Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time

Cesare Magnetti; Veronika Zimmer; Nooshin Ghavami; Emily Skelton; Jacqueline Matthew; Karen Lloyd; Jo Hajnal; Julia A. Schnabel; Alberto Gomez

Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time

Cesare Magnetti, Veronika Zimmer, Nooshin Ghavami, Emily Skelton, Jacqueline Matthew, Karen Lloyd, Jo Hajnal, Julia A. Schnabel, Alberto Gomez

Research output: Working paper/Preprint › Preprint

21 Downloads (Pure)

Abstract

We present a computational method for real-time, patient-specific simulation of 2D ultrasound (US) images. The method uses a large number of tracked ultrasound images to learn a function that maps position and orientation of the transducer to ultrasound images. This is a first step towards realistic patient-specific simulations that will enable improved training and retrospective examination of complex cases. Our models can simulate a 2D image in under 4ms (well within real-time constraints), and produce simulated images that preserve the content (anatomical structures and artefacts) of real ultrasound images.

Original language	Undefined/Unknown
Publication status	Published - 11 May 2020

Keywords

eess.IV
94A08 (Primary) 68U10, 97R40 (Secondary)

Access to Document

2005.04931v1

1 Conference paper

Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time
Magnetti, C., Zimmer, V., Ghavami, N., Skelton, E., Matthew, J., Lloyd, K., Hajnal, J., Schnabel, J. A. & Gomez, A., 1 Jan 2020, Medical Image Understanding and Analysis - 24th Annual Conference, MIUA 2020, Proceedings. Papiez, B. W., Namburete, A. I. L., Yaqub, M., Noble, J. A. & Yaqub, M. (eds.). SPRINGER, p. 423-435 13 p. (Communications in Computer and Information Science; vol. 1248 CCIS).
Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review
1 Citation (Scopus)

Cite this

@techreport{2be640a55ae74e7cac2c76d9e3ea73ab,

title = "Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time",

abstract = " We present a computational method for real-time, patient-specific simulation of 2D ultrasound (US) images. The method uses a large number of tracked ultrasound images to learn a function that maps position and orientation of the transducer to ultrasound images. This is a first step towards realistic patient-specific simulations that will enable improved training and retrospective examination of complex cases. Our models can simulate a 2D image in under 4ms (well within real-time constraints), and produce simulated images that preserve the content (anatomical structures and artefacts) of real ultrasound images. ",

keywords = "eess.IV, 94A08 (Primary) 68U10, 97R40 (Secondary)",

author = "Cesare Magnetti and Veronika Zimmer and Nooshin Ghavami and Emily Skelton and Jacqueline Matthew and Karen Lloyd and Jo Hajnal and Schnabel, {Julia A.} and Alberto Gomez",

note = "Accepted for presentation at MIUA 2020",

year = "2020",

month = may,

day = "11",

language = "Undefined/Unknown",

type = "WorkingPaper",

}

TY - UNPB

T1 - Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time

AU - Magnetti, Cesare

AU - Zimmer, Veronika

AU - Ghavami, Nooshin

AU - Skelton, Emily

AU - Matthew, Jacqueline

AU - Lloyd, Karen

AU - Hajnal, Jo

AU - Schnabel, Julia A.

AU - Gomez, Alberto

N1 - Accepted for presentation at MIUA 2020

PY - 2020/5/11

Y1 - 2020/5/11

N2 - We present a computational method for real-time, patient-specific simulation of 2D ultrasound (US) images. The method uses a large number of tracked ultrasound images to learn a function that maps position and orientation of the transducer to ultrasound images. This is a first step towards realistic patient-specific simulations that will enable improved training and retrospective examination of complex cases. Our models can simulate a 2D image in under 4ms (well within real-time constraints), and produce simulated images that preserve the content (anatomical structures and artefacts) of real ultrasound images.

AB - We present a computational method for real-time, patient-specific simulation of 2D ultrasound (US) images. The method uses a large number of tracked ultrasound images to learn a function that maps position and orientation of the transducer to ultrasound images. This is a first step towards realistic patient-specific simulations that will enable improved training and retrospective examination of complex cases. Our models can simulate a 2D image in under 4ms (well within real-time constraints), and produce simulated images that preserve the content (anatomical structures and artefacts) of real ultrasound images.

KW - eess.IV

KW - 94A08 (Primary) 68U10, 97R40 (Secondary)

M3 - Preprint

BT - Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images in Real Time

ER -