MEDUSA: Multi-Scale Encoder-Decoder Self-Attention Deep Neural Network Architecture for Medical Image Analysis

Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance between diseases. In this study, we explore the concept of self-attention for tackling such subtleties in and between diseases. To this end,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Frontiers in medicine Jg. 8; S. 821120
Hauptverfasser: Aboutalebi, Hossein, Pavlova, Maya, Gunraj, Hayden, Shafiee, Mohammad Javad, Sabri, Ali, Alaref, Amer, Wong, Alexander
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Switzerland Frontiers Media SA 15.02.2022
Frontiers Media S.A
Schlagworte:
ISSN:2296-858X, 2296-858X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance between diseases. In this study, we explore the concept of self-attention for tackling such subtleties in and between diseases. To this end, we introduce, a multi-scale encoder-decoder self-attention (MEDUSA) mechanism tailored for medical image analysis. While self-attention deep convolutional neural network architectures in existing literature center around the notion of multiple isolated lightweight attention mechanisms with limited individual capacities being incorporated at different points in the network architecture, MEDUSA takes a significant departure from this notion by possessing a single, unified self-attention mechanism with significantly higher capacity with multiple attention heads feeding into different scales in the network architecture. To the best of the authors' knowledge, this is the first “single body, multi-scale heads” realization of self-attention and enables explicit global context among selective attention at different levels of representational abstractions while still enabling differing local attention context at individual levels of abstractions. With MEDUSA, we obtain state-of-the-art performance on multiple challenging medical image analysis benchmarks including COVIDx, Radiological Society of North America (RSNA) RICORD, and RSNA Pneumonia Challenge when compared to previous work. Our MEDUSA model is publicly available.
AbstractList Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance between diseases. In this study, we explore the concept of self-attention for tackling such subtleties in and between diseases. To this end, we introduce, a multi-scale encoder-decoder self-attention (MEDUSA) mechanism tailored for medical image analysis. While self-attention deep convolutional neural network architectures in existing literature center around the notion of multiple isolated lightweight attention mechanisms with limited individual capacities being incorporated at different points in the network architecture, MEDUSA takes a significant departure from this notion by possessing a single, unified self-attention mechanism with significantly higher capacity with multiple attention heads feeding into different scales in the network architecture. To the best of the authors' knowledge, this is the first “single body, multi-scale heads” realization of self-attention and enables explicit global context among selective attention at different levels of representational abstractions while still enabling differing local attention context at individual levels of abstractions. With MEDUSA, we obtain state-of-the-art performance on multiple challenging medical image analysis benchmarks including COVIDx, Radiological Society of North America (RSNA) RICORD, and RSNA Pneumonia Challenge when compared to previous work. Our MEDUSA model is publicly available.
Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance between diseases. In this study, we explore the concept of self-attention for tackling such subtleties in and between diseases. To this end, we introduce, a multi-scale encoder-decoder self-attention (MEDUSA) mechanism tailored for medical image analysis. While self-attention deep convolutional neural network architectures in existing literature center around the notion of multiple isolated lightweight attention mechanisms with limited individual capacities being incorporated at different points in the network architecture, MEDUSA takes a significant departure from this notion by possessing a single, unified self-attention mechanism with significantly higher capacity with multiple attention heads feeding into different scales in the network architecture. To the best of the authors' knowledge, this is the first "single body, multi-scale heads" realization of self-attention and enables explicit global context among selective attention at different levels of representational abstractions while still enabling differing local attention context at individual levels of abstractions. With MEDUSA, we obtain state-of-the-art performance on multiple challenging medical image analysis benchmarks including COVIDx, Radiological Society of North America (RSNA) RICORD, and RSNA Pneumonia Challenge when compared to previous work. Our MEDUSA model is publicly available.Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance between diseases. In this study, we explore the concept of self-attention for tackling such subtleties in and between diseases. To this end, we introduce, a multi-scale encoder-decoder self-attention (MEDUSA) mechanism tailored for medical image analysis. While self-attention deep convolutional neural network architectures in existing literature center around the notion of multiple isolated lightweight attention mechanisms with limited individual capacities being incorporated at different points in the network architecture, MEDUSA takes a significant departure from this notion by possessing a single, unified self-attention mechanism with significantly higher capacity with multiple attention heads feeding into different scales in the network architecture. To the best of the authors' knowledge, this is the first "single body, multi-scale heads" realization of self-attention and enables explicit global context among selective attention at different levels of representational abstractions while still enabling differing local attention context at individual levels of abstractions. With MEDUSA, we obtain state-of-the-art performance on multiple challenging medical image analysis benchmarks including COVIDx, Radiological Society of North America (RSNA) RICORD, and RSNA Pneumonia Challenge when compared to previous work. Our MEDUSA model is publicly available.
Author Sabri, Ali
Aboutalebi, Hossein
Wong, Alexander
Pavlova, Maya
Alaref, Amer
Shafiee, Mohammad Javad
Gunraj, Hayden
AuthorAffiliation 3 Department of Systems Design Engineering, University of Waterloo , Waterloo, ON , Canada
2 Waterloo AI Institute, University of Waterloo , Waterloo, ON , Canada
4 Department of Radiology, Niagara Health, McMaster University , Hamilton, ON , Canada
1 Department of Computer Science, University of Waterloo , Waterloo, ON , Canada
6 Department of Diagnostic Radiology, Thunder Bay Regional Health Sciences Centre , Thunder Bay, ON , Canada
5 Department of Diagnostic Imaging, Northern Ontario School of Medicine , Thunder Bay, ON , Canada
AuthorAffiliation_xml – name: 5 Department of Diagnostic Imaging, Northern Ontario School of Medicine , Thunder Bay, ON , Canada
– name: 3 Department of Systems Design Engineering, University of Waterloo , Waterloo, ON , Canada
– name: 1 Department of Computer Science, University of Waterloo , Waterloo, ON , Canada
– name: 2 Waterloo AI Institute, University of Waterloo , Waterloo, ON , Canada
– name: 4 Department of Radiology, Niagara Health, McMaster University , Hamilton, ON , Canada
– name: 6 Department of Diagnostic Radiology, Thunder Bay Regional Health Sciences Centre , Thunder Bay, ON , Canada
Author_xml – sequence: 1
  givenname: Hossein
  surname: Aboutalebi
  fullname: Aboutalebi, Hossein
– sequence: 2
  givenname: Maya
  surname: Pavlova
  fullname: Pavlova, Maya
– sequence: 3
  givenname: Hayden
  surname: Gunraj
  fullname: Gunraj, Hayden
– sequence: 4
  givenname: Mohammad Javad
  surname: Shafiee
  fullname: Shafiee, Mohammad Javad
– sequence: 5
  givenname: Ali
  surname: Sabri
  fullname: Sabri, Ali
– sequence: 6
  givenname: Amer
  surname: Alaref
  fullname: Alaref, Amer
– sequence: 7
  givenname: Alexander
  surname: Wong
  fullname: Wong, Alexander
BackLink https://www.ncbi.nlm.nih.gov/pubmed/35242769$$D View this record in MEDLINE/PubMed
BookMark eNp1kk1vEzEQhleoiH7QOye0EhcuG_y1uzYHpKgJEKmBQ6jEzbK949Rhsw62F9R_X6dpUVuJ01jjdx6PZ97T4mjwAxTFG4wmlHLxwW6hmxBE8IQTjAl6UZwQIpqK1_zn0aPzcXEe4wYhhCmpGaavimNaE0baRpwUw3I-u1pNP5bLsU-uWhnVQzkfjO8gVDO4i-UKeltNU4IhOT-UM4Bd-Q3GoPoc0l8ffpXTYK5dApPGAKX1oVxC5zKsXGzVGsrpoPqb6OLr4qVVfYTz-3hWXH2e_7j4Wl1-_7K4mF5WhgmUKkMVkIYTQXBOYI65rhuM2lYbwVRjGwUMdbjWQhheC21rBlZToZFVwvKanhWLA7fzaiN3wW1VuJFeOXmX8GEtVUjO9CBtxzNEc8qtZZwzTXinMGGaasyYwZn16cDajTqP3OQp5J8_gT69Gdy1XPs_knPetBRlwPt7QPC_R4hJbl000PdqAD9GSRra5KcYJVn67pl048eQhxclJS1pa0zwXvX2cUf_WnlYaxY0B4EJPsYAVhqX1H55uUHXS4zk3kJybyG5t5A8WCgXomeFD-z_ltwCDtDIjg
CitedBy_id crossref_primary_10_3390_s23052621
crossref_primary_10_3389_fmed_2022_861680
crossref_primary_10_1016_j_heliyon_2024_e30889
crossref_primary_10_3390_math11194084
crossref_primary_10_1016_j_bdr_2024_100477
crossref_primary_10_3390_s23198122
crossref_primary_10_3390_electronics12010080
Cites_doi 10.1136/thoraxjnl-2017-211280
10.1186/s12880-015-0069-9
10.1038/s41598-021-89352-8
10.1007/s42979-021-00496-w
10.1097/RLI.0000000000000574
10.1038/nature14539
10.1038/s41598-020-76550-z
10.1109/ICSIPA.2017.8120663
10.1148/radiol.2020202944
10.1016/S0140-6736(20)30183-5
10.1038/s41598-019-55972-4
10.1148/radiol.2021203957
10.1038/s41598-019-48995-4
10.1038/srep46349
10.1145/3465055
10.1109/CRV.2015.25
10.1007/978-3-319-46493-0_38
10.1145/3065386
10.1148/radiol.2020201160
10.1056/NEJMoa2002032
10.1561/9781601982957
10.2174/1573405616666200406110547
10.1007/s12652-021-03464-7
10.1109/TBME.2015.2485779
10.1007/978-3-030-36802-9_25
10.1038/s41598-021-88807-2
10.3390/cancers12020390
10.3389/frai.2021.764047
10.1038/s41586-021-03922-4
ContentType Journal Article
Copyright Copyright © 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong.
2022. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Copyright © 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong. 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong
Copyright_xml – notice: Copyright © 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong.
– notice: 2022. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
– notice: Copyright © 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong. 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong
DBID AAYXX
CITATION
NPM
3V.
7X7
7XB
88E
8C1
8FI
8FJ
8FK
ABUWG
AFKRA
AZQEC
BENPR
CCPQU
COVID
DWQXO
FYUFA
GHDGH
K9.
M0S
M1P
PHGZM
PHGZT
PIMPY
PJZUB
PKEHL
PPXIY
PQEST
PQQKQ
PQUKI
PRINS
7X8
5PM
DOA
DOI 10.3389/fmed.2021.821120
DatabaseName CrossRef
PubMed
ProQuest Central (Corporate)
Health & Medical Collection
ProQuest Central (purchase pre-March 2016)
Medical Database (Alumni Edition)
Public Health Database
Hospital Premium Collection
Hospital Premium Collection (Alumni Edition)
ProQuest Central (Alumni) (purchase pre-March 2016)
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central Essentials - QC
ProQuest Central
ProQuest One
Coronavirus Research Database
ProQuest Central
Health Research Premium Collection
Health Research Premium Collection (Alumni)
ProQuest Health & Medical Complete (Alumni)
ProQuest Health & Medical Collection
Medical Database
ProQuest Central Premium
ProQuest One Academic (New)
Publicly Available Content Database
ProQuest Health & Medical Research Collection
ProQuest One Academic Middle East (New)
ProQuest One Health & Nursing
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Academic (retired)
ProQuest One Academic UKI Edition
ProQuest Central China
MEDLINE - Academic
PubMed Central (Full Participant titles)
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
PubMed
Publicly Available Content Database
ProQuest One Academic Middle East (New)
ProQuest Central Essentials
ProQuest Health & Medical Complete (Alumni)
ProQuest Central (Alumni Edition)
ProQuest One Community College
ProQuest One Health & Nursing
ProQuest Central China
ProQuest Central
ProQuest Health & Medical Research Collection
Health Research Premium Collection
Health and Medicine Complete (Alumni Edition)
ProQuest Central Korea
Health & Medical Research Collection
ProQuest Central (New)
ProQuest Medical Library (Alumni)
ProQuest Public Health
ProQuest One Academic Eastern Edition
Coronavirus Research Database
ProQuest Hospital Collection
Health Research Premium Collection (Alumni)
ProQuest Hospital Collection (Alumni)
ProQuest Health & Medical Complete
ProQuest Medical Library
ProQuest One Academic UKI Edition
ProQuest One Academic
ProQuest One Academic (New)
ProQuest Central (Alumni)
MEDLINE - Academic
DatabaseTitleList CrossRef
PubMed
Publicly Available Content Database


MEDLINE - Academic
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: PIMPY
  name: Publicly Available Content Database
  url: http://search.proquest.com/publiccontent
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
EISSN 2296-858X
ExternalDocumentID oai_doaj_org_article_fd89c8b838ff4884b28da124b3b144c1
PMC8886730
35242769
10_3389_fmed_2021_821120
Genre Journal Article
GroupedDBID 53G
5VS
9T4
AAFWJ
AAYXX
ACGFS
ADBBV
ADRAZ
AFPKN
ALMA_UNASSIGNED_HOLDINGS
AOIJS
BAWUL
BCNDV
CITATION
DIK
GROUPED_DOAJ
HYE
KQ8
M48
M~E
OK1
PGMZT
RPM
ACXDI
IPNFZ
NPM
RIG
3V.
7X7
7XB
88E
8C1
8FI
8FJ
8FK
ABUWG
AFKRA
AZQEC
BENPR
CCPQU
COVID
DWQXO
FYUFA
K9.
M1P
PHGZM
PHGZT
PIMPY
PJZUB
PKEHL
PPXIY
PQEST
PQQKQ
PQUKI
PRINS
7X8
5PM
ID FETCH-LOGICAL-c490t-c3ae2682921c491818b561077bc94a6f6ae40d15b99c859bf54efb39b0fa9f853
IEDL.DBID DOA
ISICitedReferencesCount 10
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000783577600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2296-858X
IngestDate Fri Oct 03 12:49:10 EDT 2025
Thu Aug 21 14:10:01 EDT 2025
Fri Sep 05 12:41:41 EDT 2025
Wed Nov 19 00:43:46 EST 2025
Mon Jul 21 05:48:13 EDT 2025
Sat Nov 29 06:16:37 EST 2025
Tue Nov 18 21:44:54 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords COVID-19
chest X-ray (CXR)
diagnosis
deep neural net
computer vision
Language English
License Copyright © 2022 Aboutalebi, Pavlova, Gunraj, Shafiee, Sabri, Alaref and Wong.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c490t-c3ae2682921c491818b561077bc94a6f6ae40d15b99c859bf54efb39b0fa9f853
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Reviewed by: John Roubil, UMass Memorial Medical Center, United States; Ameer Elaimy, MetroWest Medical Center, United States
Edited by: Thomas J. FitzGerald, UMass Memorial Health Care, United States
This article was submitted to Translational Medicine, a section of the journal Frontiers in Medicine
OpenAccessLink https://doaj.org/article/fd89c8b838ff4884b28da124b3b144c1
PMID 35242769
PQID 3272751212
PQPubID 7426803
ParticipantIDs doaj_primary_oai_doaj_org_article_fd89c8b838ff4884b28da124b3b144c1
pubmedcentral_primary_oai_pubmedcentral_nih_gov_8886730
proquest_miscellaneous_2636144432
proquest_journals_3272751212
pubmed_primary_35242769
crossref_citationtrail_10_3389_fmed_2021_821120
crossref_primary_10_3389_fmed_2021_821120
PublicationCentury 2000
PublicationDate 2022-02-15
PublicationDateYYYYMMDD 2022-02-15
PublicationDate_xml – month: 02
  year: 2022
  text: 2022-02-15
  day: 15
PublicationDecade 2020
PublicationPlace Switzerland
PublicationPlace_xml – name: Switzerland
– name: Lausanne
PublicationTitle Frontiers in medicine
PublicationTitleAlternate Front Med (Lausanne)
PublicationYear 2022
Publisher Frontiers Media SA
Frontiers Media S.A
Publisher_xml – name: Frontiers Media SA
– name: Frontiers Media S.A
References Wong (B11) 2021; 7
Shen (B17) 2019; 9
Khalvati (B51) 2015; 15
Wong (B19) 2020; 296
Wong (B9) 2021
Manchin (B29) 2019
Goyal (B21) 2021; 18
Dosovitskiy (B38) 2020
Bahdanau (B27) 2014
Cameron (B53) 2015; 63
Wu (B33) 2020
Chorowski (B31) 2015
Bello (B34) 2019
Ba (B26) 2014
Khuzani (B46) 2021; 11
He (B4) 2015
Christe (B10) 2019; 54
Wang (B44) 2020; 10
Vig (B30) 2019
He (B43) 2016
Warren (B20) 2018; 73
Rajpurkar (B49) 2017
Kingma (B42) 2014
Hu (B37) 2019
Simonyan (B5) 2014
Hashemzadeh (B14) 2021; 11
Ronneberger (B40) 2015
Zagoruyko (B7) 2016
Sarkar (B45) 2021; 2
Wong (B52) 2021
LeCun (B1) 2015; 521
Hooda (B8) 2017
Vaswani (B25) 2017
Pavlova (B41) 2021
Yoo (B16) 2019; 9
Devlin (B28) 2018
Zahoor (B18) 2020; 16
Ramachandran (B35) 2019
Monti (B50) 2020; 12
Team (B47) 2018
Hu (B36)
Huang (B22) 2020; 395
Zhang (B12) 2017; 7
Tsai (B48) 2021; 299
Zhang (B24) 2021; 298
Chaudhari (B39) 2021; 12
Kumar (B13) 2015
Woo (B32) 2018
Krizhevsky (B6) 2012; 25
Guan (B23) 2020; 382
Bengio (B2) 2009
Glorot (B3) 2010
Elmarakeby (B15) 2021; 598
References_xml – start-page: 171105225
  year: 2017
  ident: B49
  article-title: Chexnet: radiologist-level pneumonia detection on chest X-rays with deep learning
  publication-title: arXiv preprint.
– start-page: 181108939
  year: 2018
  ident: B47
  article-title: Pneumonia detection in chest radiographs
  publication-title: arXiv preprint.
– volume: 73
  start-page: 840
  year: 2018
  ident: B20
  article-title: Severity scoring of lung oedema on the chest radiograph is associated with clinical outcomes in ARDS
  publication-title: Thorax.
  doi: 10.1136/thoraxjnl-2017-211280
– volume: 15
  start-page: 1
  year: 2015
  ident: B51
  article-title: Multiparametric MRI for prostate cancer detection: new insights into the combined use of a radiomic approach with advanced acquisition protocol
  publication-title: BMC Med Imag.
  doi: 10.1186/s12880-015-0069-9
– volume: 11
  start-page: 1
  year: 2021
  ident: B14
  article-title: A combined microfluidic deep learning approach for lung cancer cell high throughput screening toward automatic cancer screening applications
  publication-title: Sci Rep.
  doi: 10.1038/s41598-021-89352-8
– volume: 2
  start-page: 1
  year: 2021
  ident: B45
  article-title: Identification of images of COVID-19 from chest X-rays using deep learning: comparing COGNEX visionpro deep learning 1
  publication-title: SN Comput Sci.
  doi: 10.1007/s42979-021-00496-w
– start-page: 201011929
  year: 2020
  ident: B38
  article-title: An image is worth 16x16 words: transformers for image recognition at scale
  publication-title: arXiv preprint.
– volume: 54
  start-page: 627
  year: 2019
  ident: B10
  article-title: Computer-aided diagnosis of pulmonary fibrosis using deep learning and CT images
  publication-title: Investig Radiol.
  doi: 10.1097/RLI.0000000000000574
– year: 2021
  ident: B52
  article-title: Synthetic correlated diffusion imaging hyperintensity delineates clinically significant prostate cancer
– volume: 521
  start-page: 436
  year: 2015
  ident: B1
  article-title: Deep learning
  publication-title: Nature.
  doi: 10.1038/nature14539
– year: 2021
  ident: B9
  article-title: TB-Net: a tailored, self-attention deep convolutional neural network design for detection of tuberculosis cases from chest X-ray images
  publication-title: arXiv.
– volume: 10
  start-page: 19549
  year: 2020
  ident: B44
  article-title: COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images
  publication-title: Sci Rep.
  doi: 10.1038/s41598-020-76550-z
– start-page: 497
  volume-title: 2017 IEEE International Conference on Signal and Image Processing Applications (ICSIPA)
  year: 2017
  ident: B8
  article-title: Deep-learning: a potential method for tuberculosis detection using chest radiography
  doi: 10.1109/ICSIPA.2017.8120663
– volume: 298
  start-page: E88
  year: 2021
  ident: B24
  article-title: Diagnosis of coronavirus disease 2019 pneumonia by using chest radiography: value of artificial intelligence
  publication-title: Radiology.
  doi: 10.1148/radiol.2020202944
– start-page: 160507146
  year: 2016
  ident: B7
  article-title: Wide residual networks
  publication-title: arXiv preprint.
– volume: 395
  start-page: 497
  year: 2020
  ident: B22
  article-title: Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China
  publication-title: Lancet.
  doi: 10.1016/S0140-6736(20)30183-5
– volume: 9
  start-page: 1
  year: 2019
  ident: B16
  article-title: Prostate cancer detection using deep convolutional neural networks
  publication-title: Sci Rep.
  doi: 10.1038/s41598-019-55972-4
– start-page: 5998
  volume-title: Advances in Neural Information Processing Systems
  year: 2017
  ident: B25
  article-title: Attention is all you need
– start-page: 210506640
  year: 2021
  ident: B41
  article-title: COVID-Net CXR-2: an enhanced deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images
  publication-title: arXiv preprint.
– volume: 299
  start-page: 203957
  year: 2021
  ident: B48
  article-title: The RSNA international COVID-19 open annotated radiology database (RICORD)
  publication-title: Radiology.
  doi: 10.1148/radiol.2021203957
– volume: 9
  start-page: 1
  year: 2019
  ident: B17
  article-title: Deep learning to improve breast cancer detection on screening mammography
  publication-title: Sci Rep.
  doi: 10.1038/s41598-019-48995-4
– volume: 7
  start-page: 1
  year: 2017
  ident: B12
  article-title: Radiomics-based prognosis analysis for non-small cell lung cancer
  publication-title: Sci Rep.
  doi: 10.1038/srep46349
– volume: 12
  start-page: 1
  year: 2021
  ident: B39
  article-title: An attentive survey of attention models
  publication-title: ACM Trans Intell Syst Technol (TIST).
  doi: 10.1145/3465055
– start-page: 150607503
  year: 2015
  ident: B31
  article-title: Attention-based models for speech recognition
  publication-title: arXiv preprint.
– start-page: 190604284
  year: 2019
  ident: B30
  article-title: Analyzing the structure of attention in a transformer language model
  publication-title: arXiv preprint.
– start-page: 181004805
  year: 2018
  ident: B28
  article-title: Bert: Pre-training of deep bidirectional transformers for language understanding
  publication-title: arXiv preprint.
– start-page: 7132
  volume-title: Proceedings of the IEEE conference on computer vision and pattern recognition
  ident: B36
  article-title: Squeeze-and-excitation networks
– start-page: 133
  volume-title: 2015 12th Conference on Computer and Robot Vision
  year: 2015
  ident: B13
  article-title: Lung nodule classification using deep features in CT images
  doi: 10.1109/CRV.2015.25
– start-page: 630
  volume-title: Computer Vision – ECCV 2016
  year: 2016
  ident: B43
  article-title: Identity mappings in deep residual networks
  doi: 10.1007/978-3-319-46493-0_38
– start-page: 3
  volume-title: Proceedings of the European Conference on Computer Vision (ECCV)
  year: 2018
  ident: B32
  article-title: Cbam: convolutional block attention module
– volume: 25
  start-page: 1097
  year: 2012
  ident: B6
  article-title: Imagenet classification with deep convolutional neural networks
  publication-title: Adv Neural Inf Process Syst.
  doi: 10.1145/3065386
– start-page: 234
  volume-title: International Conference on Medical Image Computing and Computer-Assisted Intervention
  year: 2015
  ident: B40
  article-title: U-net: convolutional networks for biomedical image segmentation
– volume: 296
  start-page: E72
  year: 2020
  ident: B19
  article-title: Frequency and distribution of chest radiographic findings in COVID-19 positive patients
  publication-title: Radiology.
  doi: 10.1148/radiol.2020201160
– volume: 382
  start-page: 1708
  year: 2020
  ident: B23
  article-title: Clinical characteristics of coronavirus disease 2019 in China
  publication-title: New Engl J Med.
  doi: 10.1056/NEJMoa2002032
– volume-title: Learning Deep Architectures for AI
  year: 2009
  ident: B2
  doi: 10.1561/9781601982957
– year: 2014
  ident: B26
  article-title: Multiple object recognition with visual attention
  publication-title: arXiv preprint.
– volume: 16
  start-page: 1187
  year: 2020
  ident: B18
  article-title: Breast cancer detection and classification using traditional computer vision techniques: a comprehensive review
  publication-title: Curr Med Imag.
  doi: 10.2174/1573405616666200406110547
– start-page: 14126980
  year: 2014
  ident: B42
  article-title: Adam: a method for stochastic optimization
  publication-title: arXiv preprint.
– volume: 18
  start-page: 1
  year: 2021
  ident: B21
  article-title: Detection and classification of lung diseases for pneumonia and Covid-19 using machine and deep learning techniques
  publication-title: J Ambient Intell Humaniz Comput
  doi: 10.1007/s12652-021-03464-7
– volume: 63
  start-page: 1145
  year: 2015
  ident: B53
  article-title: MAPS: a quantitative radiomics approach for prostate cancer detection
  publication-title: IEEE Trans Biomed Eng.
  doi: 10.1109/TBME.2015.2485779
– start-page: 223
  volume-title: International Conference on Neural Information Processing
  year: 2019
  ident: B29
  article-title: Reinforcement learning with attention that works: a self-supervised approach
  doi: 10.1007/978-3-030-36802-9_25
– start-page: 200603677
  year: 2020
  ident: B33
  article-title: Visual transformers: token-based image representation and processing for computer vision
  publication-title: arXiv preprint.
– start-page: 190605909
  year: 2019
  ident: B35
  article-title: Stand-alone self-attention in vision models
  publication-title: arXiv preprint.
– start-page: 3286
  volume-title: Proceedings of the IEEE/CVF international conference on computer vision
  year: 2019
  ident: B34
  article-title: Attention augmented convolutional networks
– volume: 11
  start-page: 1
  year: 2021
  ident: B46
  article-title: COVID-Classifier: An automated machine learning model to assist in the diagnosis of COVID-19 infection in chest X-ray images
  publication-title: Sci Rep.
  doi: 10.1038/s41598-021-88807-2
– year: 2015
  ident: B4
  article-title: Deep residual learning for image recognition
  publication-title: CoRR.
– start-page: 14091556
  year: 2014
  ident: B5
  article-title: Very deep convolutional networks for large-scale image recognition
  publication-title: arXiv preprint.
– volume: 12
  start-page: 390
  year: 2020
  ident: B50
  article-title: Multiparametric MRI for prostate cancer detection: new insights into the combined use of a radiomic approach with advanced acquisition protocol
  publication-title: Cancers.
  doi: 10.3390/cancers12020390
– volume: 7
  start-page: 764047
  year: 2021
  ident: B11
  article-title: Fibrosis-Net: a tailored deep convolutional neural network design for prediction of pulmonary fibrosis progression from chest CT images
  publication-title: Front Artif Intell.
  doi: 10.3389/frai.2021.764047
– start-page: 3464
  volume-title: Proceedings of the IEEE/CVF International Conference on Computer Vision
  year: 2019
  ident: B37
  article-title: Local relation networks for image recognition
– start-page: 14090473
  year: 2014
  ident: B27
  article-title: Neural machine translation by jointly learning to align and translate
  publication-title: arXiv preprint.
– start-page: 249
  volume-title: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. JMLR Workshop and Conference Proceedings
  year: 2010
  ident: B3
  article-title: Understanding the difficulty of training deep feedforward neural networks
– volume: 598
  start-page: 348
  year: 2021
  ident: B15
  article-title: Biologically informed deep neural network for prostate cancer discovery
  publication-title: Nature.
  doi: 10.1038/s41586-021-03922-4
SSID ssj0001325413
Score 2.2742894
Snippet Medical image analysis continues to hold interesting challenges given the subtle characteristics of certain diseases and the significant overlap in appearance...
SourceID doaj
pubmedcentral
proquest
pubmed
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
Enrichment Source
StartPage 821120
SubjectTerms chest X-ray (CXR)
computer vision
COVID-19
Deep learning
deep neural net
diagnosis
Disease
Hypotheses
Infections
Medical imaging
Medicine
Neural networks
Severe acute respiratory syndrome coronavirus 2
Visual perception
SummonAdditionalLinks – databaseName: Health & Medical Collection
  dbid: 7X7
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3db9MwELdgIMTL-BwEBjISLzyY1R9xbF5QoZ1AYhVSGeqbFcc2TBpp12b8_dwladcitBeeIjlOdMnPPv_Od74j5LUZBOt1DraJsDlTZSEY6LzIpA4ySa-takuyfP9STCZmNrNf-w23VR9WudaJraIO8wr3yI-kgJUWVicu3i8uGFaNQu9qX0LjJrmFZbNxnBez4mqPRYL5w2XnnQRbzAJYEdODCv7WgOWDRb63VqM2af-_mObfAZNbK9Dxvf-V_T7Z77knHXaD5QG5EeuH5M5J711_ROqT8eh0OnxH21O5bArwRTqu8dT7ko1ie6XTeJ7YsGm6MEk6inFBMcMHvHjShZTT4ZZzggIppr03iH7-BdqLrvOgPCanx-NvHz-xvh4Dq5QdNKySZRTaCCs4NAA1MB7ZV1H4yqpSJ11GNQg899ZWJrc-5SomL60fpNIm4AUHZK-e1_EpoYIHY3QIoSy9yvPguQo8VUh2eKWVzsjRGhdX9cnKsWbGuQOjBZF0iKRDJF2HZEbebJ5YdIk6run7AaHe9MMU223DfPnD9TPWpWDgM7yRJiXQcsoLE0pgQ16CsKriGTlcg-36eb9yV0hn5NXmNsxYdMOUdZxfrpzQEq1wJaHPk25cbSQBOqxEoW1Gip0RtyPq7p367GebFdzADwV1_ex6sZ6TuwIPcGBJm_yQ7DXLy_iC3K5-N2er5ct2-vwB0pgjCw
  priority: 102
  providerName: ProQuest
Title MEDUSA: Multi-Scale Encoder-Decoder Self-Attention Deep Neural Network Architecture for Medical Image Analysis
URI https://www.ncbi.nlm.nih.gov/pubmed/35242769
https://www.proquest.com/docview/3272751212
https://www.proquest.com/docview/2636144432
https://pubmed.ncbi.nlm.nih.gov/PMC8886730
https://doaj.org/article/fd89c8b838ff4884b28da124b3b144c1
Volume 8
WOSCitedRecordID wos000783577600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: DOA
  dateStart: 20140101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: M~E
  dateStart: 20140101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: Health & Medical Collection
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: 7X7
  dateStart: 20190101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/healthcomplete
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: BENPR
  dateStart: 20190101
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Public Health Database
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: 8C1
  dateStart: 20190101
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/publichealth
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Publicly Available Content Database
  customDbUrl:
  eissn: 2296-858X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001325413
  issn: 2296-858X
  databaseCode: PIMPY
  dateStart: 20190101
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/publiccontent
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3fb9MwED7BQIgXxG8CozISLzyEJbbj2Lx1aycm0aqiDJUnK45tMWlkU5vx93N20ipFCF54SVXHjdy7s---nP0dwFuZWWVEgdiEqiLlVUlTXPNcyoRlnhmheCzJ8vVTOZ_L1UotBqW-wp6wjh64E9yRt1LV0kgmvUdj44ZKW6FTMswgFqgj8MlKNQBT8e0KQ-CTsy4viShMoZpcIAal-XuJmCeU9x74oUjX_6cY8_etkgPfc_oQHvRBIxl3g30Et1zzGO7N-rT4E2hm08n5cvyBxOO06RLl7si0CcfV1-nExU-ydJc-Hbdtt7-RTJy7JoGaAx887_aCk_Egq0AwmiV9Goec_cBlh2wJTJ7C-en0y8nHtC-kkNZcZW1as8pRIamiOTagT5cmhE1laWrFK-FF5Xhm88IoFHWhjC-484Ypk_lKeXToz-CguWrcCyA0t1IKa21VGV4UFpVgc1-HKCWvBRcJHG3FquueZTwUu7jUiDaCInRQhA6K0J0iEni3-8V1x7Dxl77HQVO7foEbOzagxejeYvS_LCaBw62edT9hN5pRDOQw-MlpAm92t3GqhfxJ1birm42mggX4zBn2ed6ZxW4kGMdyWgqVQLlnMHtD3b_TXHyPdN4SBYrr7Mv_8d9ewX0azmeEijXFIRy06xv3Gu7WP9uLzXoEt8tVGa8Sr_IkH8Gd4-l88XkUZw9-W5zNFt9-AV8MGz4
linkProvider Directory of Open Access Journals
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V1LbxMxELaqFAEX3o-FAkaCA4elWdu7ayMhFEiqRk2iSGlROZn12oZKZROSFMSf4jcys480Qai3HjhF2nUi7-bzNzOe8XyEvJBtq0wSQ2zCVByKLGUhcJ4LeWK55yZRopRk-ThIRyN5fKzGW-R3cxYGyyobTiyJ2k5z3CPf5QwsLViniL2bfQ9RNQqzq42ERgWLA_frJ4Rsi7f9Lvy_Lxnb6x1-2A9rVYEwF6q9DHOeOZZIplgEF8DASYM-RJqaXIks8UnmRNtGsVEql7EyPhbOG65M22fKS1SJAMrfFgj2Ftke94fjT-e7OhwCrohX-VCI_hTAw2FDUha9lhBroaz4mv0rZQL-5dv-XaK5ZvP2bv5vb-sWuVF717RTLYfbZMsVd8jVYV0_cJcUw173aNJ5Q8tzx-EEAOpor8Bz_fOw68pPOnGnPuwsl1UhKO06N6PYwwR-eFQVzdPOWvqFgttP63wX7X8DfqZNp5d75OhSnvY-aRXTwj0klEVWysRam2VGxLE1kbCRz9Gdi_JEJAHZbXCg87odO6qCnGoIyxA5GpGjETm6Qk5AXq2-MatakVww9j1CazUOm4iXF6bzL7rmJO2thMcwkkvvgceFYdJm4O8ZDpMVeRSQnQZcuma2hT5HVkCer24DJ2GiKSvc9GyhWcJxn0FwGPOgwvFqJuDwC5YmKiDpBsI3prp5pzj5WvY9l_BCwSA9unhaz8i1_cPhQA_6o4PH5DrD4yoo4BPvkNZyfuaekCv5j-XJYv60XryUfL7sFfAHpyOA9w
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=MEDUSA%3A+Multi-Scale+Encoder-Decoder+Self-Attention+Deep+Neural+Network+Architecture+for+Medical+Image+Analysis&rft.jtitle=Frontiers+in+medicine&rft.au=Aboutalebi%2C+Hossein&rft.au=Pavlova%2C+Maya&rft.au=Gunraj%2C+Hayden&rft.au=Shafiee%2C+Mohammad+Javad&rft.date=2022-02-15&rft.issn=2296-858X&rft.eissn=2296-858X&rft.volume=8&rft.spage=821120&rft_id=info:doi/10.3389%2Ffmed.2021.821120&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2296-858X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2296-858X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2296-858X&client=summon