Federated Split Vision Transformer for COVID-19 CXR Diagnosis using Task-Agnostic Training

被引:0
|
作者
Park, Sangjoon [1 ]
Kim, Gwanghyun [1 ]
Kim, Jeongsol [1 ]
Kim, Boah [1 ]
Ye, Jong Chul [1 ,2 ,3 ]
机构
[1] Korea Adv Inst Sci & Technol KAIST, Dept Bio & Brain Engn, Daejeon, South Korea
[2] Korea Adv Inst Sci & Technol KAIST, Kim Jaechul Grad Sch AI, Daejeon, South Korea
[3] Korea Adv Inst Sci & Technol KAIST, Dept Math Sci, Daejeon, South Korea
来源
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021) | 2021年 / 34卷
基金
新加坡国家研究基金会;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Federated learning, which shares the weights of the neural network across clients, is gaining attention in the healthcare sector as it enables training on a large corpus of decentralized data while maintaining data privacy. For example, this enables neural network training for COVID-19 diagnosis on chest X-ray (CXR) images without collecting patient CXR data across multiple hospitals. Unfortunately, the exchange of the weights quickly consumes the network bandwidth if highly expressive network architecture is employed. So-called split learning partially solves this problem by dividing a neural network into a client and a server part, so that the client part of the network takes up less extensive computation resources and bandwidth. However, it is not clear how to find the optimal split without sacrificing the overall network performance. To amalgamate these methods and thereby maximize their distinct strengths, here we show that the Vision Transformer, a recently developed deep learning architecture with straightforward decomposable configuration, is ideally suitable for split learning without sacrificing performance. Even under the non-independent and identically distributed data distribution which emulates a real collaboration between hospitals using CXR datasets from multiple sources, the proposed framework was able to attain performance comparable to data-centralized training. In addition, the proposed framework along with heterogeneous multi-task clients also improves individual task performances including the diagnosis of COVID-19, eliminating the need for sharing large weights with innumerable parameters. Our results affirm the suitability of Transformer for collaborative learning in medical imaging and pave the way forward for future real-world implementations.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] Task-Agnostic Vision Transformer for Distributed Learning of Image Processing
    Kim, Boah
    Kim, Jeongsol
    Ye, Jong Chul
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2023, 32 : 203 - 218
  • [2] An interpretable multi-task system for clinically applicable COVID-19 diagnosis using CXR
    Zhuang, Yan
    Rahman, Md Fashiar
    Wen, Yuxin
    Pokojovy, Michael
    McCaffrey, Peter
    Vo, Alexander
    Walser, Eric
    Moen, Scott
    Xu, Honglun
    Tseng, Tzu-Liang
    JOURNAL OF X-RAY SCIENCE AND TECHNOLOGY, 2022, 30 (05) : 847 - 862
  • [3] Deep Learning Approaches for Automated Diagnosis of COVID-19 Using Imbalanced Training CXR Data
    Sharma, Ajay
    Mishra, Pramod Kumar
    ADVANCED NETWORK TECHNOLOGIES AND INTELLIGENT COMPUTING, ANTIC 2021, 2022, 1534 : 453 - 472
  • [4] Federated Learning for COVID-19 on Heterogeneous CXR Images with Noise
    Ding, Mengqing
    Li, Juan
    Yi, Changyan
    Cai, Jun
    ICC 2023-IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS, 2023, : 3413 - 3418
  • [5] Towards robust diagnosis of COVID-19 using vision self-attention transformer
    Fozia Mehboob
    Abdul Rauf
    Richard Jiang
    Abdul Khader Jilani Saudagar
    Khalid Mahmood Malik
    Muhammad Badruddin Khan
    Mozaherul Hoque Abdul Hasnat
    Abdullah AlTameem
    Mohammed AlKhathami
    Scientific Reports, 12
  • [6] Towards robust diagnosis of COVID-19 using vision self-attention transformer
    Mehboob, Fozia
    Rauf, Abdul
    Jiang, Richard
    Saudagar, Abdul Khader Jilani
    Malik, Khalid Mahmood
    Khan, Muhammad Badruddin
    Hasnat, Mozaherul Hoque Abdul
    AlTameem, Abdullah
    AlKhathami, Mohammed
    SCIENTIFIC REPORTS, 2022, 12 (01)
  • [7] COVID-Transformer: Interpretable COVID-19 Detection Using Vision Transformer for Healthcare
    Shome, Debaditya
    Kar, T.
    Mohanty, Sachi Nandan
    Tiwari, Prayag
    Muhammad, Khan
    AlTameem, Abdullah
    Zhang, Yazhou
    Saudagar, Abdul Khader Jilani
    INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH, 2021, 18 (21)
  • [8] Deep Learning COVID-19 Features on CXR Using Limited Training Data Sets
    Oh, Yujin
    Park, Sangjoon
    Ye, Jong Chul
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2020, 39 (08) : 2688 - 2700
  • [9] A CNN-transformer fusion network for COVID-19 CXR image classification
    Cao, Kai
    Deng, Tao
    Zhang, Chuanlin
    Lu, Limeng
    Li, Lin
    PLOS ONE, 2022, 17 (10):
  • [10] CXR-based Diagnosis of COVID-19 using Deep Learning with CycleGAN for Data Augmentation
    Chirila, Laura
    Cristea, Darius-Luca
    Banias, Ovidiu
    2021 INTERNATIONAL CONFERENCE ON E-HEALTH AND BIOENGINEERING (EHB 2021), 9TH EDITION, 2021,