Progressive Transfer Learning and Adversarial Domain Adaptation for Cross-Domain Skin Disease Classification

被引:64
|
作者
Gu, Yanyang [1 ]
Ge, Zongyuan [2 ]
Bonnington, C. Paul [2 ,3 ]
Zhou, Jun [1 ]
机构
[1] Griffith Univ, Nathan, Qld 4111, Australia
[2] Monash Univ, Clayton, Vic 3800, Australia
[3] Airdoc, Shanghai 200000, Peoples R China
关键词
Skin; Adaptation models; Melanoma; Deep learning; Data models; Automatic melanoma detection; dermoscopy image; cycle-GAN; deep learning; transfer learning; domain adaptation; DEEP; SEGMENTATION;
D O I
10.1109/JBHI.2019.2942429
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Deep learning has been used to analyze and diagnose various skin diseases through medical imaging. However, recent researches show that a well-trained deep learning model may not generalize well to data from different cohorts due to domain shift. Simple data fusion techniques such as combining disease samples from different data sources are not effective to solve this problem. In this paper, we present two methods for a novel task of cross-domain skin disease recognition. Starting from a fully supervised deep convolutional neural network classifier pre-trained on ImageNet, we explore a two-step progressive transfer learning technique by fine-tuning the network on two skin disease datasets. We then propose to adopt adversarial learning as a domain adaptation technique to perform invariant attribute translation from source to target domain in order to improve the recognition performance. In order to evaluate these two methods, we analyze generalization capability of the trained model on melanoma detection, cancer detection, and cross-modality learning tasks on two skin image datasets collected from different clinical settings and cohorts with different disease distributions. The experiments prove the effectiveness of our method in solving the domain shift problem.
引用
收藏
页码:1379 / 1393
页数:15
相关论文
共 50 条
  • [1] Adversarial Domain Adaptation with Semantic Consistency for Cross-Domain Image Classification
    Cao, Manliang
    Zhou, Xiangdong
    Xu, Yiming
    Pang, Yue
    Yao, Bo
    [J]. PROCEEDINGS OF THE 28TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT (CIKM '19), 2019, : 259 - 268
  • [2] Graph Domain Adversarial Transfer Network for Cross-Domain Sentiment Classification
    Tang, Hengliang
    Mi, Yuan
    Xue, Fei
    Cao, Yang
    [J]. IEEE ACCESS, 2021, 9 (09): : 33051 - 33060
  • [3] Adversarial transfer learning for cross-domain visual recognition
    Wang, Shanshan
    Zhang, Lei
    Fu, Jingru
    [J]. KNOWLEDGE-BASED SYSTEMS, 2020, 204
  • [4] Unsupervised Energy-based Adversarial Domain Adaptation for Cross-domain Text Classification
    Zou, Han
    Yang, Jianfei
    Wu, Xiaojian
    [J]. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL-IJCNLP 2021, 2021, : 1208 - 1218
  • [5] Feature transfer based adversarial domain adaptation method for cross-domain road extraction
    Wang, Shuyang
    Mu, Xiaodong
    He, Hao
    Yang, Dongfang
    Zhao, Peng
    [J]. GEOCARTO INTERNATIONAL, 2022, 37 (02) : 445 - 455
  • [6] Cross-Domain Graph Convolutions for Adversarial Unsupervised Domain Adaptation
    Zhu, Ronghang
    Jiang, Xiaodong
    Lu, Jiasen
    Li, Sheng
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 34 (08) : 3847 - 3858
  • [7] Cross-Domain Adversarial Learning for Zero-Shot Classification
    Liu H.
    Zheng Q.
    Luo M.
    Zhao H.
    Xiao Y.
    Lü Y.
    [J]. Jisuanji Yanjiu yu Fazhan/Computer Research and Development, 2019, 56 (12): : 2521 - 2535
  • [8] Cross-domain Recommendation via Adversarial Adaptation
    Su, Hongzu
    Zhang, Yifei
    Yang, Xuejiao
    Hua, Hua
    Wang, Shuangyang
    Li, Jingjing
    [J]. PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2022, 2022, : 1808 - 1817
  • [9] Softly Associative Transfer Learning for Cross-Domain Classification
    Wang, Deqing
    Lu, Chenwei
    Wu, Junjie
    Liu, Hongfu
    Zhang, Wenjie
    Zhuang, Fuzhen
    Zhang, Hui
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2020, 50 (11) : 4709 - 4721
  • [10] Dyadic Transfer Learning for Cross-Domain Image Classification
    Wang, Hua
    Nie, Feiping
    Huang, Heng
    Ding, Chris
    [J]. 2011 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2011, : 551 - 556