An Improved VGG Model for Skin Cancer Detection

• Published in: Neural processing letters Vol. 55; no. 4; pp. 3715 - 3732
• Main Authors: Tabrizchi, Hamed, Parvizpour, Sepideh, Razmara, Jafar
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2023; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial Intelligence; Artificial neural networks; Automation; Cancer; Classification; Comparative studies; Complex Systems; Computational Intelligence; Computer Science; Deep learning; Dermatology; Literature reviews; Machine learning; Medical diagnosis; Medical imaging; Medical screening; Melanoma; Neural networks; Optimization; Performance evaluation; Skin cancer; Success; Computer-aided diagnosis systems; Convolutional neural network; VGG-16; Melanoma; Skin cancer
• ISSN: 1370-4621; 1573-773X
• DOI: 10.1007/s11063-022-10927-1

Skin cancer is one of the most prevalent malignancies in humans and is generally diagnosed through visual means. Since it is essential to detect this type of cancer in its early phases, one of the challenging tasks in developing and designing digital medical systems is the development of an automated classification system for skin lesions. For the automated detection of melanoma, a serious form of skin cancer, using dermoscopic images, convolutional neural network (CNN) models are getting noticed more than ever before. This study presents a new model for the early detection of skin cancer on the basis of processing dermoscopic images. The model works based on a well-known CNN-based architecture called the VGG-16 network. The proposed framework employs an enhanced architecture of VGG-16 to develop a model, which contributes to the improvement of accuracy in skin cancer detection. To evaluate the proposed technique, we have conducted a comparative study between our method and a number of previously introduced techniques on the International Skin Image Collaboration dataset. According to the results, the proposed model outperforms the compared alternative techniques in terms of accuracy.