Lymph node detection and segmentation in chest CT data using discriminative learning and a spatial prior

Feulner, Johannes; Zhou, S. Kevin; Hammon, Matthias; Hornegger, Joachim; Comanicìu, Dorin

doi:10.1016/j.media.2012.11.001

Cited by 63 publications

(50 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In comparison, the direct one-shot 3D detection (Barbu et al, 2012, Feulner et al, 2013) saturates at ~65% sensitivity at full FP range. Recently, the availability of large-scale annotated training sets and the accessibility of affordable parallel computing resources via GPUs has made it feasible to train deep Convolution Neural Networks (CNNs) and achieve great advances in challenging ImageNet recognition tasks (Krizhevsky et al, 2012, Zeiler and Fergus, 2013).…”

Section: Introductionmentioning

confidence: 98%

“…Particularly, lymph nodes have large within-class appearance, location or pose variations, and low contrast from surrounding anatomies over a patient population. This results in many false-positives (FP), to assure a moderately high detection sensitivity (Feuerstein et al, 2009), or only limited sensitivity levels (Barbu et al, 2012, Feulner et al, 2013). The good sensitivities achieved at low FP range in Barbu et al (2012) are not directly comparable with the other studies since Barbu et al (2012) reports on axillary, pelvic, and only some parts of the abdominal regions, while others evaluate only on mediastinum (Feuerstein et al, 2012, Feulner et al, 2013, Feuerstein et al, 2009) or abdomen (Nakamura et al, 2013).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A New 2.5D Representation for Lymph Node Detection Using Random Sets of Deep Convolutional Neural Network Observations

Roth

Lü

Seff

et al. 2014

Lecture Notes in Computer Science

393

298

View full text Add to dashboard Cite

Automated Lymph Node (LN) detection is an important clinical diagnostic task but very challenging due to the low contrast of surrounding structures in Computed Tomography (CT) and to their varying sizes, poses, shapes and sparsely distributed locations. State-of-the-art studies show the performance range of 52.9% sensitivity at 3.1 false-positives per volume (FP/vol.), or 60.9% at 6.1 FP/vol. for mediastinal LN, by one-shot boosting on 3D HAAR features. In this paper, we first operate a preliminary candidate generation stage, towards ~100% sensitivity at the cost of high FP levels (~40 per patient), to harvest volumes of interest (VOI). Our 2.5D approach consequently decomposes any 3D VOI by resampling 2D reformatted orthogonal views N times, via scale, random translations, and rotations with respect to the VOI centroid coordinates. These random views are then used to train a deep Convolutional Neural Network (CNN) classifier. In testing, the CNN is employed to assign LN probabilities for all N random views that can be simply averaged (as a set) to compute the final classification probability per VOI. We validate the approach on two datasets: 90 CT volumes with 388 mediastinal LNs and 86 patients with 595 abdominal LNs. We achieve sensitivities of 70%/83% at 3 FP/vol. and 84%/90% at 6 FP/vol. in mediastinum and abdomen respectively, which drastically improves over the previous state-of-theart work.

show abstract

Section: Introductionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

A New 2.5D Representation for Lymph Node Detection Using Random Sets of Deep Convolutional Neural Network Observations

Roth

Lü

Seff

et al. 2014

Lecture Notes in Computer Science

393

298

View full text Add to dashboard Cite

show abstract

“…In the second approach, known as the continuous convex relaxation method, the image is treated in a continuous domain, and the optimal labeling minimization problem, initially nonconvex, is relaxed to obtain an equivalent convex minimization problem, which is solved via continuous max-flow formulation. The scientific literature shows how these types of mathematical models have been used to segment different tumors, mainly in the lungs [4,26,27,42,44,48], liver [29], lymph nodes [5,13,16,17,54], prostate [20,25,31], brain [12] and breast [49]. Some studies used level-set and active contour methods [13,20,25,26,34,44,48], while others were based on the graph-cut method [4,7,12,16,17,27,31,42,54].…”

Section: Introductionmentioning

confidence: 99%

Three-dimensional segmentation of retroperitoneal masses using continuous convex relaxation and accumulated gradient distance for radiotherapy planning

Suárez-Mejías

Pérez-Carrasco

Serrano

et al. 2016

Med Biol Eng Comput

View full text Add to dashboard Cite

An innovative algorithm has been developed for the segmentation of retroperitoneal tumors in 3D radiological images. This algorithm makes it possible for radiation oncologists and surgeons semiautomatically to select tumors for possible future radiation treatment and surgery. It is based on continuous convex relaxation methodology, the main novelty being the introduction of accumulated gradient distance, with intensity and gradient information being incorporated into the segmentation process. The algorithm was used to segment 26 CT image volumes. The results were compared with manual contouring of the same tumors. The proposed algorithm achieved 90 % sensitivity, 100 % specificity and 84 % positive predictive value, obtaining a mean distance to the closest point of 3.20 pixels. The algorithm's dependence on the initial manual contour was also analyzed, with results showing that the algorithm substantially reduced the variability of the manual segmentation carried out by different specialists. The algorithm was also compared with four benchmark algorithms (thresholding, edge-based level-set, region-based level-set and continuous max-flow with two labels). To the best of our knowledge, this is the first time the segmentation of retroperitoneal tumors for radiotherapy planning has been addressed.

show abstract

“…Recent work on LN CADe has varied according to the feature types and learning algorithms used for training. [1,8] utilize direct 3D information from CT scans, performing boosting-based feature selection over a pool of 50-60 thousand 3D Haar wavelet features. Due to the curse of dimensionality (analyzed in [14]), such approaches can result in systems with limited sensitivity (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…Due to the curse of dimensionality (analyzed in [14]), such approaches can result in systems with limited sensitivity (e.g. 60.9% at 6.1 FP/scan for mediastinal LNs in [8]). Circumventing 3D feature computation during LN classification, [14] implements a shallow hierarchy of linear models operating on 2D slices or views of LN candidate volumes of interest (VOIs) with histograms of oriented gradients (HOG) [3] features.…”

Section: Introductionmentioning

confidence: 99%

Leveraging Mid-Level Semantic Boundary Cues for Automated Lymph Node Detection

Seff

Lü

Barbu

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Histograms of oriented gradients (HOG) are widely employed image descriptors in modern computer-aided diagnosis systems. Built upon a set of local, robust statistics of low-level image gradients, HOG features are usually computed on raw intensity images. In this paper, we explore a learned image transformation scheme for producing higher-level inputs to HOG. Leveraging semantic object boundary cues, our methods compute data-driven image feature maps via a supervised boundary detector. Compared with the raw image map, boundary cues offer mid-level, more object-specific visual responses that can be suited for subsequent HOG encoding. We validate integrations of several image transformation maps with an application of computer-aided detection of lymph nodes on thoracoabdominal CT images. Our experiments demonstrate that semantic boundary cues based HOG descriptors complement and enrich the raw intensity alone. We observe an overall system with substantially improved results (∼78% versus 60% recall at 3 FP/volume for two target regions). The proposed system also moderately outperforms the state-of-the-art deep convolutional neural network (CNN) system in the mediastinum region, without relying on data augmentation and requiring significantly fewer training samples.

show abstract

Lymph node detection and segmentation in chest CT data using discriminative learning and a spatial prior

Cited by 63 publications

References 30 publications

A New 2.5D Representation for Lymph Node Detection Using Random Sets of Deep Convolutional Neural Network Observations

A New 2.5D Representation for Lymph Node Detection Using Random Sets of Deep Convolutional Neural Network Observations

Three-dimensional segmentation of retroperitoneal masses using continuous convex relaxation and accumulated gradient distance for radiotherapy planning

Leveraging Mid-Level Semantic Boundary Cues for Automated Lymph Node Detection

Contact Info

Product

Resources

About