Prostate segmentation from 3D transrectal ultrasound using statistical shape models and various appearance models

Due to the high noise and artifacts typically encountered in ultrasound images, segmenting objects from this modality is one of the most challenging tasks in medical image analysis. Model-based approaches like statistical shape models (SSMs) incorporate prior knowledge that supports object detection in case of incomplete evidence from the image data. How well the model adapts to an unseen image is primarily determined by the suitability of the used appearance model, which evaluates the goodness of fit during model evolution. In this paper, we compare two gradient profile models with a region-based approach featuring local histograms to detect the prostate in 3D transrectal ultrasound (TRUS) images. All models are used within an SSM segmentation framework with optimal surface detection for outlier removal. Evaluation was performed using cross-validation on 35 datasets. While the histogram model failed in 10 cases, both gradient models had only 2 failures and reached an average surface distance of 1.16 ± 0.38 mm in comparison with interactively generated reference contours.

[1]  Edward L. Chaney,et al.  Histogram Statistics of Local Model-Relative Image Regions , 2005, DSSCV.

[2]  Osamu Ukimura,et al.  Real-time transrectal ultrasound guidance during laparoscopic radical prostatectomy: impact on surgical margins. , 2006, The Journal of urology.

[3]  W D Richard,et al.  Automated texture-based segmentation of ultrasound images of the prostate. , 1996, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[4]  Hans-Peter Meinzer,et al.  A Shape-Guided Deformable Model with Evolutionary Algorithm Initialization for 3D Soft Tissue Segmentation , 2007, IPMI.

[5]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[6]  Mingyue Ding,et al.  3D TRUS Image Segmentation in Prostate Brachytherapy , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[7]  Aaron Fenster,et al.  Prostate boundary segmentation from ultrasound images using 2D active shape models: Optimisation and extension to 3D , 2006, Comput. Methods Programs Biomed..

[8]  Hans-Peter Meinzer,et al.  Interactive constraints for 3D-simplex meshes , 2005, SPIE Medical Imaging.

[9]  Roberto Brunelli,et al.  Histograms analysis for image retrieval , 2001, Pattern Recognit..

[10]  Hans-Peter Meinzer,et al.  3D Active Shape Models Using Gradient Descent Optimization of Description Length , 2005, IPMI.

[11]  Max A. Viergever,et al.  Active-shape-model-based segmentation of abdominal aortic aneurysms in CTA images , 2002, SPIE Medical Imaging.

[12]  Yongmin Kim,et al.  Parametric shape modeling using deformable superellipses for prostate segmentation , 2004, IEEE Transactions on Medical Imaging.

[13]  David Djajaputra,et al.  Ultrasound-based guidance of intensity-modulated radiation therapy. , 2006, Medical dosimetry : official journal of the American Association of Medical Dosimetrists.

[14]  Timothy F. Cootes,et al.  Active Shape Model Search using Local Grey-Level Models: A Quantitative Evaluation , 1993, BMVC.

[15]  Dinggang Shen,et al.  Deformable segmentation of 3-D ultrasound prostate images using statistical texture matching method , 2006, IEEE Transactions on Medical Imaging.

[16]  Mingyue Ding,et al.  Fast prostate segmentation in 3D TRUS images based on continuity constraint using an autoregressive model. , 2007, Medical physics.

[17]  Tao Zhang,et al.  Model-based segmentation of medical imagery by matching distributions , 2005, IEEE Transactions on Medical Imaging.