Universität zu Köln

Laukamp, Kai Roman, Thiele, Frank, Shakirin, Georgy, Zopfs, David ORCID: 0000-0001-9978-7453, Faymonville, Andrea, Timmer, Marco, Maintz, David, Perkuhn, Michael and Borggrefe, Jan ORCID: 0000-0003-2908-7560 (2019). Fully automated detection and segmentation of meningiomas using deep learning on routine multiparametric MRI. Eur. Radiol., 29 (1). S. 124 - 133. NEW YORK: SPRINGER. ISSN 1432-1084

Full text not available from this repository.

Abstract

ObjectivesMagnetic resonance imaging (MRI) is the method of choice for imaging meningiomas. Volumetric assessment of meningiomas is highly relevant for therapy planning and monitoring. We used a multiparametric deep-learning model (DLM) on routine MRI data including images from diverse referring institutions to investigate DLM performance in automated detection and segmentation of meningiomas in comparison to manual segmentations.MethodsWe included 56 of 136 consecutive preoperative MRI datasets [T1/T2-weighted, T1-weighted contrast-enhanced (T1CE), FLAIR] of meningiomas that were treated surgically at the University Hospital Cologne and graded histologically as tumour grade I (n = 38) or grade II (n = 18). The DLM was trained on an independent dataset of 249 glioma cases and segmented different tumour classes as defined in the brain tumour image segmentation benchmark (BRATS benchmark). The DLM was based on the DeepMedic architecture. Results were compared to manual segmentations by two radiologists in a consensus reading in FLAIR and T1CE.ResultsThe DLM detected meningiomas in 55 of 56 cases. Further, automated segmentations correlated strongly with manual segmentations: average Dice coefficients were 0.81 0.10 (range, 0.46-0.93) for the total tumour volume (union of tumour volume in FLAIR and T1CE) and 0.78 0.19 (range, 0.27-0.95) for contrast-enhancing tumour volume in T1CE.ConclusionsThe DLM yielded accurate automated detection and segmentation of meningioma tissue despite diverse scanner data and thereby may improve and facilitate therapy planning as well as monitoring of this highly frequent tumour entity.Key Points center dot Deep learning allows for accurate meningioma detection and segmentation center dot Deep learning helps clinicians to assess patients with meningiomas center dot Meningioma monitoring and treatment planning can be improved

Item Type:	Journal Article
Creators:	Creators Email ORCID ORCID Put Code Laukamp, Kai Roman UNSPECIFIED UNSPECIFIED UNSPECIFIED Thiele, Frank UNSPECIFIED UNSPECIFIED UNSPECIFIED Shakirin, Georgy UNSPECIFIED UNSPECIFIED UNSPECIFIED Zopfs, David UNSPECIFIED orcid.org/0000-0001-9978-7453 UNSPECIFIED Faymonville, Andrea UNSPECIFIED UNSPECIFIED UNSPECIFIED Timmer, Marco UNSPECIFIED UNSPECIFIED UNSPECIFIED Maintz, David UNSPECIFIED UNSPECIFIED UNSPECIFIED Perkuhn, Michael UNSPECIFIED UNSPECIFIED UNSPECIFIED Borggrefe, Jan UNSPECIFIED orcid.org/0000-0003-2908-7560 UNSPECIFIED
URN:	urn:nbn:de:hbz:38-140729
DOI:	10.1007/s00330-018-5595-8
Journal or Publication Title:	Eur. Radiol.
Volume:	29
Number:	1
Page Range:	S. 124 - 133
Date:	2019
Publisher:	SPRINGER
Place of Publication:	NEW YORK
ISSN:	1432-1084
Language:	English
Faculty:	Unspecified
Divisions:	Unspecified
Subjects:	no entry
Uncontrolled Keywords:	Keywords Language BRAIN-TUMOR SEGMENTATION; COMPUTER-AIDED DETECTION; CLASSIFICATION; DIAGNOSIS; GRADE; SYSTEM; CNN Multiple languages Radiology, Nuclear Medicine & Medical Imaging Multiple languages
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/14072