Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images

Abstract Detecting skin melanoma in the early stage using dermoscopic images presents a complex challenge due to the inherent variability in images. Utilizing dermatology datasets, the study aimed to develop Automated Diagnostic Systems for early skin cancer detection. Existing methods often struggl...

Full description

Saved in:
Bibliographic Details
Main Authors: Md Abdullah All Mahmud, Sadia Afrin, M. F. Mridha, Sultan Alfarhood, Dunren Che, Mejdl Safran
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-09938-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849343020711804928
author Md Abdullah All Mahmud
Sadia Afrin
M. F. Mridha
Sultan Alfarhood
Dunren Che
Mejdl Safran
author_facet Md Abdullah All Mahmud
Sadia Afrin
M. F. Mridha
Sultan Alfarhood
Dunren Che
Mejdl Safran
author_sort Md Abdullah All Mahmud
collection DOAJ
description Abstract Detecting skin melanoma in the early stage using dermoscopic images presents a complex challenge due to the inherent variability in images. Utilizing dermatology datasets, the study aimed to develop Automated Diagnostic Systems for early skin cancer detection. Existing methods often struggle with diverse skin types, cancer stages, and imaging conditions, highlighting a critical gap in reliability and explainability. The novel approach proposed through this research addresses this gap by utilizing a proposed model with advanced layers, including Global Average Pooling, Batch Normalization, Dropout, and dense layers with ReLU and Swish activations to improve model performance. The proposed model achieved accuracies of 95.23% and 96.48% for the two different datasets, demonstrating its robustness, reliability, and strong performance across other performance metrics. Explainable AI techniques such as Gradient-weighted Class Activation Mapping and Saliency Maps offered insights into the model’s decision- making process. These advancements enhance skin cancer diagnostics, provide medical experts with resources for early detection, improve clinical outcomes, and increase acceptance of Deep Learning-based diagnostics in healthcare.
format Article
id doaj-art-951a67de79bb4b76bb2c519472f34ae9
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-951a67de79bb4b76bb2c519472f34ae92025-08-20T03:43:11ZengNature PortfolioScientific Reports2045-23222025-07-0115112010.1038/s41598-025-09938-4Explainable deep learning approaches for high precision early melanoma detection using dermoscopic imagesMd Abdullah All Mahmud0Sadia Afrin1M. F. Mridha2Sultan Alfarhood3Dunren Che4Mejdl Safran5Department of Computer Science, American International University-Bangladesh Department of Computer Science and Engineering, World University of BangladeshDepartment of Computer Science, American International University-BangladeshDepartment of Computer Science, College of Computer and Information Sciences, King Saud UniversityDepartment of Electrical Engineering and Computer Science, Texas A&M University-KingsvilleDepartment of Computer Science, College of Computer and Information Sciences, King Saud UniversityAbstract Detecting skin melanoma in the early stage using dermoscopic images presents a complex challenge due to the inherent variability in images. Utilizing dermatology datasets, the study aimed to develop Automated Diagnostic Systems for early skin cancer detection. Existing methods often struggle with diverse skin types, cancer stages, and imaging conditions, highlighting a critical gap in reliability and explainability. The novel approach proposed through this research addresses this gap by utilizing a proposed model with advanced layers, including Global Average Pooling, Batch Normalization, Dropout, and dense layers with ReLU and Swish activations to improve model performance. The proposed model achieved accuracies of 95.23% and 96.48% for the two different datasets, demonstrating its robustness, reliability, and strong performance across other performance metrics. Explainable AI techniques such as Gradient-weighted Class Activation Mapping and Saliency Maps offered insights into the model’s decision- making process. These advancements enhance skin cancer diagnostics, provide medical experts with resources for early detection, improve clinical outcomes, and increase acceptance of Deep Learning-based diagnostics in healthcare.https://doi.org/10.1038/s41598-025-09938-4Dermoscopic imagesDeep learningEarly-stage melanoma detectionExplainable AISwish activation
spellingShingle Md Abdullah All Mahmud
Sadia Afrin
M. F. Mridha
Sultan Alfarhood
Dunren Che
Mejdl Safran
Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
Scientific Reports
Dermoscopic images
Deep learning
Early-stage melanoma detection
Explainable AI
Swish activation
title Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
title_full Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
title_fullStr Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
title_full_unstemmed Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
title_short Explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
title_sort explainable deep learning approaches for high precision early melanoma detection using dermoscopic images
topic Dermoscopic images
Deep learning
Early-stage melanoma detection
Explainable AI
Swish activation
url https://doi.org/10.1038/s41598-025-09938-4
work_keys_str_mv AT mdabdullahallmahmud explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages
AT sadiaafrin explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages
AT mfmridha explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages
AT sultanalfarhood explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages
AT dunrenche explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages
AT mejdlsafran explainabledeeplearningapproachesforhighprecisionearlymelanomadetectionusingdermoscopicimages