Automatic 3D Prostate MR Image Segmentation Using Graph Cuts and Level Sets with Shape Prior

Xiong, Wei; Li, Anthony Lianjie; Ong, Sim Heng; Sun, Yeneng

doi:10.1007/978-3-319-03731-8_20

Cited by 7 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In response, Toth et al [17] proposed a novel landmark-free AAM. Other methods that have been applied in deformable models include level set [18,19], and active contour models [20]. Computer vision with object segmentation [21][22][23] and saliency detection [24][25][26][27]) has progressed markedly in recent years and so many investigators applied transfer learning from general images to biomedical images for segmentation.…”

Section: Related Workmentioning

confidence: 99%

A propagation-DNN: Deep combination learning of multi-level features for MR prostate segmentation

Yan

Wang

Kim

et al. 2019

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

Background and Objective: Prostate segmentation on Magnetic Resonance (MR) imaging is problematic because disease changes the shape and boundaries of the gland and it can be difficult to separate the prostate from surrounding tissues. We propose an automated model that extracts and combines multi-level features in a deep neural network to segment prostate on MR images. Methods: Our proposed model, the Propagation Deep Neural Network (P-DNN), incorporates the optimal combination of multilevel feature extraction as a single model. High level features from the convolved data using DNN are extracted for prostate localization and shape recognition, while labeling propagation, by low level cues, is embedded into a deep layer to delineate the prostate boundary. Results: A well-recognized benchmarking dataset (50 training data and 30 testing data from patients) was used to evaluate the P-DNN. When compared it to existing DNN methods, the P-DNN statistically outperformed the baseline DNN models with an average improvement in the DSC of 3.19%. When compared to the state-of-the-art non-DNN prostate segmentation methods, P-DNN was competitive by achieving 89.9 ±2.8% DSC and 6.84±2.5mm HD on training sets and 84.13±5.18% DSC and 9.74±4.21mm HD on testing sets. Conclusion: Our results show that P-DNN maximizes multi-level feature extraction for prostate segmentation of MR images.

show abstract

Section: Related Workmentioning

confidence: 99%

A propagation-DNN: Deep combination learning of multi-level features for MR prostate segmentation

Yan

Wang

Kim

et al. 2019

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

“…It determines the desired object in the new image by applying a registration-based method. Other prostate region segmentation works include a level-set based model that uses shape and texture data to predict the model and positional parameters [6] and a 3-D scheme based on matched shape contents with AAM, which utilizes PCA to combine the shape information [7]. Chandra et al [8] proposed an automatic deformable model for prostate segmentation in MR images.…”

Section: Introductionmentioning

confidence: 99%

Automatic prostate segmentation using multiobjective active appearance modelin MR images

Salimi¹,

Pourmina²,

Moin³

2019

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Prostate cancer is the second largest cause of mortality among men. Prostate segmentation, i.e. the precise determination of the prostate region in magnetic resonance imaging (MRI), is generally used for prostate volume measurement, which can be used as a potential prostate cancer indicator. This paper presents a new fully automatic statistical model called the multiobjective active appearance model (MOAAM) for prostate segmentation in MR images.First, in the training stage, the appearance model, including the shape and texture model, is developed by applying principal component analysis to the training images, already outlined by a physician. Then noise and roughness are properly removed in the preprocessing step by Sticks filter and nonlinear filtering. This helps us provide the proper conditions for the prostate region detection. Finally, in order to detect the prostate region, a new multiobjective function is optimized using a suitable search algorithm. The proposed method has been applied to prostate images for segmenting the prostate boundaries. The evaluation results indicate that the presented method can yield a DSC value of 87.4 ± 5.00% , is less sensitive to the edge information and initialization, and has a stronger capture range in comparison with existing methods.

show abstract

“…Xin Liu [13] utilized region-based active contour model and shape information to segment prostate without training dataset. Wei Xiong [14] realized the accurate prostate segmentation by combining 3D graph cuts and 3D geodesic active contour shape prior level set. R. Toth and A. Madabhushi [15] presented a novel active appearance models methodology that utilized the level set to realize the most accurate segmentation.…”

Section: Introductionmentioning

confidence: 99%

MR T1 Image Segmentation of a Prostate Based on Distance Regularized Level Set Evolution

Zhang¹,

Peng²,

Liu³

et al. 2016

IJHIT

View full text Add to dashboard Cite

show abstract

Automatic 3D Prostate MR Image Segmentation Using Graph Cuts and Level Sets with Shape Prior

Cited by 7 publications

References 10 publications

A propagation-DNN: Deep combination learning of multi-level features for MR prostate segmentation

A propagation-DNN: Deep combination learning of multi-level features for MR prostate segmentation

Automatic prostate segmentation using multiobjective active appearance modelin MR images

MR T1 Image Segmentation of a Prostate Based on Distance Regularized Level Set Evolution

Contact Info

Product

Resources

About