Filling Large Discontinuities in 3D Vascular Networks Using Skeleton- and Intensity-Based Information
Segmentation of vasculature is a common task in many areas of medical imaging, but complex morphology and weak signal often lead to incomplete segmentations. In this paper, we present a new gap filling strategy for 3D vascular networks. The novelty of our
- PDF / 1,103,905 Bytes
- 8 Pages / 439.363 x 666.131 pts Page_size
- 79 Downloads / 125 Views
2
1 Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, UK Institut de Math´ematiques de Toulouse (UMR 5219), CNRS, France 3 Department of Oncology, University of Oxford, UK
Abstract. Segmentation of vasculature is a common task in many areas of medical imaging, but complex morphology and weak signal often lead to incomplete segmentations. In this paper, we present a new gap filling strategy for 3D vascular networks. The novelty of our approach is to combine both skeleton- and intensity-based information to fill large discontinuities. Our approach also does not make any hypothesis on the network topology, which is particularly important for tumour vasculature due to the chaotic arrangement of vessels within tumours. Synthetic results show that using intensity-based information, in addition to skeleton-based information, can make the detection of large discontinuities more robust. Our strategy is also shown to outperform a classic gap filling strategy on 3D Micro-CT images of preclinical tumour models.
1
Introduction
Gap filling methods for vascular networks have recently generated significant interest. Many methods for the segmentation of the vasculature rely on the generation of a likelihood or vesselness map. To obtain a final segmentation, these maps are usually binarized, meaning that important vessel information may be discarded. Under-segmentation in this sense can lead to discontinuities in the segmentation, which will have implications for any analysis of the branching structure. In this paper, we then propose a novel method to incorporate image intensity information, additional to the final segmentation, to reconnect the gaps in the segmentation. This method is motivated by the extraction of tumour vasculature which is highly leaky and poorly perfused, leading to an irregular distribution of signal within the vasculature (see Fig. 1). No strong hypotheses can therefore be made on their chaotic and highly irregular topology.
We would like to acknowledge funding from the CRUK/ EPSRC Cancer Imaging Centre in Oxford. JAS and LR also wish to acknowledge the CNRS-INSMI/John Fell Oxford University Press (OUP) Research Fund.
c Springer International Publishing Switzerland 2015 N. Navab et al. (Eds.): MICCAI 2015, Part III, LNCS 9351, pp. 157–164, 2015. DOI: 10.1007/978-3-319-24574-4_19
158
R. Bates et al.
Preliminary approaches to perform gap filling or to improve robustness in the segmentation of vascular structures were proposed in [10,13,9]. A gap filling strategy for large 3D images with a discontinuous segmented vasculature, based on a tensor voting strategy [4], was proposed in [11]. This approach, however, only makes use of the skeleton of segmented structures. Another tensor voting strategy was proposed in [7] to segment noisy tubular structures in an iterative fashion. However, this only applies to relatively small gaps. An interesting learning strategy was proposed in [5]. This approach utilizes human interactions to learn appropriate graph connectivity.
Data Loading...
