A Survey of Deep-Learning Applications in Ultrasound: Artificial Intelligence–Powered Ultrasound for Improving Clinical Workflow

Zeynettin Akkus, Jason Cai, Arunnit Boonrod, Atefeh Zeinoddini, Alexander D. Weston, Kenneth A. Philbrick, Bradley J Erickson

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Ultrasound is the most commonly used imaging modality in clinical practice because it is a nonionizing, low-cost, and portable point-of-care imaging tool that provides real-time images. Artificial intelligence (AI)–powered ultrasound is becoming more mature and getting closer to routine clinical applications in recent times because of an increased need for efficient and objective acquisition and evaluation of ultrasound images. Because ultrasound images involve operator-, patient-, and scanner-dependent variations, the adaptation of classical machine learning methods to clinical applications becomes challenging. With their self-learning ability, deep-learning (DL) methods are able to harness exponentially growing graphics processing unit computing power to identify abstract and complex imaging features. This has given rise to tremendous opportunities such as providing robust and generalizable AI models for improving image acquisition, real-time assessment of image quality, objective diagnosis and detection of diseases, and optimizing ultrasound clinical workflow. In this report, the authors review current DL approaches and research directions in rapidly advancing ultrasound technology and present their outlook on future directions and trends for DL techniques to further improve diagnosis, reduce health care cost, and optimize ultrasound clinical workflow.

Original languageEnglish (US)
Pages (from-to)1318-1328
Number of pages11
JournalJournal of the American College of Radiology
Volume16
Issue number9
DOIs
StatePublished - Sep 1 2019

Fingerprint

Workflow
Learning
Artificial Intelligence
Point-of-Care Systems
Aptitude
Health Care Costs
Technology
Costs and Cost Analysis
Surveys and Questionnaires
Research
Direction compound

Keywords

  • Artificial intelligence in ultrasound
  • breast lesion
  • deep learning in ultrasound
  • liver lesion
  • thyroid nodule

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

A Survey of Deep-Learning Applications in Ultrasound : Artificial Intelligence–Powered Ultrasound for Improving Clinical Workflow. / Akkus, Zeynettin; Cai, Jason; Boonrod, Arunnit; Zeinoddini, Atefeh; Weston, Alexander D.; Philbrick, Kenneth A.; Erickson, Bradley J.

In: Journal of the American College of Radiology, Vol. 16, No. 9, 01.09.2019, p. 1318-1328.

Research output: Contribution to journalArticle

Akkus, Zeynettin ; Cai, Jason ; Boonrod, Arunnit ; Zeinoddini, Atefeh ; Weston, Alexander D. ; Philbrick, Kenneth A. ; Erickson, Bradley J. / A Survey of Deep-Learning Applications in Ultrasound : Artificial Intelligence–Powered Ultrasound for Improving Clinical Workflow. In: Journal of the American College of Radiology. 2019 ; Vol. 16, No. 9. pp. 1318-1328.
@article{cd46a7ccfb01423ca79c4746b3ae31d7,
title = "A Survey of Deep-Learning Applications in Ultrasound: Artificial Intelligence–Powered Ultrasound for Improving Clinical Workflow",
abstract = "Ultrasound is the most commonly used imaging modality in clinical practice because it is a nonionizing, low-cost, and portable point-of-care imaging tool that provides real-time images. Artificial intelligence (AI)–powered ultrasound is becoming more mature and getting closer to routine clinical applications in recent times because of an increased need for efficient and objective acquisition and evaluation of ultrasound images. Because ultrasound images involve operator-, patient-, and scanner-dependent variations, the adaptation of classical machine learning methods to clinical applications becomes challenging. With their self-learning ability, deep-learning (DL) methods are able to harness exponentially growing graphics processing unit computing power to identify abstract and complex imaging features. This has given rise to tremendous opportunities such as providing robust and generalizable AI models for improving image acquisition, real-time assessment of image quality, objective diagnosis and detection of diseases, and optimizing ultrasound clinical workflow. In this report, the authors review current DL approaches and research directions in rapidly advancing ultrasound technology and present their outlook on future directions and trends for DL techniques to further improve diagnosis, reduce health care cost, and optimize ultrasound clinical workflow.",
keywords = "Artificial intelligence in ultrasound, breast lesion, deep learning in ultrasound, liver lesion, thyroid nodule",
author = "Zeynettin Akkus and Jason Cai and Arunnit Boonrod and Atefeh Zeinoddini and Weston, {Alexander D.} and Philbrick, {Kenneth A.} and Erickson, {Bradley J}",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.jacr.2019.06.004",
language = "English (US)",
volume = "16",
pages = "1318--1328",
journal = "JACR Journal of the American College of Radiology",
issn = "1558-349X",
publisher = "Elsevier BV",
number = "9",

}

TY - JOUR

T1 - A Survey of Deep-Learning Applications in Ultrasound

T2 - Artificial Intelligence–Powered Ultrasound for Improving Clinical Workflow

AU - Akkus, Zeynettin

AU - Cai, Jason

AU - Boonrod, Arunnit

AU - Zeinoddini, Atefeh

AU - Weston, Alexander D.

AU - Philbrick, Kenneth A.

AU - Erickson, Bradley J

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Ultrasound is the most commonly used imaging modality in clinical practice because it is a nonionizing, low-cost, and portable point-of-care imaging tool that provides real-time images. Artificial intelligence (AI)–powered ultrasound is becoming more mature and getting closer to routine clinical applications in recent times because of an increased need for efficient and objective acquisition and evaluation of ultrasound images. Because ultrasound images involve operator-, patient-, and scanner-dependent variations, the adaptation of classical machine learning methods to clinical applications becomes challenging. With their self-learning ability, deep-learning (DL) methods are able to harness exponentially growing graphics processing unit computing power to identify abstract and complex imaging features. This has given rise to tremendous opportunities such as providing robust and generalizable AI models for improving image acquisition, real-time assessment of image quality, objective diagnosis and detection of diseases, and optimizing ultrasound clinical workflow. In this report, the authors review current DL approaches and research directions in rapidly advancing ultrasound technology and present their outlook on future directions and trends for DL techniques to further improve diagnosis, reduce health care cost, and optimize ultrasound clinical workflow.

AB - Ultrasound is the most commonly used imaging modality in clinical practice because it is a nonionizing, low-cost, and portable point-of-care imaging tool that provides real-time images. Artificial intelligence (AI)–powered ultrasound is becoming more mature and getting closer to routine clinical applications in recent times because of an increased need for efficient and objective acquisition and evaluation of ultrasound images. Because ultrasound images involve operator-, patient-, and scanner-dependent variations, the adaptation of classical machine learning methods to clinical applications becomes challenging. With their self-learning ability, deep-learning (DL) methods are able to harness exponentially growing graphics processing unit computing power to identify abstract and complex imaging features. This has given rise to tremendous opportunities such as providing robust and generalizable AI models for improving image acquisition, real-time assessment of image quality, objective diagnosis and detection of diseases, and optimizing ultrasound clinical workflow. In this report, the authors review current DL approaches and research directions in rapidly advancing ultrasound technology and present their outlook on future directions and trends for DL techniques to further improve diagnosis, reduce health care cost, and optimize ultrasound clinical workflow.

KW - Artificial intelligence in ultrasound

KW - breast lesion

KW - deep learning in ultrasound

KW - liver lesion

KW - thyroid nodule

UR - http://www.scopus.com/inward/record.url?scp=85071017712&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071017712&partnerID=8YFLogxK

U2 - 10.1016/j.jacr.2019.06.004

DO - 10.1016/j.jacr.2019.06.004

M3 - Article

C2 - 31492410

AN - SCOPUS:85071017712

VL - 16

SP - 1318

EP - 1328

JO - JACR Journal of the American College of Radiology

JF - JACR Journal of the American College of Radiology

SN - 1558-349X

IS - 9

ER -