Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data

Jen Wei Kuo, Yao Wang, Orlando Aristizabal, Jeffrey A. Ketterling, Jonathan Mamou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Volumetric analysis of brain ventricles is important to the study of normal and abnormal development of the central nervous system of mouse embryos. High-frequency ultrasound (HFU) is frequently used to image embryos because HFU is real-time, non-invasive, and provides fine-resolution images. However, manual segmentation of ventricles from 3D HFU volumes remains challenging and time consuming. In this study, in utero and in vivo volumetric ultrasound data were acquired from pregnant mice using a 5-element, 40-MHz annular array. An automatic segmentation algorithm based on active shape model (ASM) was developed to segment the brain ventricles of the embryos; ASM allows us to efficiently 'learn' from training data (i.e., manually segmented data). The algorithm was further enhanced by using detail-preserving reference shapes (also learned from training data) and region growing constrained by the reference shape. The hybrid algorithm was applied to three 12.5-day-old embryos. Results were qualitatively analyzed and compared with manual segmentation results in regions typically difficult to segment (e.g., thin brain ventricle connections). In addition, quantitative analysis using the Dice similarity coefficient (DSC) was used to compare the automatic segmentation results with manual segmentation. We obtained average DSC values of 0.848±0.015 for the brain ventricles and our method produced morphologically accurate results. Therefore, our method could streamline current HFU longitudinal studies of brain development that require manual segmentation.

Original languageEnglish (US)
Title of host publication2013 IEEE International Ultrasonics Symposium, IUS 2013
Pages1781-1784
Number of pages4
DOIs
StatePublished - 2013
Event2013 IEEE International Ultrasonics Symposium, IUS 2013 - Prague, Czech Republic
Duration: Jul 21 2013Jul 25 2013

Other

Other2013 IEEE International Ultrasonics Symposium, IUS 2013
CountryCzech Republic
CityPrague
Period7/21/137/25/13

Fingerprint

embryos
brain
mice
education
volumetric analysis
central nervous system
image resolution
coefficients
preserving
quantitative analysis

Keywords

  • Brain
  • High-frequency ultrasound
  • Mouse embryo
  • Segmentation

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Cite this

Kuo, J. W., Wang, Y., Aristizabal, O., Ketterling, J. A., & Mamou, J. (2013). Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data. In 2013 IEEE International Ultrasonics Symposium, IUS 2013 (pp. 1781-1784). [6725197] https://doi.org/10.1109/ULTSYM.2013.0454

Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data. / Kuo, Jen Wei; Wang, Yao; Aristizabal, Orlando; Ketterling, Jeffrey A.; Mamou, Jonathan.

2013 IEEE International Ultrasonics Symposium, IUS 2013. 2013. p. 1781-1784 6725197.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kuo, JW, Wang, Y, Aristizabal, O, Ketterling, JA & Mamou, J 2013, Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data. in 2013 IEEE International Ultrasonics Symposium, IUS 2013., 6725197, pp. 1781-1784, 2013 IEEE International Ultrasonics Symposium, IUS 2013, Prague, Czech Republic, 7/21/13. https://doi.org/10.1109/ULTSYM.2013.0454
Kuo JW, Wang Y, Aristizabal O, Ketterling JA, Mamou J. Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data. In 2013 IEEE International Ultrasonics Symposium, IUS 2013. 2013. p. 1781-1784. 6725197 https://doi.org/10.1109/ULTSYM.2013.0454
Kuo, Jen Wei ; Wang, Yao ; Aristizabal, Orlando ; Ketterling, Jeffrey A. ; Mamou, Jonathan. / Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data. 2013 IEEE International Ultrasonics Symposium, IUS 2013. 2013. pp. 1781-1784
@inproceedings{ef0a2d0b3f8a494fb0591592b5f18a38,
title = "Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data",
abstract = "Volumetric analysis of brain ventricles is important to the study of normal and abnormal development of the central nervous system of mouse embryos. High-frequency ultrasound (HFU) is frequently used to image embryos because HFU is real-time, non-invasive, and provides fine-resolution images. However, manual segmentation of ventricles from 3D HFU volumes remains challenging and time consuming. In this study, in utero and in vivo volumetric ultrasound data were acquired from pregnant mice using a 5-element, 40-MHz annular array. An automatic segmentation algorithm based on active shape model (ASM) was developed to segment the brain ventricles of the embryos; ASM allows us to efficiently 'learn' from training data (i.e., manually segmented data). The algorithm was further enhanced by using detail-preserving reference shapes (also learned from training data) and region growing constrained by the reference shape. The hybrid algorithm was applied to three 12.5-day-old embryos. Results were qualitatively analyzed and compared with manual segmentation results in regions typically difficult to segment (e.g., thin brain ventricle connections). In addition, quantitative analysis using the Dice similarity coefficient (DSC) was used to compare the automatic segmentation results with manual segmentation. We obtained average DSC values of 0.848±0.015 for the brain ventricles and our method produced morphologically accurate results. Therefore, our method could streamline current HFU longitudinal studies of brain development that require manual segmentation.",
keywords = "Brain, High-frequency ultrasound, Mouse embryo, Segmentation",
author = "Kuo, {Jen Wei} and Yao Wang and Orlando Aristizabal and Ketterling, {Jeffrey A.} and Jonathan Mamou",
year = "2013",
doi = "10.1109/ULTSYM.2013.0454",
language = "English (US)",
isbn = "9781467356862",
pages = "1781--1784",
booktitle = "2013 IEEE International Ultrasonics Symposium, IUS 2013",

}

TY - GEN

T1 - Automatic mouse embryo brain ventricle segmentation from 3D 40-MHz ultrasound data

AU - Kuo, Jen Wei

AU - Wang, Yao

AU - Aristizabal, Orlando

AU - Ketterling, Jeffrey A.

AU - Mamou, Jonathan

PY - 2013

Y1 - 2013

N2 - Volumetric analysis of brain ventricles is important to the study of normal and abnormal development of the central nervous system of mouse embryos. High-frequency ultrasound (HFU) is frequently used to image embryos because HFU is real-time, non-invasive, and provides fine-resolution images. However, manual segmentation of ventricles from 3D HFU volumes remains challenging and time consuming. In this study, in utero and in vivo volumetric ultrasound data were acquired from pregnant mice using a 5-element, 40-MHz annular array. An automatic segmentation algorithm based on active shape model (ASM) was developed to segment the brain ventricles of the embryos; ASM allows us to efficiently 'learn' from training data (i.e., manually segmented data). The algorithm was further enhanced by using detail-preserving reference shapes (also learned from training data) and region growing constrained by the reference shape. The hybrid algorithm was applied to three 12.5-day-old embryos. Results were qualitatively analyzed and compared with manual segmentation results in regions typically difficult to segment (e.g., thin brain ventricle connections). In addition, quantitative analysis using the Dice similarity coefficient (DSC) was used to compare the automatic segmentation results with manual segmentation. We obtained average DSC values of 0.848±0.015 for the brain ventricles and our method produced morphologically accurate results. Therefore, our method could streamline current HFU longitudinal studies of brain development that require manual segmentation.

AB - Volumetric analysis of brain ventricles is important to the study of normal and abnormal development of the central nervous system of mouse embryos. High-frequency ultrasound (HFU) is frequently used to image embryos because HFU is real-time, non-invasive, and provides fine-resolution images. However, manual segmentation of ventricles from 3D HFU volumes remains challenging and time consuming. In this study, in utero and in vivo volumetric ultrasound data were acquired from pregnant mice using a 5-element, 40-MHz annular array. An automatic segmentation algorithm based on active shape model (ASM) was developed to segment the brain ventricles of the embryos; ASM allows us to efficiently 'learn' from training data (i.e., manually segmented data). The algorithm was further enhanced by using detail-preserving reference shapes (also learned from training data) and region growing constrained by the reference shape. The hybrid algorithm was applied to three 12.5-day-old embryos. Results were qualitatively analyzed and compared with manual segmentation results in regions typically difficult to segment (e.g., thin brain ventricle connections). In addition, quantitative analysis using the Dice similarity coefficient (DSC) was used to compare the automatic segmentation results with manual segmentation. We obtained average DSC values of 0.848±0.015 for the brain ventricles and our method produced morphologically accurate results. Therefore, our method could streamline current HFU longitudinal studies of brain development that require manual segmentation.

KW - Brain

KW - High-frequency ultrasound

KW - Mouse embryo

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=84894386305&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894386305&partnerID=8YFLogxK

U2 - 10.1109/ULTSYM.2013.0454

DO - 10.1109/ULTSYM.2013.0454

M3 - Conference contribution

SN - 9781467356862

SP - 1781

EP - 1784

BT - 2013 IEEE International Ultrasonics Symposium, IUS 2013

ER -