An Automated Segmentation Algorithm for CT Volumes of Livers with Atypical Shapes and Large Pathological Lesions

Shun UMETSU; Akinobu SHIMIZU; Hidefumi WATANABE; Hidefumi KOBATAKE; Shigeru NAWANO

doi:10.1587/transinf.E97.D.951

An Automated Segmentation Algorithm for CT Volumes of Livers with Atypical Shapes and Large Pathological Lesions

Shun UMETSU, Akinobu SHIMIZU, Hidefumi WATANABE, Hidefumi KOBATAKE, Shigeru NAWANO

Full Text Views

0

Share
Cite this

Summary :

This paper presents a novel liver segmentation algorithm that achieves higher performance than conventional algorithms in the segmentation of cases with unusual liver shapes and/or large liver lesions. An L1 norm was introduced to the mean squared difference to find the most relevant cases with an input case from a training dataset. A patient-specific probabilistic atlas was generated from the retrieved cases to compensate for livers with unusual shapes, which accounts for liver shape more specifically than a conventional probabilistic atlas that is averaged over a number of training cases. To make the above process robust against large pathological lesions, we incorporated a novel term based on a set of “lesion bases” proposed in this study that account for the differences from normal liver parenchyma. Subsequently, the patient-specific probabilistic atlas was forwarded to a graph-cuts-based fine segmentation step, in which a penalty function was computed from the probabilistic atlas. A leave-one-out test using clinical abdominal CT volumes was conducted to validate the performance, and proved that the proposed segmentation algorithm with the proposed patient-specific atlas reinforced by the lesion bases outperformed the conventional algorithm with a statistically significant difference.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.4 pp.951-963

Publication Date: 2014/04/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.951

Type of Manuscript: PAPER

Category: Biological Engineering

Authors

Shun UMETSU
  Tokyo University of Agriculture and Technology
Akinobu SHIMIZU
  Tokyo University of Agriculture and Technology
Hidefumi WATANABE
  Tokyo University of Agriculture and Technology
Hidefumi KOBATAKE
  Institute of National Colleges of Technology
Shigeru NAWANO
  International University of Health and Welfare

Keyword

CT volume, liver segmentation, lasso, lesion, probabilistic atlas

Cite this

Copy

Shun UMETSU, Akinobu SHIMIZU, Hidefumi WATANABE, Hidefumi KOBATAKE, Shigeru NAWANO, "An Automated Segmentation Algorithm for CT Volumes of Livers with Atypical Shapes and Large Pathological Lesions" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 4, pp. 951-963, April 2014, doi: 10.1587/transinf.E97.D.951.
Abstract: This paper presents a novel liver segmentation algorithm that achieves higher performance than conventional algorithms in the segmentation of cases with unusual liver shapes and/or large liver lesions. An L1 norm was introduced to the mean squared difference to find the most relevant cases with an input case from a training dataset. A patient-specific probabilistic atlas was generated from the retrieved cases to compensate for livers with unusual shapes, which accounts for liver shape more specifically than a conventional probabilistic atlas that is averaged over a number of training cases. To make the above process robust against large pathological lesions, we incorporated a novel term based on a set of “lesion bases” proposed in this study that account for the differences from normal liver parenchyma. Subsequently, the patient-specific probabilistic atlas was forwarded to a graph-cuts-based fine segmentation step, in which a penalty function was computed from the probabilistic atlas. A leave-one-out test using clinical abdominal CT volumes was conducted to validate the performance, and proved that the proposed segmentation algorithm with the proposed patient-specific atlas reinforced by the lesion bases outperformed the conventional algorithm with a statistically significant difference.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E97.D.951/_p

Copy

@ARTICLE{e97-d_4_951,
author={Shun UMETSU, Akinobu SHIMIZU, Hidefumi WATANABE, Hidefumi KOBATAKE, Shigeru NAWANO, },
journal={IEICE TRANSACTIONS on Information},
title={An Automated Segmentation Algorithm for CT Volumes of Livers with Atypical Shapes and Large Pathological Lesions},
year={2014},
volume={E97-D},
number={4},
pages={951-963},
abstract={This paper presents a novel liver segmentation algorithm that achieves higher performance than conventional algorithms in the segmentation of cases with unusual liver shapes and/or large liver lesions. An L1 norm was introduced to the mean squared difference to find the most relevant cases with an input case from a training dataset. A patient-specific probabilistic atlas was generated from the retrieved cases to compensate for livers with unusual shapes, which accounts for liver shape more specifically than a conventional probabilistic atlas that is averaged over a number of training cases. To make the above process robust against large pathological lesions, we incorporated a novel term based on a set of “lesion bases” proposed in this study that account for the differences from normal liver parenchyma. Subsequently, the patient-specific probabilistic atlas was forwarded to a graph-cuts-based fine segmentation step, in which a penalty function was computed from the probabilistic atlas. A leave-one-out test using clinical abdominal CT volumes was conducted to validate the performance, and proved that the proposed segmentation algorithm with the proposed patient-specific atlas reinforced by the lesion bases outperformed the conventional algorithm with a statistically significant difference.},
keywords={},
doi={10.1587/transinf.E97.D.951},
ISSN={1745-1361},
month={April},}

Copy

TY - JOUR
TI - An Automated Segmentation Algorithm for CT Volumes of Livers with Atypical Shapes and Large Pathological Lesions
T2 - IEICE TRANSACTIONS on Information
SP - 951
EP - 963
AU - Shun UMETSU
AU - Akinobu SHIMIZU
AU - Hidefumi WATANABE
AU - Hidefumi KOBATAKE
AU - Shigeru NAWANO
PY - 2014
DO - 10.1587/transinf.E97.D.951
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2014
AB - This paper presents a novel liver segmentation algorithm that achieves higher performance than conventional algorithms in the segmentation of cases with unusual liver shapes and/or large liver lesions. An L1 norm was introduced to the mean squared difference to find the most relevant cases with an input case from a training dataset. A patient-specific probabilistic atlas was generated from the retrieved cases to compensate for livers with unusual shapes, which accounts for liver shape more specifically than a conventional probabilistic atlas that is averaged over a number of training cases. To make the above process robust against large pathological lesions, we incorporated a novel term based on a set of “lesion bases” proposed in this study that account for the differences from normal liver parenchyma. Subsequently, the patient-specific probabilistic atlas was forwarded to a graph-cuts-based fine segmentation step, in which a penalty function was computed from the probabilistic atlas. A leave-one-out test using clinical abdominal CT volumes was conducted to validate the performance, and proved that the proposed segmentation algorithm with the proposed patient-specific atlas reinforced by the lesion bases outperformed the conventional algorithm with a statistically significant difference.
ER -