Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries

Thomas L.A. van den Heuvel, Hezkiel Petros, Stefano Santini, Chris L. de Korte, Bram van Ginneken

Research output: Contribution to journalArticleAcademicpeer-review

1 Citation (Scopus)

Abstract

Ultrasound imaging remains out of reach for most pregnant women in developing countries because it requires a trained sonographer to acquire and interpret the images. We address this problem by presenting a system that can automatically estimate the fetal head circumference (HC) from data obtained with use of the obstetric sweep protocol (OSP). The OSP consists of multiple pre-defined sweeps with the ultrasound transducer over the abdomen of the pregnant woman. The OSP can be taught within a day to any health care worker without prior knowledge of ultrasound. An experienced sonographer acquired both the standard plane—to obtain the reference HC—and the OSP from 183 pregnant women in St. Luke's Hospital, Wolisso, Ethiopia. The OSP data, which will most likely not contain the standard plane, was used to automatically estimate HC using two fully convolutional neural networks. First, a VGG-Net-inspired network was trained to automatically detect the frames that contained the fetal head. Second, a U-net-inspired network was trained to automatically measure the HC for all frames in which the first network detected a fetal head. The HC was estimated from these frame measurements, and the curve of Hadlock was used to determine gestational age (GA). The results indicated that most automatically estimated GAs fell within the P2.5–P97.5 interval of the Hadlock curve compared with the GAs obtained from the reference HC, so it is possible to automatically estimate GA from OSP data. Our method therefore has potential application for providing maternal care in resource-constrained countries.

Original languageEnglish
Pages (from-to)773-785
Number of pages13
JournalUltrasound in medicine and biology
Volume45
Issue number3
Early online date17 Dec 2018
DOIs
Publication statusPublished - 1 Mar 2019

Fingerprint

circumferences
learning
resources
Obstetrics
Hand
Head
Learning
Pregnant Women
estimates
Ethiopia
Gestational Age
abdomen
curves
health
transducers
Transducers
Abdomen
intervals
Developing Countries
Ultrasonography

Keywords

  • Computer-aided detection and diagnosis
  • Fetus
  • Machine learning
  • Neural network
  • Obstetric Sweep Protocol
  • Prenatal
  • Resource-limited countries
  • Segmentation
  • Ultrasound

Cite this

van den Heuvel, Thomas L.A. ; Petros, Hezkiel ; Santini, Stefano ; de Korte, Chris L. ; van Ginneken, Bram. / Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries. In: Ultrasound in medicine and biology. 2019 ; Vol. 45, No. 3. pp. 773-785.
@article{5c38e193929d49d082e3e6a7a6af8496,
title = "Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries",
abstract = "Ultrasound imaging remains out of reach for most pregnant women in developing countries because it requires a trained sonographer to acquire and interpret the images. We address this problem by presenting a system that can automatically estimate the fetal head circumference (HC) from data obtained with use of the obstetric sweep protocol (OSP). The OSP consists of multiple pre-defined sweeps with the ultrasound transducer over the abdomen of the pregnant woman. The OSP can be taught within a day to any health care worker without prior knowledge of ultrasound. An experienced sonographer acquired both the standard plane—to obtain the reference HC—and the OSP from 183 pregnant women in St. Luke's Hospital, Wolisso, Ethiopia. The OSP data, which will most likely not contain the standard plane, was used to automatically estimate HC using two fully convolutional neural networks. First, a VGG-Net-inspired network was trained to automatically detect the frames that contained the fetal head. Second, a U-net-inspired network was trained to automatically measure the HC for all frames in which the first network detected a fetal head. The HC was estimated from these frame measurements, and the curve of Hadlock was used to determine gestational age (GA). The results indicated that most automatically estimated GAs fell within the P2.5–P97.5 interval of the Hadlock curve compared with the GAs obtained from the reference HC, so it is possible to automatically estimate GA from OSP data. Our method therefore has potential application for providing maternal care in resource-constrained countries.",
keywords = "Computer-aided detection and diagnosis, Fetus, Machine learning, Neural network, Obstetric Sweep Protocol, Prenatal, Resource-limited countries, Segmentation, Ultrasound",
author = "{van den Heuvel}, {Thomas L.A.} and Hezkiel Petros and Stefano Santini and {de Korte}, {Chris L.} and {van Ginneken}, Bram",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.ultrasmedbio.2018.09.015",
language = "English",
volume = "45",
pages = "773--785",
journal = "Ultrasound in medicine and biology",
issn = "0301-5629",
publisher = "Elsevier",
number = "3",

}

Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries. / van den Heuvel, Thomas L.A.; Petros, Hezkiel; Santini, Stefano; de Korte, Chris L.; van Ginneken, Bram.

In: Ultrasound in medicine and biology, Vol. 45, No. 3, 01.03.2019, p. 773-785.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Automated Fetal Head Detection and Circumference Estimation from Free-Hand Ultrasound Sweeps Using Deep Learning in Resource-Limited Countries

AU - van den Heuvel, Thomas L.A.

AU - Petros, Hezkiel

AU - Santini, Stefano

AU - de Korte, Chris L.

AU - van Ginneken, Bram

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Ultrasound imaging remains out of reach for most pregnant women in developing countries because it requires a trained sonographer to acquire and interpret the images. We address this problem by presenting a system that can automatically estimate the fetal head circumference (HC) from data obtained with use of the obstetric sweep protocol (OSP). The OSP consists of multiple pre-defined sweeps with the ultrasound transducer over the abdomen of the pregnant woman. The OSP can be taught within a day to any health care worker without prior knowledge of ultrasound. An experienced sonographer acquired both the standard plane—to obtain the reference HC—and the OSP from 183 pregnant women in St. Luke's Hospital, Wolisso, Ethiopia. The OSP data, which will most likely not contain the standard plane, was used to automatically estimate HC using two fully convolutional neural networks. First, a VGG-Net-inspired network was trained to automatically detect the frames that contained the fetal head. Second, a U-net-inspired network was trained to automatically measure the HC for all frames in which the first network detected a fetal head. The HC was estimated from these frame measurements, and the curve of Hadlock was used to determine gestational age (GA). The results indicated that most automatically estimated GAs fell within the P2.5–P97.5 interval of the Hadlock curve compared with the GAs obtained from the reference HC, so it is possible to automatically estimate GA from OSP data. Our method therefore has potential application for providing maternal care in resource-constrained countries.

AB - Ultrasound imaging remains out of reach for most pregnant women in developing countries because it requires a trained sonographer to acquire and interpret the images. We address this problem by presenting a system that can automatically estimate the fetal head circumference (HC) from data obtained with use of the obstetric sweep protocol (OSP). The OSP consists of multiple pre-defined sweeps with the ultrasound transducer over the abdomen of the pregnant woman. The OSP can be taught within a day to any health care worker without prior knowledge of ultrasound. An experienced sonographer acquired both the standard plane—to obtain the reference HC—and the OSP from 183 pregnant women in St. Luke's Hospital, Wolisso, Ethiopia. The OSP data, which will most likely not contain the standard plane, was used to automatically estimate HC using two fully convolutional neural networks. First, a VGG-Net-inspired network was trained to automatically detect the frames that contained the fetal head. Second, a U-net-inspired network was trained to automatically measure the HC for all frames in which the first network detected a fetal head. The HC was estimated from these frame measurements, and the curve of Hadlock was used to determine gestational age (GA). The results indicated that most automatically estimated GAs fell within the P2.5–P97.5 interval of the Hadlock curve compared with the GAs obtained from the reference HC, so it is possible to automatically estimate GA from OSP data. Our method therefore has potential application for providing maternal care in resource-constrained countries.

KW - Computer-aided detection and diagnosis

KW - Fetus

KW - Machine learning

KW - Neural network

KW - Obstetric Sweep Protocol

KW - Prenatal

KW - Resource-limited countries

KW - Segmentation

KW - Ultrasound

UR - http://www.scopus.com/inward/record.url?scp=85058454105&partnerID=8YFLogxK

U2 - 10.1016/j.ultrasmedbio.2018.09.015

DO - 10.1016/j.ultrasmedbio.2018.09.015

M3 - Article

VL - 45

SP - 773

EP - 785

JO - Ultrasound in medicine and biology

JF - Ultrasound in medicine and biology

SN - 0301-5629

IS - 3

ER -