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

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.

[1]  Joon Beom Seo,et al.  Multi-task vision transformer using low-level chest X-ray feature corpus for COVID-19 diagnosis and severity quantification , 2021, Medical Image Analysis.

[2]  N. Adami,et al.  BS-Net: Learning COVID-19 pneumonia severity on a large chest X-ray dataset , 2021, Medical Image Analysis.

[3]  Kiran Kumar Chandriah,et al.  Maximizing a deep submodular function optimization with a weighted MAX-SAT problem for trajectory clustering and motion segmentation , 2021, Applied Intelligence.

[4]  Yan Wang,et al.  TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation , 2021, ArXiv.

[5]  Priyanka Mary Mammen,et al.  Federated Learning: Opportunities and Challenges , 2021, ArXiv.

[6]  Fahad Shahbaz Khan,et al.  Transformers in Vision: A Survey , 2021, ACM Comput. Surv..

[7]  Jingwei Sun,et al.  Soteria: Provable Defense against Privacy Leakage in Federated Learning from Representation Perspective , 2020, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Chang Xu,et al.  Pre-Trained Image Processing Transformer , 2020, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Jong Chul Ye,et al.  Deep learning for tomographic image reconstruction , 2020, Nature Machine Intelligence.

[10]  S. Gelly,et al.  An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale , 2020, ICLR.

[11]  Angelica I. Avilés-Rivero,et al.  Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scans , 2020, Nature Machine Intelligence.

[12]  Jerry L Prince,et al.  A Review of Deep Learning in Medical Imaging: Imaging Traits, Technology Trends, Case Studies With Progress Highlights, and Future Promises , 2020, Proceedings of the IEEE.

[13]  Haifang Li,et al.  Deep transfer learning artificial intelligence accurately stages COVID-19 lung disease severity on portable chest radiographs , 2020, PloS one.

[14]  Daniel J. Beutel,et al.  Flower: A Friendly Federated Learning Research Framework , 2020, 2007.14390.

[15]  Ender Konukoglu,et al.  Joint reconstruction and bias field correction for undersampled MR imaging , 2020, MICCAI.

[16]  Miguel Cazorla,et al.  BIMCV COVID-19+: a large annotated dataset of RX and CT images from COVID-19 patients , 2020, ArXiv.

[17]  Yi Li,et al.  Weakly Supervised Lesion Localization With Probabilistic-CAM Pooling , 2020, ArXiv.

[18]  Dmytro Poplavskiy,et al.  Deep Learning for Automatic Pneumonia Detection , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[19]  R. Maroldi,et al.  COVID-19 outbreak in Italy: experimental chest X-ray scoring system for quantifying and monitoring disease progression , 2020, La radiologia medica.

[20]  Surya Nepal,et al.  End-to-End Evaluation of Federated Learning and Split Learning for Internet of Things , 2020, 2020 International Symposium on Reliable Distributed Systems (SRDS).

[21]  Richard D Riley,et al.  Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal , 2020, BMJ.

[22]  Bo Zhao,et al.  iDLG: Improved Deep Leakage from Gradients , 2020, ArXiv.

[23]  W. Curran,et al.  Deep learning in medical image registration: a review , 2019, Physics in medicine and biology.

[24]  Aaron Carass,et al.  DeepHarmony: A deep learning approach to contrast harmonization across scanner changes. , 2019, Magnetic resonance imaging.

[25]  Xiaowei Ding,et al.  Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation , 2019, Medical Image Anal..

[26]  Anit Kumar Sahu,et al.  Federated Learning: Challenges, Methods, and Future Directions , 2019, IEEE Signal Processing Magazine.

[27]  Luc Rocher,et al.  Estimating the success of re-identifications in incomplete datasets using generative models , 2019, Nature Communications.

[28]  Joachim M. Buhmann,et al.  Variational Federated Multi-Task Learning , 2019, ArXiv.

[29]  Xiaodong Liu,et al.  Multi-Task Deep Neural Networks for Natural Language Understanding , 2019, ACL.

[30]  Yifan Yu,et al.  CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison , 2019, AAAI.

[31]  Yang Song,et al.  Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning , 2018, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[32]  Ramesh Raskar,et al.  Split learning for health: Distributed deep learning without sharing raw patient data , 2018, ArXiv.

[33]  Sebastian Caldas,et al.  LEAF: A Benchmark for Federated Settings , 2018, ArXiv.

[34]  Hubert Eichner,et al.  Federated Learning for Mobile Keyboard Prediction , 2018, ArXiv.

[35]  Yasaman Khazaeni,et al.  Probabilistic Federated Neural Matching , 2018 .

[36]  Spyridon Bakas,et al.  Multi-Institutional Deep Learning Modeling Without Sharing Patient Data: A Feasibility Study on Brain Tumor Segmentation , 2018, BrainLes@MICCAI.

[37]  Geraint Rees,et al.  Clinically applicable deep learning for diagnosis and referral in retinal disease , 2018, Nature Medicine.

[38]  Ramesh Raskar,et al.  Distributed learning of deep neural network over multiple agents , 2018, J. Netw. Comput. Appl..

[39]  Ramakanth Pasunuru,et al.  Soft Layer-Specific Multi-Task Summarization with Entailment and Question Generation , 2018, ACL.

[40]  Shaohua Kevin Zhou,et al.  Less is More: Simultaneous View Classification and Landmark Detection for Abdominal Ultrasound Images , 2018, MICCAI.

[41]  Bruce R. Rosen,et al.  Distributed deep learning networks among institutions for medical imaging , 2018, J. Am. Medical Informatics Assoc..

[42]  Andrew J. Davison,et al.  End-To-End Multi-Task Learning With Attention , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Alexander Rakhlin,et al.  Automatic Instrument Segmentation in Robot-Assisted Surgery Using Deep Learning , 2018, bioRxiv.

[44]  Pengtao Xie,et al.  On the Automatic Generation of Medical Imaging Reports , 2017, ACL.

[45]  C. Pal,et al.  Deep Learning: A Primer for Radiologists. , 2017, Radiographics : a review publication of the Radiological Society of North America, Inc.

[46]  Sarvar Patel,et al.  Practical Secure Aggregation for Privacy-Preserving Machine Learning , 2017, IACR Cryptol. ePrint Arch..

[47]  H. Brendan McMahan,et al.  Learning Differentially Private Recurrent Language Models , 2017, ICLR.

[48]  Kaiming He,et al.  Focal Loss for Dense Object Detection , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[49]  Chen Sun,et al.  Revisiting Unreasonable Effectiveness of Data in Deep Learning Era , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[50]  Shiho Moriai,et al.  Privacy-Preserving Deep Learning: Revisited and Enhanced , 2017, ATIS.

[51]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[52]  Ameet Talwalkar,et al.  Federated Multi-Task Learning , 2017, NIPS.

[53]  Joachim Bingel,et al.  Latent Multi-Task Architecture Learning , 2017, AAAI.

[54]  Andrea Vedaldi,et al.  Learning multiple visual domains with residual adapters , 2017, NIPS.

[55]  Le Lu,et al.  ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[56]  Peter Richtárik,et al.  Federated Learning: Strategies for Improving Communication Efficiency , 2016, ArXiv.

[57]  A. Gupta,et al.  Cross-Stitch Networks for Multi-task Learning , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[58]  Blaise Agüera y Arcas,et al.  Communication-Efficient Learning of Deep Networks from Decentralized Data , 2016, AISTATS.

[59]  Quoc V. Le,et al.  Multi-task Sequence to Sequence Learning , 2015, ICLR.

[60]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[61]  Polina Golland,et al.  BrainPrint: A discriminative characterization of brain morphology , 2015, NeuroImage.

[62]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[63]  R. Fergus,et al.  Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[64]  J. Grefenstette,et al.  A systematic review of barriers to data sharing in public health , 2014, BMC Public Health.

[65]  Massimiliano Pontil,et al.  Regularized multi--task learning , 2004, KDD.

[66]  Phillip Rogaway,et al.  Authenticated-encryption with associated-data , 2002, CCS '02.

[67]  Rich Caruana,et al.  Multitask Learning , 1997, Machine Learning.

[68]  T. Tasdizen,et al.  Segmentation , 2014, International Journal of Computer Assisted Radiology and Surgery.