Hybrid CNN–Vision Transformer Architecture for Accurate and Early Skin Cancer Detection

Read full text Summarise with VIRAT
Authors: Aya E. Fawzy, Mohammed E. Almandouh, M. Herajy, Mohamed Eisa
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Skin cancer is one of the major and fastest growing global public health issues. The most crucial factor for the patient's survival is the early and accurate detection of skin cancer. Unfortunately, it is still challenging for dermatologists to distinguish skin cancer lesions from skin benign ones visually since lesion types look very similar and medical professionals are overloaded with diagnostic cases. As a result of these challenges, many intelligent computers aided diagnostic systems that use deep learning techniques for medical image analysis have been developed. This paper presents a hybrid deep learning method designed to perform skin lesion classification automatically from images, which combines Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). The model utilizes CNNs for extracting local spatial features and the Vision Transformers abilities to obtain global contextual information. The proposed method has been trained and validated on a public dermoscopic image benchmark dataset HAM10000 containing a total of 10015 images. From the experimental results, the single model controlled to reach the classification accuracy of 94.24% which makes it a very promising computer aided tool for the early detection of skin cancer.

Fields of Study & Subjects

Cutaneous Melanoma Detection and Management AI in cancer detection Advanced Neural Network Applications

Publication details

Journal / source
Alfarama Journal of Basic & Applied Sciences
Volume
0
Issue
0
Publication date
26 March 2026
Language
EN