Details
Main Authors: Iftekhar Ahmed, Biggo Bushon Routh, Md. Saidur Rahman Kohinoor, Shadman Sakib, Md Mahfuzur Rahman, Farag Azzedin
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10772229/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850164945990189056
author Iftekhar Ahmed
Biggo Bushon Routh
Md. Saidur Rahman Kohinoor
Shadman Sakib
Md Mahfuzur Rahman
Farag Azzedin
author_facet Iftekhar Ahmed
Biggo Bushon Routh
Md. Saidur Rahman Kohinoor
Shadman Sakib
Md Mahfuzur Rahman
Farag Azzedin
author_sort Iftekhar Ahmed
collection DOAJ
description Skin cancer, with its rising global prevalence, remains a crucial healthcare challenge, necessitating efficient and early detection for better patient outcomes. While deep convolutional neural networks have advanced image classification, current models struggle with diverse lesion types, variable image quality, and dataset imbalances. Artifacts like hair can further obscure important features. This research addresses the problem and introduces a novel deep learning approach for accurate skin cancer classification by combining ResNet50V2, MobileNetV2, and EfficientNetV2 models. Our proposed architecture leverages the unique feature extraction capabilities of these models. It incorporates an attention mechanism to dynamically emphasize relevant features, enhancing focus, and promoting synergistic interactions among diverse feature sets. As such, our ensemble architectural approach outperforms other state-of-the-art models with high precision, recall, and F1-score metrics. Notably, the model demonstrates significant precision for Dermatofibroma (92% with 96% recall) and Vascular lesions (99% with 99% recall), highlighting its robustness across varied lesion types. Additionally, comprehensive image preprocessing techniques, including image resampling, black-hat filtering, thresholding, morphological closing, inpainting, and overall hair artifact removal, ensure the dataset’s quality and reliability. Validated through statistical significance testing and prototyped with an mHealth solution, this research heralds a significant stride in skin cancer diagnosis with the potential of attention-enhanced ensemble architectures.
format Article
id doaj-art-bcaeea5ab7cb41a2835ee88327fba4e9
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-bcaeea5ab7cb41a2835ee88327fba4e92025-08-20T02:21:51ZengIEEEIEEE Access2169-35362024-01-011218100918102410.1109/ACCESS.2024.351022410772229Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer ClassificationIftekhar Ahmed0https://orcid.org/0009-0004-6081-0707Biggo Bushon Routh1Md. Saidur Rahman Kohinoor2https://orcid.org/0000-0002-0405-8539Shadman Sakib3https://orcid.org/0000-0002-8794-1414Md Mahfuzur Rahman4https://orcid.org/0000-0002-4334-6402Farag Azzedin5https://orcid.org/0000-0001-9712-439XDepartment of Computer Science and Engineering, Leading University, Sylhet, BangladeshDepartment of Computer Science and Engineering, Leading University, Sylhet, BangladeshDepartment of Information and Computer Science, King Fahd University of Petroleum and Minerals, Dhahran, Saudi ArabiaInterdisciplinary Computer Science (InteX) Research Lab, Sylhet, BangladeshDepartment of Information and Computer Science, King Fahd University of Petroleum and Minerals, Dhahran, Saudi ArabiaDepartment of Information and Computer Science, King Fahd University of Petroleum and Minerals, Dhahran, Saudi ArabiaSkin cancer, with its rising global prevalence, remains a crucial healthcare challenge, necessitating efficient and early detection for better patient outcomes. While deep convolutional neural networks have advanced image classification, current models struggle with diverse lesion types, variable image quality, and dataset imbalances. Artifacts like hair can further obscure important features. This research addresses the problem and introduces a novel deep learning approach for accurate skin cancer classification by combining ResNet50V2, MobileNetV2, and EfficientNetV2 models. Our proposed architecture leverages the unique feature extraction capabilities of these models. It incorporates an attention mechanism to dynamically emphasize relevant features, enhancing focus, and promoting synergistic interactions among diverse feature sets. As such, our ensemble architectural approach outperforms other state-of-the-art models with high precision, recall, and F1-score metrics. Notably, the model demonstrates significant precision for Dermatofibroma (92% with 96% recall) and Vascular lesions (99% with 99% recall), highlighting its robustness across varied lesion types. Additionally, comprehensive image preprocessing techniques, including image resampling, black-hat filtering, thresholding, morphological closing, inpainting, and overall hair artifact removal, ensure the dataset’s quality and reliability. Validated through statistical significance testing and prototyped with an mHealth solution, this research heralds a significant stride in skin cancer diagnosis with the potential of attention-enhanced ensemble architectures.https://ieeexplore.ieee.org/document/10772229/Skin cancerdeep learningtransfer learningmulti-model fusionfeature extractionimage preprocessing
spellingShingle Iftekhar Ahmed
Biggo Bushon Routh
Md. Saidur Rahman Kohinoor
Shadman Sakib
Md Mahfuzur Rahman
Farag Azzedin
Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
IEEE Access
Skin cancer
deep learning
transfer learning
multi-model fusion
feature extraction
image preprocessing
title Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
title_full Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
title_fullStr Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
title_full_unstemmed Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
title_short Multi-Model Attentional Fusion Ensemble for Accurate Skin Cancer Classification
title_sort multi model attentional fusion ensemble for accurate skin cancer classification
topic Skin cancer
deep learning
transfer learning
multi-model fusion
feature extraction
image preprocessing
url https://ieeexplore.ieee.org/document/10772229/
work_keys_str_mv AT iftekharahmed multimodelattentionalfusionensembleforaccurateskincancerclassification
AT biggobushonrouth multimodelattentionalfusionensembleforaccurateskincancerclassification
AT mdsaidurrahmankohinoor multimodelattentionalfusionensembleforaccurateskincancerclassification
AT shadmansakib multimodelattentionalfusionensembleforaccurateskincancerclassification
AT mdmahfuzurrahman multimodelattentionalfusionensembleforaccurateskincancerclassification
AT faragazzedin multimodelattentionalfusionensembleforaccurateskincancerclassification