Show simple item record

dc.contributor.authorDaza, Laura
dc.contributor.authorPérez, Juan C.
dc.contributor.authorArbeláez, Pablo
dc.date.accessioned2021-10-04T06:19:03Z
dc.date.available2021-07-26T13:40:05Z
dc.date.available2021-10-04T06:19:03Z
dc.date.issued2021-09-21
dc.identifier.citationDaza, L., Pérez, J. C., & Arbeláez, P. (2021). Towards Robust General Medical Image Segmentation. Lecture Notes in Computer Science, 3–13. doi:10.1007/978-3-030-87199-4_1
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.doi10.1007/978-3-030-87199-4_1
dc.identifier.urihttp://hdl.handle.net/10754/670288
dc.description.abstractThe reliability of Deep Learning systems depends on their accuracy but also on their robustness against adversarial perturbations to the input data. Several attacks and defenses have been proposed to improve the performance of Deep Neural Networks under the presence of adversarial noise in the natural image domain. However, robustness in computer-aided diagnosis for volumetric data has only been explored for specific tasks and with limited attacks. We propose a new framework to assess the robustness of general medical image segmentation systems. Our contributions are two-fold: (i) we propose a new benchmark to evaluate robustness in the context of the Medical Segmentation Decathlon (MSD) by extending the recent AutoAttack natural image classification framework to the domain of volumetric data segmentation, and (ii) we present a novel lattice architecture for RObust Generic medical image segmentation (ROG). Our results show that ROG is capable of generalizing across different tasks of the MSD and largely surpasses the state-of-the-art under sophisticated adversarial attacks.
dc.description.sponsorshipWe thank Amazon Web Services (AWS) for a computational research grant used for the development of this project.
dc.publisherSpringer International Publishing
dc.relation.urlhttps://link.springer.com/10.1007/978-3-030-87199-4_1
dc.rightsArchived with thanks to Springer International Publishing
dc.subjectRobustness assessment
dc.subjectadversarial training
dc.subjectadversarial attacks
dc.subjectgeneral medical segmentation
dc.titleTowards Robust General Medical Image Segmentation
dc.typeConference Paper
dc.contributor.departmentKing Abdullah University of Science and Technology (KAUST), Saudi Arabia
dc.conference.dateSeptember 27 to October 1, 2021
dc.conference.nameMedical Image Computing and Computer Assisted Intervention – MICCAI 2021
dc.conference.locationVirtual
dc.eprint.versionPre-print
dc.contributor.institutionUniversidad de los Andes, Colombia
dc.identifier.pages3-13
dc.identifier.arxivid2107.04263
kaust.personPérez, Juan C.
refterms.dateFOA2021-10-04T06:19:03Z


Files in this item

Thumbnail
Name:
2107.04263.pdf
Size:
644.6Kb
Format:
PDF
Description:
Preprint

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version