What does this research mean for the field?

Deep learning models, particularly hybrid and ensemble frameworks, achieve high accuracy in skin cancer classification and segmentation, but their clinical applicability is currently hindered by persistent challenges such as class imbalance, data leakage, and insufficient clinical validation. Novelty: ClaimNovelty.SYNTHESIS. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This review aims to evaluate the effectiveness of deep learning applications in skin cancer detection and diagnosis.

June 1, 2026Open Access

A comprehensive review of deep learning applications in the segmentation and classification of skin cancer

Key Points

This review aims to evaluate the effectiveness of deep learning applications in skin cancer detection and diagnosis.
Systematic review of 77 studies from databases including Scopus, IEEE, PubMed, and MDPI.
Analysis of segmentation and classification techniques, primarily focusing on convolutional neural networks and U-Net frameworks.
Assessment of performance metrics and risk-of-bias across studies, with data drawn from benchmark datasets like ISIC and HAM10000.
Classification approaches achieved an average accuracy of 96% on the ISIC dataset and 93% on the HAM10000 dataset.
Hybrid and ensemble frameworks outperformed standalone models in segmentation and classification tasks.
Identified challenges such as class imbalance and inadequate clinical validation affecting the reliability of AI applications.

Abstract

Skin cancer (SC) is one of the most prevalent forms of cancer worldwide. Both melanoma and non-melanoma types pose major challenges for early detection, accurate diagnosis, and proper treatment. Conventional diagnostic approaches, such as biopsy and visual examination, are often time-consuming, subjective, and prone to human error. Recent advances in artificial intelligence (AI) and deep learning (DL) have greatly improved the accuracy of SC diagnosis. This systematic review explores the applications of DL techniques in the segmentation and classification of skin lesions between 2014 and 2024. Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines and applying predefined inclusion and exclusion criteria, a total of 77 experimental studies out of 540 were analyzed from major databases, including Scopus, IEEE, PubMed, and MDPI. Convolutional neural networks (CNNs) were identified as the most widely used for classification, while U-Net and its variants dominated segmentation tasks. Hybrid and ensemble frameworks demonstrated superior performance across benchmark datasets, like the ISIC archive and HAM10000. Moreover, this work incorporates a formal risk-of-bias analysis, revealing critical concerns about class imbalance and data leakage. Almost all reviewed studies for the classification task achieved an average accuracy of 96% for the ISIC dataset, while the HAM10000 dataset attained an average accuracy of 93%. Despite these advances, challenges such as class imbalance, limited dataset diversity, and insufficient clinical validation persist. Addressing these issues through data augmentation, explainable AI, and federated learning could further enhance the generalizability and clinical applicability of AI-driven diagnosis systems. Additionally, this study identifies a clear paradigm shift from standalone CNNs to hybrid frameworks and multi-source feature fusion strategies, aiming to improve SC diagnosis.

Read Full Paperexternally

Demander à l'IA

Bookmark

View Full Paper

Cite This Study

Saleh et al. (Fri,) studied this question.

synapsesocial.com/papers/6a1d218f02fbce91306378a3 https://doi.org/https://doi.org/10.1088/2057-1976/ae74d6

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

Demander à l'IA

Bookmark

View Full Paper