Federated and secure cloud services for building medical image classifiers on an intercontinental infrastructure

Autor(es):Mostrar menos -

Blanquer, Ignacio ^[1] ; Brasileiro, Francisco ^[2] ; Brito, Andrey ^[2] ; Calatrava, Amanda ^[1] ; Carvalho, Andre ^[3] ; Fetzer, Christof ^[4] ; Figueiredo, Flavio ^[5] ; Guimaraes, Ronny Petterson ^[3] ; Marinho, Leandro ^[2] ; Meira, Jr., Wagner ^[5] ; Silva, Altigran ^[3] ; Alberich-Bayarri, Angel ^[6] ; Camacho-Ramos, Eduardo ^[6] ; Jimenez-Pastor, Ana ^[6] ; Ribeiro, Antonio Luiz L. ^[5] ; Nascimento, Bruno Ramos ^[5] ; Silva, Fabio ^[2]

Afiliação do(s) autor(es)

^[1] Univ Politecn Valencia, Valencia - Spain

^[2] Univ Fed Campina Grande, Campina Grande, Paraiba - Brazil

^[3] Univ Fed Amazonas, Manaus, Amazonas - Brazil

^[4] Tech Univ Dresden, Dresden - Germany

^[5] Univ Fed Minas Gerais, Belo Horizonte, MG - Brazil

^[6] QUIBIM, Valencia - Spain

• Tipo de documento: Artigo Científico
• Fonte: FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE; v. 110, p. 119-134, SEP 2020.
• Citações Web of Science: 0

Resumo

Medical data processing has found a new dimension with the extensive use of machine-learning techniques to classify and extract features. Machine learning strongly benefits from computing accelerators. However, such accelerators are not easily available at hospital premises, although they can be easily found on public cloud infrastructures or research centers. Nevertheless, the sensitivity of medical data poses several challenges on the access to such data, requiring security guarantees and isolation. In this paper we present an architecture that addresses this problem. It keeps critical data encrypted in memory and disk, which can only be accessed inside trusted execution environments protected by hardware extensions. Data is anonymized inside these environments and securely transferred to external sites that host accelerator devices, keeping the same network space and reducing security risks even in untrusted cloud backends. Results on the processing of data in different scenarios are presented and discussed. The results are demonstrated on a geographically-wide deployment provided by the ATMOSPHERE project. (C) 2020 Elsevier B.V. All rights reserved. (AU)

• Processo FAPESP: 15/24461-2 - Estudo de modelos de negócios para federação de serviços para suporte a e-Ciência
• Beneficiário: Francisco Vilar Brasileiro
• Modalidade de apoio: Auxílio à Pesquisa - Temático