Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Breast cancer screening with digital breast tomosynthesis : comparison of different reading strategies implementing artificial intelligence

Dahlblom, Victor LU orcid ; Dustler, Magnus LU ; Tingberg, Anders LU and Zackrisson, Sophia LU (2023) In European Radiology 33(5). p.3754-3765
Abstract

OBJECTIVES: Digital breast tomosynthesis (DBT) can detect more cancers than the current standard breast screening method, digital mammography (DM); however, it can substantially increase the reading workload and thus hinder implementation in screening. Artificial intelligence (AI) might be a solution. The aim of this study was to retrospectively test different ways of using AI in a screening workflow.

METHODS: An AI system was used to analyse 14,772 double-read single-view DBT examinations from a screening trial with paired DM double reading. Three scenarios were studied: if AI can identify normal cases that can be excluded from human reading; if AI can replace the second reader; if AI can replace both readers. The number of... (More)

OBJECTIVES: Digital breast tomosynthesis (DBT) can detect more cancers than the current standard breast screening method, digital mammography (DM); however, it can substantially increase the reading workload and thus hinder implementation in screening. Artificial intelligence (AI) might be a solution. The aim of this study was to retrospectively test different ways of using AI in a screening workflow.

METHODS: An AI system was used to analyse 14,772 double-read single-view DBT examinations from a screening trial with paired DM double reading. Three scenarios were studied: if AI can identify normal cases that can be excluded from human reading; if AI can replace the second reader; if AI can replace both readers. The number of detected cancers and false positives was compared with DM or DBT double reading.

RESULTS: By excluding normal cases and only reading 50.5% (7460/14,772) of all examinations, 95% (121/127) of the DBT double reading detected cancers could be detected. Compared to DM screening, 27% (26/95) more cancers could be detected (p < 0.001) while keeping recall rates at the same level. With AI replacing the second reader, 95% (120/127) of the DBT double reading detected cancers could be detected-26% (25/95) more than DM screening (p < 0.001)-while increasing recall rates by 53%. AI alone with DBT has a sensitivity similar to DM double reading (p = 0.689).

CONCLUSION: AI can open up possibilities for implementing DBT screening and detecting more cancers with the total reading workload unchanged. Considering the potential legal and psychological implications, replacing the second reader with AI would probably be most the feasible approach.

KEY POINTS: • Breast cancer screening with digital breast tomosynthesis and artificial intelligence can detect more cancers than mammography screening without increasing screen-reading workload. • Artificial intelligence can either exclude low-risk cases from double reading or replace the second reader. • Retrospective study based on paired mammography and digital breast tomosynthesis screening data.

(Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Humans, Female, Breast Neoplasms/diagnosis, Retrospective Studies, Artificial Intelligence, Early Detection of Cancer/methods, Breast/diagnostic imaging, Mammography/methods, Mass Screening/methods
in
European Radiology
volume
33
issue
5
pages
3754 - 3765
publisher
Springer
external identifiers
  • scopus:85143699851
  • pmid:36502459
ISSN
0938-7994
DOI
10.1007/s00330-022-09316-y
language
English
LU publication?
yes
additional info
© 2022. The Author(s).
id
2a4f6af7-29e5-4227-85ee-d2b4bfa62bdb
date added to LUP
2023-08-31 09:45:42
date last changed
2024-04-20 03:19:20
@article{2a4f6af7-29e5-4227-85ee-d2b4bfa62bdb,
  abstract     = {{<p>OBJECTIVES: Digital breast tomosynthesis (DBT) can detect more cancers than the current standard breast screening method, digital mammography (DM); however, it can substantially increase the reading workload and thus hinder implementation in screening. Artificial intelligence (AI) might be a solution. The aim of this study was to retrospectively test different ways of using AI in a screening workflow.</p><p>METHODS: An AI system was used to analyse 14,772 double-read single-view DBT examinations from a screening trial with paired DM double reading. Three scenarios were studied: if AI can identify normal cases that can be excluded from human reading; if AI can replace the second reader; if AI can replace both readers. The number of detected cancers and false positives was compared with DM or DBT double reading.</p><p>RESULTS: By excluding normal cases and only reading 50.5% (7460/14,772) of all examinations, 95% (121/127) of the DBT double reading detected cancers could be detected. Compared to DM screening, 27% (26/95) more cancers could be detected (p &lt; 0.001) while keeping recall rates at the same level. With AI replacing the second reader, 95% (120/127) of the DBT double reading detected cancers could be detected-26% (25/95) more than DM screening (p &lt; 0.001)-while increasing recall rates by 53%. AI alone with DBT has a sensitivity similar to DM double reading (p = 0.689).</p><p>CONCLUSION: AI can open up possibilities for implementing DBT screening and detecting more cancers with the total reading workload unchanged. Considering the potential legal and psychological implications, replacing the second reader with AI would probably be most the feasible approach.</p><p>KEY POINTS: • Breast cancer screening with digital breast tomosynthesis and artificial intelligence can detect more cancers than mammography screening without increasing screen-reading workload. • Artificial intelligence can either exclude low-risk cases from double reading or replace the second reader. • Retrospective study based on paired mammography and digital breast tomosynthesis screening data.</p>}},
  author       = {{Dahlblom, Victor and Dustler, Magnus and Tingberg, Anders and Zackrisson, Sophia}},
  issn         = {{0938-7994}},
  keywords     = {{Humans; Female; Breast Neoplasms/diagnosis; Retrospective Studies; Artificial Intelligence; Early Detection of Cancer/methods; Breast/diagnostic imaging; Mammography/methods; Mass Screening/methods}},
  language     = {{eng}},
  number       = {{5}},
  pages        = {{3754--3765}},
  publisher    = {{Springer}},
  series       = {{European Radiology}},
  title        = {{Breast cancer screening with digital breast tomosynthesis : comparison of different reading strategies implementing artificial intelligence}},
  url          = {{http://dx.doi.org/10.1007/s00330-022-09316-y}},
  doi          = {{10.1007/s00330-022-09316-y}},
  volume       = {{33}},
  year         = {{2023}},
}