A novel approach for melanoma detection utilizing GAN synthesis and vision transformer

Rui Wang; Xiaofei Chen; Xiangyang Wang; Haiquan Wang; Chunhua Qian; Liucheng Yao; Kecheng Zhang

doi:10.1016/j.compbiomed.2024.108572

A novel approach for melanoma detection utilizing GAN synthesis and vision transformer

Comput Biol Med. 2024 Jun:176:108572. doi: 10.1016/j.compbiomed.2024.108572. Epub 2024 May 9.

Authors

Rui Wang¹, Xiaofei Chen², Xiangyang Wang³, Haiquan Wang⁴, Chunhua Qian⁵, Liucheng Yao⁶, Kecheng Zhang⁷

Affiliations

¹ School of Communication and Information Engineering, Shanghai University, No. 99 Shangda Rd., Shanghai, 200444, China. Electronic address: rwang@shu.edu.cn.
² School of Communication and Information Engineering, Shanghai University, No. 99 Shangda Rd., Shanghai, 200444, China. Electronic address: Chenxiaofer@shu.edu.cn.
³ School of Communication and Information Engineering, Shanghai University, No. 99 Shangda Rd., Shanghai, 200444, China. Electronic address: wangxiangyang@shu.edu.cn.
⁴ Department of General Surgery, Shanghai General Hospital of Shanghai Jiaotong University, No. 100 Haining Rd., Shanghai, 200080, China. Electronic address: wanghaiquan2008@163.com.
⁵ Department of Endocrinology and Metabolism, Shanghai Tenth People's Hospital, Tongji University, School of Medicine, No. 301 Middle Yanchang Rd., Shanghai, 200072, China. Electronic address: chqian2003@126.com.
⁶ School of Communication and Information Engineering, Shanghai University, No. 99 Shangda Rd., Shanghai, 200444, China. Electronic address: 1579909850@shu.edu.cn.
⁷ School of Communication and Information Engineering, Shanghai University, No. 99 Shangda Rd., Shanghai, 200444, China. Electronic address: zkc111111@shu.edu.cn.

PMID: 38749327
DOI: 10.1016/j.compbiomed.2024.108572

Abstract

Background and objective: Melanoma, a malignant form of skin cancer, is a critical health concern worldwide. Early and accurate detection plays a pivotal role in improving patient's conditions. Current diagnosis of skin cancer largely relies on visual inspections such as dermoscopy examinations, clinical screening and histopathological examinations. However, these approaches are characterized by low efficiency, high costs, and a lack of guaranteed accuracy. Consequently, deep learning based techniques have emerged in the field of melanoma detection, successfully aiding in improving the accuracy of diagnosis. However, the high similarity between benign and malignant melanomas, combined with the class imbalance issue in skin lesion datasets, present a significant challenge in further improving the diagnosis accuracy. We propose a two-stage framework for melanoma detection to address these issues.

Methods: In the first stage, we use Style Generative Adversarial Networks with Adaptive discriminator augmentation synthesis to generate realistic and diverse melanoma images, which are then combined with the original dataset to create an augmented dataset. In the second stage, we utilize a vision Transformer of BatchFormer to extract features and detect melanoma or non-melanoma skin lesions on the augmented dataset obtained in the previous step, specifically, we employed a dual-branch training strategy in this process.

Results: Our experimental results on the ISIC2020 dataset demonstrate the effectiveness of the proposed approach, showing a significant improvement in melanoma detection. The method achieved an accuracy of 98.43%, an AUC value of 98.63%, and an F1 value of 99.01%, surpassing some existing methods.

Conclusion: The method is feasible, efficient, and achieves early melanoma screening. It significantly enhances detection accuracy and can assist physicians in diagnosis to a great extent.

Keywords: BatchFormer; Condition-StyleGAN2-ADA; Melanoma detection; Vision transformer.

MeSH terms

Deep Learning
Dermoscopy / methods
Humans
Image Interpretation, Computer-Assisted / methods
Melanoma* / diagnosis
Melanoma* / diagnostic imaging
Skin Neoplasms* / diagnosis
Skin Neoplasms* / diagnostic imaging