A magnetic resonance spectroscopy driven initialization scheme for active shape model based prostate segmentation

Segmentation of the prostate boundary on clinical images is useful in a large number of applications including calculation of prostate volume pre- and post-treatment, to detect extra-capsular spread, and for creating patient-specific anatomical models. Manual segmentation of the prostate boundary is, however, time consuming and subject to inter- and intra-reader variability. T2-weighted (T2-w) magnetic resonance (MR) structural imaging (MRI) and MR spectroscopy (MRS) have recently emerged as promising modalities for detection of prostate cancer in vivo. MRS data consists of spectral signals measuring relative metabolic concentrations, and the metavoxels near the prostate have distinct spectral signals from metavoxels outside the prostate. Active Shape Models (ASM's) have become very popular segmentation methods for biomedical imagery. However, ASMs require careful initialization and are extremely sensitive to model initialization. The primary contribution of this paper is a scheme to automatically initialize an ASM for prostate segmentation on endorectal in vivo multi-protocol MRI via automated identification of MR spectra that lie within the prostate. A replicated clustering scheme is employed to distinguish prostatic from extra-prostatic MR spectra in the midgland. The spatial locations of the prostate spectra so identified are used as the initial ROI for a 2D ASM. The midgland initializations are used to define a ROI that is then scaled in 3D to cover the base and apex of the prostate. A multi-feature ASM employing statistical texture features is then used to drive the edge detection instead of just image intensity information alone. Quantitative comparison with another recent ASM initialization method by Cosio showed that our scheme resulted in a superior average segmentation performance on a total of 388 2D MRI sections obtained from 32 3D endorectal in vivo patient studies. Initialization of a 2D ASM via our MRS-based clustering scheme resulted in an average overlap accuracy (true positive ratio) of 0.60, while the scheme of Cosio yielded a corresponding average accuracy of 0.56 over 388 2D MR image sections. During an ASM segmentation, using no initialization resulted in an overlap of 0.53, using the Cosio based methodology resulted in an overlap of 0.60, and using the MRS-based methodology resulted in an overlap of 0.67, with a paired Student's t-test indicating statistical significance to a high degree for all results. We also show that the final ASM segmentation result is highly correlated (as high as 0.90) to the initialization scheme.

[1]  David G. Stork,et al.  Pattern Classification , 1973 .

[2]  Olivier Ledoit,et al.  A well-conditioned estimator for large-dimensional covariance matrices , 2004 .

[3]  Timothy F. Cootes,et al.  Multi-resolution search with active shape models , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[4]  Milan Sonka,et al.  Object localization and border detection criteria design in edge-based image segmentation: automated learning from examples , 2000, IEEE Transactions on Medical Imaging.

[5]  Arthur Hung,et al.  A comparison of CT scan to transrectal ultrasound-measured prostate volume in untreated prostate cancer. , 2003, International journal of radiation oncology, biology, physics.

[6]  Anant Madabhushi,et al.  A hierarchical spectral clustering and nonlinear dimensionality reduction scheme for detection of prostate cancer from magnetic resonance spectroscopy (MRS). , 2009, Medical physics.

[7]  Jayaram K. Udupa,et al.  New methods of MR image intensity standardization via generalized scale , 2005, SPIE Medical Imaging.

[8]  William M. Wells,et al.  Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation , 2004, IEEE Transactions on Medical Imaging.

[9]  A S Dewalle,et al.  3D delineation of prostate, rectum and bladder on MR images. , 2008, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[10]  R. Zwiggelaar,et al.  Prostate segmentation: a comparative study , 2003 .

[11]  Rajeev Kumar,et al.  Potential of magnetic resonance spectroscopic imaging in predicting absence of prostate cancer in men with serum prostate-specific antigen between 4 and 10 ng/ml: a follow-up study. , 2008, Urology.

[12]  Aaron Fenster,et al.  Prostate segmentation algorithm using dyadic wavelet transform and discrete dynamic contour. , 2004, Physics in medicine and biology.

[13]  J. Vilanova,et al.  Prostate cancer detection: magnetic resonance (MR) spectroscopic imaging , 2007, Abdominal Imaging.

[14]  Ying Lu,et al.  High-grade prostatic intraepithelial neoplasia in patients with prostate cancer: MR and MR spectroscopic imaging features--initial experience. , 2007, Radiology.

[15]  A. Evans,et al.  Unsupervised segmentation of the prostate using MR images based on level set with a shape prior , 2009, 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[16]  Johan Montagnat,et al.  Volumetric medical images segmentation using shape constrained deformable models , 1997, CVRMed.

[17]  Timothy F. Cootes,et al.  Statistical models of appearance for computer vision , 1999 .

[18]  Melanie Mitchell,et al.  An introduction to genetic algorithms , 1996 .

[19]  J E Wickham,et al.  Transrectal ultrasonography: why are estimates of prostate volume and dimension so inaccurate? , 1996, British journal of urology.

[20]  Jayaram K. Udupa,et al.  Generalized scale: Theory, algorithms, and application to image inhomogeneity correction , 2006, Comput. Vis. Image Underst..

[21]  R. Viard,et al.  Automatic 3D segmentation of prostate in MRI combining a priori knowledge, Markov fields and Bayesian framework , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[22]  Anant Madabhushi,et al.  WERITAS: weighted ensemble of regional image textures for ASM segmentation , 2009, Medical Imaging.

[23]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[24]  J. Kurhanewicz,et al.  Combined magnetic resonance imaging and spectroscopic imaging approach to molecular imaging of prostate cancer , 2002, Journal of magnetic resonance imaging : JMRI.

[25]  Reyer Zwiggelaar,et al.  Semi-automatic Segmentation of the Prostate , 2003, IbPRIA.

[26]  Dorothea Heiss-Czedik,et al.  An Introduction to Genetic Algorithms. , 1997, Artificial Life.

[27]  Maximilien Vermandel,et al.  3D delineation of prostate, rectum and bladder on MR images , 2008, Comput. Medical Imaging Graph..

[28]  Yongmin Kim,et al.  Edge-guided boundary delineation in prostate ultrasound images , 2000, IEEE Transactions on Medical Imaging.

[29]  Dimitris N. Metaxas,et al.  A 3D global-to-local deformable mesh model based registration and anatomy-constrained segmentation method for image guided prostate radiotherapy. , 2010, Medical physics.

[30]  William M. Wells,et al.  Validation of Image Segmentation and Expert Quality with an Expectation-Maximization Algorithm , 2002, MICCAI.

[31]  J. Kurhanewicz,et al.  Magnetic Resonance Imaging and Spectroscopic Imaging of Prostate Cancer , 2001, Cancer investigation.

[32]  Timothy F. Cootes,et al.  Using grey-level models to improve active shape model search , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[33]  Gabriel Thomas,et al.  Semi automatic MRI prostate segmentation based on wavelet multiscale products , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[34]  Alexandra L Hanlon,et al.  Relationship between prostate volume, prostate-specific antigen nadir, and biochemical control. , 2002, International journal of radiation oncology, biology, physics.

[35]  J.S. Suri,et al.  Three Different Strategies for Real-Time Prostate Capsule Volume Computation from 3-D End-Fire Transrectal Ultrasound , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[36]  Milan Sonka,et al.  3-D active appearance models: segmentation of cardiac MR and ultrasound images , 2002, IEEE Transactions on Medical Imaging.

[37]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[38]  Jayaram K. Udupa,et al.  Interplay between intensity standardization and inhomogeneity correction in MR image processing , 2005, IEEE Transactions on Medical Imaging.

[39]  Fernando Arámbula Cosío,et al.  Automatic initialization of an active shape model of the prostate , 2008, Medical Image Anal..

[40]  Alejandro F. Frangi,et al.  Active shape model segmentation with optimal features , 2002, IEEE Transactions on Medical Imaging.

[41]  Paul Suetens,et al.  Minimal Shape and Intensity Cost Path Segmentation , 2007, IEEE Transactions on Medical Imaging.

[42]  W. Catalona,et al.  Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. , 1991, The New England journal of medicine.

[43]  Gabor Fichtinger,et al.  A Coupled Global Registration and Segmentation Framework With Application to Magnetic Resonance Prostate Imagery , 2010, IEEE Transactions on Medical Imaging.

[44]  J. Hogg Magnetic resonance imaging. , 1994, Journal of the Royal Naval Medical Service.

[45]  Marleen de Bruijne,et al.  Adapting Active Shape Models for 3D Segmentation of Tubular Structures in Medical Images , 2003, IPMI.

[46]  Dimitris N. Metaxas,et al.  Automated detection of prostatic adenocarcinoma from high-resolution ex vivo MRI , 2005, IEEE Transactions on Medical Imaging.

[47]  Wen Gao,et al.  An improved active shape model for face alignment , 2002, Proceedings. Fourth IEEE International Conference on Multimodal Interfaces.

[48]  Hervé Delingette,et al.  Automatic Segmentation of Bladder and Prostate Using Coupled 3D Deformable Models , 2007, MICCAI.

[49]  Christopher J. Taylor,et al.  Automatic Measurement of Vertebral Shape using Active Shape Models , 1996, BMVC.

[50]  Anant Madabhushi,et al.  A Hierarchical Unsupervised Spectral Clustering Scheme for Detection of Prostate Cancer from Magnetic Resonance Spectroscopy (MRS) , 2007, MICCAI.

[51]  Jinghao Zhou,et al.  Object-constrained meshless deformable algorithm for high speed 3D nonrigid registration between CT and CBCT. , 2009, Medical physics.

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

[53]  Hee-Won Kim,et al.  In Vivo Prostate Magnetic Resonance Imaging and Magnetic Resonance Spectroscopy at 3 Tesla Using a Transceive Pelvic Phased Array Coil: Preliminary Results , 2003, Investigative radiology.

[54]  J. Kurhanewicz,et al.  Prostate cancer: sextant localization with MR imaging, MR spectroscopy, and 11C-choline PET/CT. , 2007, Radiology.

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

[56]  Mark Fisher,et al.  Improving ASM Search Using Mixture Models for Grey-Level Profiles , 2005, IbPRIA.

[57]  Timothy F. Cootes,et al.  The Use of Active Shape Models for Locating Structures in Medical Images , 1993, IPMI.

[58]  P. Carroll,et al.  Three-dimensional H-1 MR spectroscopic imaging of the in situ human prostate with high (0.24-0.7-cm3) spatial resolution. , 1996, Radiology.

[59]  A Fenster,et al.  Prostate boundary segmentation from 2D ultrasound images. , 2000, Medical physics.

[60]  Anant Madabhushi,et al.  Consensus-Locally Linear Embedding (C-LLE): Application to Prostate Cancer Detection on Magnetic Resonance Spectroscopy , 2008, MICCAI.

[61]  Aaron Fenster,et al.  Prostate boundary segmentation from 3D ultrasound images. , 2003, Medical physics.

[62]  Stefan Klein,et al.  Automatic segmentation of the prostate in 3D MR images by atlas matching using localized mutual information. , 2008, Medical physics.

[63]  William Wells,et al.  Detection of prostate cancer by integration of line-scan diffusion, T2-mapping and T2-weighted magnetic resonance imaging; a multichannel statistical classifier. , 2003, Medical physics.

[64]  M. Zaider,et al.  Treatment planning for prostate implants using magnetic-resonance spectroscopy imaging. , 2000, International journal of radiation oncology, biology, physics.

[65]  Shigeo Abe DrEng Pattern Classification , 2001, Springer London.

[66]  James S. Duncan,et al.  Medical Image Analysis , 1999, IEEE Pulse.