A GAN-Based Data Augmentation Method for Imbalanced Multi-Class Skin Lesion Classification

被引:19
作者
Su, Qichen [1 ]
Hamed, Haza Nuzly Abdull [1 ]
Isa, Mohd Adham [1 ]
Hao, Xue [1 ]
Dai, Xin [1 ]
机构
[1] Univ Teknol Malaysia UTM, Fac Comp, Johor Baharu 81310, Malaysia
关键词
Skin; Training; Lesions; Image color analysis; Data augmentation; Generative adversarial networks; Data models; data imbalance; generative adversarial networks; skin lesion classification; DIAGNOSIS;
D O I
10.1109/ACCESS.2024.3360215
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Skin cancer is one of the most common types of cancer globally. Despite the remarkable advancements of deep learning methods in computer vision, automatic diagnosis of skin diseases still faces challenges such as limited data and class imbalance. Generative Adversarial Networks (GANs), which can synthesize realistic data, appear as an alternative to mitigate these issues. However, for imbalanced data, unconditional GANs either generate uneven data distribution or neglect universal knowledge of the whole dataset, while state-of-the-art (SOTA) conditional GANs suffer from performance degradation due to the mode collapse of minority classes. This paper proposes a two-stage GAN-based method to synthesize fine-grained and diverse 256x256 pixels skin lesion images for the imbalanced dataset, named Self-Transfer GAN (STGAN). STGAN first learns universal knowledge from all classes then transfers this shared knowledge to each class and fuses it with class-specific knowledge to synthesize high-quality images. Furthermore, based on STGAN, a framework to enhance the classification performance is established. Both data generation and classification tasks are evaluated on HAM10000 dataset. In terms of Frechet Inception Distance (FID), Inception Score, Precision, and Recall, STGAN improved by 16%, 16%, 4%, 33% compared with SOTA conditional StyleGAN2. For classification, the STGAN-based framework achieves remarkable results, with an Accuracy of 98.23%, Sensitivity of 88.85%, Precision of 90.23%, F1-score of 89.48%, and Specificity of 98.34%.
引用
收藏
页码:16498 / 16513
页数:16
相关论文
共 50 条
[41]   Data Fusion Generative Adversarial Network for Multi-Class Imbalanced Fault Diagnosis of Rotating Machinery [J].
Liu, Qianjun ;
Ma, Guijun ;
Cheng, Cheng .
IEEE ACCESS, 2020, 8 :70111-70124
[42]   Freshwater Microscopic Algae Detection Based on Deep Neural Network with GAN-Based Augmentation for Imbalanced Algal Data [J].
Fung, Benjamin S. B. ;
Chan, Wang Hin ;
Lo, Irene M. C. ;
Tsang, Danny H. K. .
ACS ES&T WATER, 2023, 4 (03) :982-990
[43]   Deep CNN and Data Augmentation for Skin Lesion Classification [J].
Tri-Cong Pham ;
Luong, Chi-Mai ;
Visani, Muriel ;
Van-Dung Hoang .
INTELLIGENT INFORMATION AND DATABASE SYSTEMS, ACIIDS 2018, PT II, 2018, 10752 :573-582
[44]   Effectiveness of Data Augmentation in Multi-class Face Recognition [J].
Rahmatullah, Prinanda ;
Abidin, Taufik Fuadi ;
Misbullah, Alim ;
Nazaruddin .
2021 5TH INTERNATIONAL CONFERENCE ON INFORMATICS AND COMPUTATIONAL SCIENCES (ICICOS 2021), 2021,
[45]   An Integrated GAN-Based Approach to Imbalanced Disk Failure Data [J].
Yuan, Shuangshuang ;
Wu, Peng ;
Chen, Yuehui ;
Zhang, Liqiang ;
Wang, Jian .
INTELLIGENT COMPUTING THEORIES AND APPLICATION, ICIC 2022, PT II, 2022, 13394 :615-627
[46]   A NOVEL GAN-BASED DATA AUGMENTATION ALGORITHM FOR SEMICONDUCTOR DEFECT INSPECTION [J].
Liu, Yang ;
Guan, Yuanjun ;
Han, Tianyan ;
Ma, Can ;
Wang, Jiayi ;
Wang, Tao ;
Yi, Qianchuan ;
Hu, Lilei .
CONFERENCE OF SCIENCE & TECHNOLOGY FOR INTEGRATED CIRCUITS, 2024 CSTIC, 2024,
[47]   EnvGAN: a GAN-based augmentation to improve environmental sound classification [J].
Aswathy Madhu ;
Suresh K. .
Artificial Intelligence Review, 2022, 55 :6301-6320
[48]   Binary Imbalanced Data Classification Based on Modified D2GAN Oversampling and Classifier Fusion [J].
Zhai, Junhai ;
Qi, Jiaxing ;
Zhang, Sufang .
IEEE ACCESS, 2020, 8 :169456-169469
[49]   EFAM-Net: A Multi-Class Skin Lesion Classification Model Utilizing Enhanced Feature Fusion and Attention Mechanisms [J].
Ji, Zhanlin ;
Wang, Xuan ;
Liu, Chunling ;
Wang, Zhiwu ;
Yuan, Na ;
Ganchev, Ivan .
IEEE ACCESS, 2024, 12 :143029-143041
[50]   Data Augmentation Using Generative Adversarial Networks for Multi-Class Segmentation of Lung Confocal IF Images [J].
Katsuma, Daiki ;
Kawanaka, Hiroharu ;
Prasath, V. B. Surya ;
Aronow, Bruce J. .
JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS, 2022, 26 (02) :138-146