Crossing-Preserving Multi-scale Vesselness

Hannink, Julius; Duits, Remco; Bekkers, EJ Erik

doi:10.1007/978-3-319-10470-6_75

Cited by 36 publications

(35 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various detectors of tubular image structures compute the contrast between the regions inside and outside the tube or ridge [2,3,9]. We extend this idea to curved-support Gaussian models by computing the secondorder directional derivative in the gradient direction at each pixel.…”

Section: Methodsmentioning

confidence: 99%

“…Most of the methods estimate a local tubularity measure (e.g. vesselness in [2,9]) based on hand-crafted features (henceforth, HCFs) modelling local geometrical properties of ideal tubular structures; enhancement filters are then built based on such models [2,3,5,9]. Recently, combining hand-crafted and learned filters, or using HCF in the learning process, has been shown to improve detection [7,8].…”

Section: Introductionmentioning

confidence: 98%

“…for lesion quantification and evaluation of therapy progress [1][2][3][4][5][6][7][8][9]. Most of the methods estimate a local tubularity measure (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Scale and Curvature Invariant Ridge Detector for Tortuous and Fragmented Structures

Annunziata

Kheirkhah

Hamrah

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Segmenting dendritic trees and corneal nerve fibres is challenging due to their uneven and irregular appearance in the respective image modalities. State-of-the-art approaches use hand-crafted features based on local assumptions that are often violated by tortuous and pointlike structures, e.g., straight tubular shape. We propose a novel ridge detector, SCIRD, which is simultaneously rotation, scale and curvature invariant, and relaxes shape assumptions to achieve enhancement of target image structures. Experimental results on three datasets show that our approach outperforms state-of-the-art hand-crafted methods on tortuous and point-like structures, especially when captured at low resolution or limited signal-to-noise ratio and in the presence of other non-target structures.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Scale and Curvature Invariant Ridge Detector for Tortuous and Fragmented Structures

Annunziata

Kheirkhah

Hamrah

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Fast and accurate curvilinear structure segmentation is therefore needed, but different characteristics of tortuous curvilinear structures across image modalities makes segmentation challenging. In fact, tortuousity violates one of the basic assumptions of most tubular structure detectors, namely locally straight tubular shape (Frangi et al, 1998;Soares et al, 2006;Law and Chung, 2008;Hannink et al, 2014). Further issues include the presence of non-target structures (clutter), low resolution, noise and non-uniform illumination.…”

Section: A C C E P T E D Mmentioning

confidence: 99%

“…However, performance tend to degrade at crossings or bifurcations since this approach only looks for elongated structures. To overcome this issue, Hannink et al (2014) proposed to segment crossings/bifurcations with multiscale invertible orientation scores and apply vesselness filters to maps of the latter. Optimally Oriented Flux (OOF) was recently proposed by Law and Chung (2008) to improve detection of adjacent structures with vesselness measures.…”

Section: Curvilinear Structure Segmentationmentioning

confidence: 99%