Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study

M. J. A. Jansen; H.J. Kuijf; J. P. W. Pluim

doi:10.1109/ISBI.2018.8363565

Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study

M. J. A. Jansen, H.J. Kuijf, J. P. W. Pluim

Medical Image Analysis

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

2 Citations (Scopus)

Abstract

Focal liver lesion classification is an important part of diagnostics. In clinical practice, T2-weighted (T2W) and dynamic contrast enhanced (DCE) MR images are used to determine the type of lesion. For automatic liver lesion classification only T2W images are exploited. In this feasibility study, a multi-modal approach for automatic lesion classification of five lesion classes (adenoma, cyst, haemangioma, HCC, and metastasis) is studied. Features are derived from four sets: (A) non-corrected, and (B) motion corrected DCE-MRI, (C) T2W images, and (D) B+C combined, originating from 43 patients. An extremely randomized forest is used as classifier. The results show that motion corrected DCE-MRI features are a valuable addition to the T2W features, and improve the accuracy in discriminating benign and malignant lesions, as well as the classification of the five lesion classes. The multimodal approach shows promising results for an automatic liver lesion classification.

Original language	English
Title of host publication	2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Pages	245-248
Number of pages	4
ISBN (Electronic)	978-1-5386-3636-7
ISBN (Print)	978-1-5386-3637-4
DOIs	https://doi.org/10.1109/ISBI.2018.8363565
Publication status	Published - 23 May 2018
Event	15th IEEE International Symposium on Biomedical Imaging (ISBI 2018) - Omni Shoreham Hotel, Washington, United States Duration: 4 Apr 2018 → 7 Apr 2018 Conference number: 15 https://biomedicalimaging.org/2018/

Conference

Conference	15th IEEE International Symposium on Biomedical Imaging (ISBI 2018)
Abbreviated title	ISBI18
Country/Territory	United States
City	Washington
Period	4/04/18 → 7/04/18
Internet address	https://biomedicalimaging.org/2018/

Keywords

Liver
classification
DCE-MRI
Classification

Access to Document

10.1109/ISBI.2018.8363565

Cite this

@inproceedings{fb1cfbfe9dec416abe45cd6a46e80cc7,

title = "Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study",

abstract = "Focal liver lesion classification is an important part of diagnostics. In clinical practice, T2-weighted (T2W) and dynamic contrast enhanced (DCE) MR images are used to determine the type of lesion. For automatic liver lesion classification only T2W images are exploited. In this feasibility study, a multi-modal approach for automatic lesion classification of five lesion classes (adenoma, cyst, haemangioma, HCC, and metastasis) is studied. Features are derived from four sets: (A) non-corrected, and (B) motion corrected DCE-MRI, (C) T2W images, and (D) B+C combined, originating from 43 patients. An extremely randomized forest is used as classifier. The results show that motion corrected DCE-MRI features are a valuable addition to the T2W features, and improve the accuracy in discriminating benign and malignant lesions, as well as the classification of the five lesion classes. The multimodal approach shows promising results for an automatic liver lesion classification.",

keywords = "Liver, classification, DCE-MRI, Classification",

author = "Jansen, {M. J. A.} and H.J. Kuijf and Pluim, {J. P. W.}",

year = "2018",

month = may,

day = "23",

doi = "10.1109/ISBI.2018.8363565",

language = "English",

isbn = "978-1-5386-3637-4",

pages = "245--248",

booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "15th IEEE International Symposium on Biomedical Imaging (ISBI 2018), ISBI18 ; Conference date: 04-04-2018 Through 07-04-2018",

url = "https://biomedicalimaging.org/2018/",

}

Jansen, MJA, Kuijf, HJ & Pluim, JPW 2018, Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study. in 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Institute of Electrical and Electronics Engineers, Piscataway, pp. 245-248, 15th IEEE International Symposium on Biomedical Imaging (ISBI 2018), Washington, District of Columbia, United States, 4/04/18. https://doi.org/10.1109/ISBI.2018.8363565

Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study. / Jansen, M. J. A.; Kuijf, H.J.; Pluim, J. P. W.
2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Piscataway: Institute of Electrical and Electronics Engineers, 2018. p. 245-248.

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

TY - GEN

T1 - Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images

T2 - 15th IEEE International Symposium on Biomedical Imaging (ISBI 2018)

AU - Jansen, M. J. A.

AU - Kuijf, H.J.

AU - Pluim, J. P. W.

N1 - Conference code: 15

PY - 2018/5/23

Y1 - 2018/5/23

N2 - Focal liver lesion classification is an important part of diagnostics. In clinical practice, T2-weighted (T2W) and dynamic contrast enhanced (DCE) MR images are used to determine the type of lesion. For automatic liver lesion classification only T2W images are exploited. In this feasibility study, a multi-modal approach for automatic lesion classification of five lesion classes (adenoma, cyst, haemangioma, HCC, and metastasis) is studied. Features are derived from four sets: (A) non-corrected, and (B) motion corrected DCE-MRI, (C) T2W images, and (D) B+C combined, originating from 43 patients. An extremely randomized forest is used as classifier. The results show that motion corrected DCE-MRI features are a valuable addition to the T2W features, and improve the accuracy in discriminating benign and malignant lesions, as well as the classification of the five lesion classes. The multimodal approach shows promising results for an automatic liver lesion classification.

AB - Focal liver lesion classification is an important part of diagnostics. In clinical practice, T2-weighted (T2W) and dynamic contrast enhanced (DCE) MR images are used to determine the type of lesion. For automatic liver lesion classification only T2W images are exploited. In this feasibility study, a multi-modal approach for automatic lesion classification of five lesion classes (adenoma, cyst, haemangioma, HCC, and metastasis) is studied. Features are derived from four sets: (A) non-corrected, and (B) motion corrected DCE-MRI, (C) T2W images, and (D) B+C combined, originating from 43 patients. An extremely randomized forest is used as classifier. The results show that motion corrected DCE-MRI features are a valuable addition to the T2W features, and improve the accuracy in discriminating benign and malignant lesions, as well as the classification of the five lesion classes. The multimodal approach shows promising results for an automatic liver lesion classification.

KW - Liver

KW - classification

KW - DCE-MRI

KW - Classification

UR - http://www.scopus.com/inward/record.url?scp=85048105298&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2018.8363565

DO - 10.1109/ISBI.2018.8363565

M3 - Conference contribution

SN - 978-1-5386-3637-4

SP - 245

EP - 248

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

Y2 - 4 April 2018 through 7 April 2018

ER -

Automatic classification of focal liver lesions based on clinical DCE-MR and T2-weighted images: a feasibility study

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this