Liver segmentation and metastases detection in MR images using convolutional neural networks

Mariëlle J.A. Jansen (Corresponding author), Hugo J. Kuijf, Maarten Niekel, Wouter B. Veldhuis, Frank J. Wessels, Max A. Viergever, Josien P.W. Pluim

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

Uittreksel

Primary tumors have a high likelihood of developing metastases in the liver, and early detection of these metastases is crucial for patient outcome. We propose a method based on convolutional neural networks to detect liver metastases. First, the liver is automatically segmented using the six phases of abdominal dynamic contrast-enhanced (DCE) MR images. Next, DCE-MR and diffusion weighted MR images are used for metastases detection within the liver mask. The liver segmentations have a median Dice similarity coefficient of 0.95 compared with manual annotations. The metastases detection method has a sensitivity of 99.8% with a median of two false positives per image. The combination of the two MR sequences in a dual pathway network is proven valuable for the detection of liver metastases. In conclusion, a high quality liver segmentation can be obtained in which we can successfully detect liver metastases.

TaalEngels
Artikelnummer044003
Aantal pagina's10
TijdschriftJournal of Medical Imaging
Volume6
Nummer van het tijdschrift4
DOI's
StatusGepubliceerd - okt 2019

Vingerafdruk

Neoplasm Metastasis
Liver
Masks
Neoplasms

Citeer dit

Jansen, M. J. A., Kuijf, H. J., Niekel, M., Veldhuis, W. B., Wessels, F. J., Viergever, M. A., & Pluim, J. P. W. (2019). Liver segmentation and metastases detection in MR images using convolutional neural networks. Journal of Medical Imaging, 6(4), [044003]. DOI: 10.1117/1.JMI.6.4.044003
Jansen, Mariëlle J.A. ; Kuijf, Hugo J. ; Niekel, Maarten ; Veldhuis, Wouter B. ; Wessels, Frank J. ; Viergever, Max A. ; Pluim, Josien P.W./ Liver segmentation and metastases detection in MR images using convolutional neural networks. In: Journal of Medical Imaging. 2019 ; Vol. 6, Nr. 4.
@article{0d01071771db44d5a43823849481ea1b,
title = "Liver segmentation and metastases detection in MR images using convolutional neural networks",
abstract = "Primary tumors have a high likelihood of developing metastases in the liver, and early detection of these metastases is crucial for patient outcome. We propose a method based on convolutional neural networks to detect liver metastases. First, the liver is automatically segmented using the six phases of abdominal dynamic contrast-enhanced (DCE) MR images. Next, DCE-MR and diffusion weighted MR images are used for metastases detection within the liver mask. The liver segmentations have a median Dice similarity coefficient of 0.95 compared with manual annotations. The metastases detection method has a sensitivity of 99.8{\%} with a median of two false positives per image. The combination of the two MR sequences in a dual pathway network is proven valuable for the detection of liver metastases. In conclusion, a high quality liver segmentation can be obtained in which we can successfully detect liver metastases.",
author = "Jansen, {Mari{\"e}lle J.A.} and Kuijf, {Hugo J.} and Maarten Niekel and Veldhuis, {Wouter B.} and Wessels, {Frank J.} and Viergever, {Max A.} and Pluim, {Josien P.W.}",
note = "{\circledC} 2019 Society of Photo-Optical Instrumentation Engineers (SPIE).",
year = "2019",
month = "10",
doi = "10.1117/1.JMI.6.4.044003",
language = "English",
volume = "6",
journal = "Journal of Medical Imaging",
issn = "2329-4302",
publisher = "SPIE",
number = "4",

}

Jansen, MJA, Kuijf, HJ, Niekel, M, Veldhuis, WB, Wessels, FJ, Viergever, MA & Pluim, JPW 2019, 'Liver segmentation and metastases detection in MR images using convolutional neural networks' Journal of Medical Imaging, vol. 6, nr. 4, 044003. DOI: 10.1117/1.JMI.6.4.044003

Liver segmentation and metastases detection in MR images using convolutional neural networks. / Jansen, Mariëlle J.A. (Corresponding author); Kuijf, Hugo J.; Niekel, Maarten; Veldhuis, Wouter B.; Wessels, Frank J.; Viergever, Max A.; Pluim, Josien P.W.

In: Journal of Medical Imaging, Vol. 6, Nr. 4, 044003, 10.2019.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Liver segmentation and metastases detection in MR images using convolutional neural networks

AU - Jansen,Mariëlle J.A.

AU - Kuijf,Hugo J.

AU - Niekel,Maarten

AU - Veldhuis,Wouter B.

AU - Wessels,Frank J.

AU - Viergever,Max A.

AU - Pluim,Josien P.W.

N1 - © 2019 Society of Photo-Optical Instrumentation Engineers (SPIE).

PY - 2019/10

Y1 - 2019/10

N2 - Primary tumors have a high likelihood of developing metastases in the liver, and early detection of these metastases is crucial for patient outcome. We propose a method based on convolutional neural networks to detect liver metastases. First, the liver is automatically segmented using the six phases of abdominal dynamic contrast-enhanced (DCE) MR images. Next, DCE-MR and diffusion weighted MR images are used for metastases detection within the liver mask. The liver segmentations have a median Dice similarity coefficient of 0.95 compared with manual annotations. The metastases detection method has a sensitivity of 99.8% with a median of two false positives per image. The combination of the two MR sequences in a dual pathway network is proven valuable for the detection of liver metastases. In conclusion, a high quality liver segmentation can be obtained in which we can successfully detect liver metastases.

AB - Primary tumors have a high likelihood of developing metastases in the liver, and early detection of these metastases is crucial for patient outcome. We propose a method based on convolutional neural networks to detect liver metastases. First, the liver is automatically segmented using the six phases of abdominal dynamic contrast-enhanced (DCE) MR images. Next, DCE-MR and diffusion weighted MR images are used for metastases detection within the liver mask. The liver segmentations have a median Dice similarity coefficient of 0.95 compared with manual annotations. The metastases detection method has a sensitivity of 99.8% with a median of two false positives per image. The combination of the two MR sequences in a dual pathway network is proven valuable for the detection of liver metastases. In conclusion, a high quality liver segmentation can be obtained in which we can successfully detect liver metastases.

U2 - 10.1117/1.JMI.6.4.044003

DO - 10.1117/1.JMI.6.4.044003

M3 - Article

VL - 6

JO - Journal of Medical Imaging

T2 - Journal of Medical Imaging

JF - Journal of Medical Imaging

SN - 2329-4302

IS - 4

M1 - 044003

ER -

Jansen MJA, Kuijf HJ, Niekel M, Veldhuis WB, Wessels FJ, Viergever MA et al. Liver segmentation and metastases detection in MR images using convolutional neural networks. Journal of Medical Imaging. 2019 okt;6(4). 044003. Beschikbaar vanaf, DOI: 10.1117/1.JMI.6.4.044003