Automatic slice identification in 3D medical images with a ConvNet regressor

Bob D. de Vos; Max A. Viergever; Pim A. de Jong; Ivana Išgum

doi:10.1007/978-3-319-46976-8_17

Automatic slice identification in 3D medical images with a ConvNet regressor

Bob D. de Vos^*
, Max A. Viergever
, Pim A. de Jong
, Ivana Išgum

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Identification of anatomical regions of interest is a prerequisite in many medical image analysis tasks. We propose a method that automatically identifies a slice of interest (SOI) in 3D images with a convolutional neural network (ConvNet) regressor. In 150 chest CT scans two reference slices were manually identified: one containing the aortic root and another superior to the aortic arch. In two independent experiments, the ConvNet regressor was trained with 100 CTs to determine the distance between each slice and the SOI in a CT. To identify the SOI, a first order polynomial was fitted through the obtained distances. In 50 test scans, the mean distances between the reference and the automatically identified slices were 5.7mm (4.0 slices) for the aortic root and 5.6mm (3.7 slices) for the aortic arch. The method shows similar results for both tasks and could be used for automatic slice identification.

Original language	English
Title of host publication	Deep Learning and Data Labeling for Medical Applications
Subtitle of host publication	First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings
Editors	Gustavo Carneiro, Diana Mateus, Loïc Peter, Andrew Bradley
Publisher	Springer-Verlag
Pages	161-169
Number of pages	9
ISBN (Electronic)	978-3-319-46976-8
ISBN (Print)	9783319469751
DOIs	https://doi.org/10.1007/978-3-319-46976-8_17
Publication status	Published - 2016
Event	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 - Athens, Greece Duration: 21 Oct 2016 → 21 Oct 2016

Publication series

Name	Lecture Notes in Computer Science
Volume	10008
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Conference

Conference	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016
Country/Territory	Greece
City	Athens
Period	21/10/16 → 21/10/16

Keywords

Convolutional neural network
Deep learning
Detection
Localization
Regression
Slice identification

Access to Document

10.1007/978-3-319-46976-8_17

Cite this

de Vos, B. D., Viergever, M. A., de Jong, P. A., & Išgum, I. (2016). Automatic slice identification in 3D medical images with a ConvNet regressor. In G. Carneiro, D. Mateus, L. Peter, & A. Bradley (Eds.), Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings (pp. 161-169). (Lecture Notes in Computer Science ; Vol. 10008 ). Springer-Verlag. https://doi.org/10.1007/978-3-319-46976-8_17

de Vos, Bob D. ; Viergever, Max A. ; de Jong, Pim A. et al. / Automatic slice identification in 3D medical images with a ConvNet regressor. Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings. editor / Gustavo Carneiro ; Diana Mateus ; Loïc Peter ; Andrew Bradley. Springer-Verlag, 2016. pp. 161-169 (Lecture Notes in Computer Science ).

@inproceedings{e7125732ee2c43f390d8a3c740601911,

title = "Automatic slice identification in 3D medical images with a ConvNet regressor",

abstract = "Identification of anatomical regions of interest is a prerequisite in many medical image analysis tasks. We propose a method that automatically identifies a slice of interest (SOI) in 3D images with a convolutional neural network (ConvNet) regressor. In 150 chest CT scans two reference slices were manually identified: one containing the aortic root and another superior to the aortic arch. In two independent experiments, the ConvNet regressor was trained with 100 CTs to determine the distance between each slice and the SOI in a CT. To identify the SOI, a first order polynomial was fitted through the obtained distances. In 50 test scans, the mean distances between the reference and the automatically identified slices were 5.7mm (4.0 slices) for the aortic root and 5.6mm (3.7 slices) for the aortic arch. The method shows similar results for both tasks and could be used for automatic slice identification.",

keywords = "Convolutional neural network, Deep learning, Detection, Localization, Regression, Slice identification",

author = "\{de Vos\}, \{Bob D.\} and Viergever, \{Max A.\} and \{de Jong\}, \{Pim A.\} and Ivana I{\v s}gum",

year = "2016",

doi = "10.1007/978-3-319-46976-8\_17",

language = "English",

isbn = "9783319469751",

series = "Lecture Notes in Computer Science ",

publisher = "Springer-Verlag",

pages = "161--169",

editor = "Carneiro, \{Gustavo \} and Mateus, \{Diana \} and Peter, \{Lo{\"i}c \} and \{ Bradley\}, Andrew",

booktitle = "Deep Learning and Data Labeling for Medical Applications",

address = "Germany",

note = "1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016 ; Conference date: 21-10-2016 Through 21-10-2016",

}

de Vos, BD, Viergever, MA , de Jong, PA & Išgum, I 2016, Automatic slice identification in 3D medical images with a ConvNet regressor. in G Carneiro, D Mateus, L Peter & A Bradley (eds), Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings. Lecture Notes in Computer Science , vol. 10008 , Springer-Verlag, pp. 161-169, 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016, Athens, Greece, 21/10/16. https://doi.org/10.1007/978-3-319-46976-8_17

Automatic slice identification in 3D medical images with a ConvNet regressor. / de Vos, Bob D.; Viergever, Max A.; de Jong, Pim A. et al.
Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings. ed. / Gustavo Carneiro; Diana Mateus; Loïc Peter; Andrew Bradley. Springer-Verlag, 2016. p. 161-169 (Lecture Notes in Computer Science ; Vol. 10008 ).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Automatic slice identification in 3D medical images with a ConvNet regressor

AU - de Vos, Bob D.

AU - Viergever, Max A.

AU - de Jong, Pim A.

AU - Išgum, Ivana

PY - 2016

Y1 - 2016

N2 - Identification of anatomical regions of interest is a prerequisite in many medical image analysis tasks. We propose a method that automatically identifies a slice of interest (SOI) in 3D images with a convolutional neural network (ConvNet) regressor. In 150 chest CT scans two reference slices were manually identified: one containing the aortic root and another superior to the aortic arch. In two independent experiments, the ConvNet regressor was trained with 100 CTs to determine the distance between each slice and the SOI in a CT. To identify the SOI, a first order polynomial was fitted through the obtained distances. In 50 test scans, the mean distances between the reference and the automatically identified slices were 5.7mm (4.0 slices) for the aortic root and 5.6mm (3.7 slices) for the aortic arch. The method shows similar results for both tasks and could be used for automatic slice identification.

AB - Identification of anatomical regions of interest is a prerequisite in many medical image analysis tasks. We propose a method that automatically identifies a slice of interest (SOI) in 3D images with a convolutional neural network (ConvNet) regressor. In 150 chest CT scans two reference slices were manually identified: one containing the aortic root and another superior to the aortic arch. In two independent experiments, the ConvNet regressor was trained with 100 CTs to determine the distance between each slice and the SOI in a CT. To identify the SOI, a first order polynomial was fitted through the obtained distances. In 50 test scans, the mean distances between the reference and the automatically identified slices were 5.7mm (4.0 slices) for the aortic root and 5.6mm (3.7 slices) for the aortic arch. The method shows similar results for both tasks and could be used for automatic slice identification.

KW - Convolutional neural network

KW - Deep learning

KW - Detection

KW - Localization

KW - Regression

KW - Slice identification

UR - https://www.scopus.com/pages/publications/84992530168

U2 - 10.1007/978-3-319-46976-8_17

DO - 10.1007/978-3-319-46976-8_17

M3 - Conference contribution

AN - SCOPUS:84992530168

SN - 9783319469751

T3 - Lecture Notes in Computer Science

SP - 161

EP - 169

BT - Deep Learning and Data Labeling for Medical Applications

A2 - Carneiro, Gustavo

A2 - Mateus, Diana

A2 - Peter, Loïc

A2 - Bradley, Andrew

PB - Springer-Verlag

T2 - 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2016

Y2 - 21 October 2016 through 21 October 2016

ER -

de Vos BD, Viergever MA , de Jong PA, Išgum I. Automatic slice identification in 3D medical images with a ConvNet regressor. In Carneiro G, Mateus D, Peter L, Bradley A, editors, Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings. Springer-Verlag. 2016. p. 161-169. (Lecture Notes in Computer Science ). doi: 10.1007/978-3-319-46976-8_17

Automatic slice identification in 3D medical images with a ConvNet regressor

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this