Automatic Detection and Segmentation of Axillary Lymph Nodes

Barbu, Adrian; Suehling, Michael; Xu, Xun; Liu, David; Zhou, S. Kevin; Comanicìu, Dorin

doi:10.1007/978-3-642-15705-9_4

Cited by 30 publications

(23 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The third classifier uses gradient-aligned features (GAF) introduced in [1]. The idea is to extract the features at locations of gradient jumps, because these are hints for the object boundary.…”

Section: Candidate Generationmentioning

confidence: 99%

“…Next, 100 points are randomly sampled from the surface of the segmentation. As proposed in [1], the points are sorted by their gradient magnitude to enumerate them. The surface normal at each point is sampled at seven positions with a spacing of 1mm between the samples.…”

Section: Segmentation Based Features As Final Stage In the Detection mentioning

confidence: 99%

“…In [3], lymph nodes are detected and segmented by fitting a mass-spring model at different positions. In this paper, we follow [6,1] that proposed two data driven approaches. As in [6], a discriminative model of the lymph node appearance is combined with a spatial prior probability that contains anatomical knowledge about where lymph nodes are likely to appear.…”

Section: Introductionmentioning

confidence: 99%

“…As in [6], a discriminative model of the lymph node appearance is combined with a spatial prior probability that contains anatomical knowledge about where lymph nodes are likely to appear. Similar to [1], detection and segmentation is combined to improve the detection performance. We introduce a feature set that is extracted from a candidate segmentation.…”

Section: Introductionmentioning

confidence: 99%

“…In contrast to [1], we do not fit a sphere to a lymph node. Instead, we adapt graph cuts to the problem of lymph node segmentation.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Segmentation Based Features for Lymph Node Detection from 3-D Chest CT

Feulner

Zhou

Hammon

et al. 2011

Machine Learning in Medical Imaging

Self Cite

View full text Add to dashboard Cite

Abstract. Lymph nodes routinely need to be considered in clinical practice in all kinds of oncological examinations. Automatic detection of lymph nodes from chest CT data is however challenging because of low contrast and clutter. Sliding window detectors using traditional features easily get confused by similar structures like muscles and vessels. It recently has been proposed to combine segmentation and detection to improve the detection performance. Features extracted from a segmentation that is initialized with a detection candidate can be used to train a classifier that decides whether the detection is a true or false positive. In this paper, the graph cuts method is adapted to the problem of lymph nodes segmentation. We propose a setting that requires only a single positive seed and at the same time solves the small cut problem of graph cuts. Furthermore, we propose a feature set that is extracted from the candidate segmentation. A classifier is trained on this feature set and used to reject false alarms. Cross validation on 54 CT datasets showed that the proposed system reaches a detection rate of 60.9% with only 6.1 false alarms per volume image, which is better than the current state of the art of mediastinal lymph node detection.

show abstract

Section: Candidate Generationmentioning

confidence: 99%

Section: Segmentation Based Features As Final Stage In the Detection mentioning

confidence: 99%