Mohamad Ali POURMINA, Mohammad-Shahram MOIN, Ahad SALIMI

Automatic prostate segmentation using multiobjective active appearance model in MR images

Prostate cancer is the second largest cause of mortality among men. Prostate segmentation, i.e. theprecise determination of the prostate region in magnetic resonance imaging (MRI), is generally used for prostate volumemeasurement, which can be used as a potential prostate cancer indicator. This paper presents a new fully automaticstatistical 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 applyingprincipal component analysis to the training images, already outlined by a physician. Then noise and roughness areproperly removed in the preprocessing step by Sticks filter and nonlinear filtering. This helps us provide the properconditions for the prostate region detection. Finally, in order to detect the prostate region, a new multiobjectivefunction is optimized using a suitable search algorithm. The proposed method has been applied to prostate images forsegmenting the prostate boundaries. The evaluation results indicate that the presented method can yield a DSC value of87.4 ± 5.00%, is less sensitive to the edge information and initialization, and has a stronger capture range in comparisonwith existing methods.

PDF

___

[1] Siegel RL, Miller KD, Jemal A. Cancer statistics. CA Cancer Journal for Clinicians 2018; 68 (1): 7-30. doi: 10.3322 /caac.21442
[2] Ghose S, Oliver A, Marti R, Liado X, Vilanova JC et al. A survey of prostate segmentation methodologies in ultrasound, magnetic resonance and computed tomography images. Computer Methods and Programs in Biomedicine 2012; 108 (1): 262-287. doi: 10.3322 /caac.21442
[3] Litjens G, Toth R, Van de ven W, Hoeks C, Kerkstra S et al. Evaluation of prostate segmentation algorithms for MRI. Medical Image Analysis 2014; 18 (2): 359-373. doi: 10.3322 /caac.21442
[4] Toth R, Ribault J, Gentile J, Sperling D, Madabhushi A. Simultaneous segmentation of prostatic zones using active appearance models with multiple coupled levelsets. Computer Vision and Image Understanding 2013; 117 (1): 1051-1060. doi: 10.1016/j.cviu.2012.11.013
[5] Toth R, Madabhushi A. Multi-feature landmark-free active appearance models: application to prostate MRI segmentation. IEEE Transactions on Medical Imaging 2012; 31 (8): 1638-1650. doi: 10.1109/TMI.2012.2201498
[6] Maan B, van der Heijden F. Prostate MR image segmentation using 3d active appearance models. In: MICCAI 2012 Grand Challenge: Prostate MRI Image Segmentation; USA; 2012. pp. 44-51. doi: 10.1007/978-3-319-03731-8_20
[7] Korsager AS, Stephansen UL, Carl J, Ostergaard LR. The use of an active appearance model for automated prostate segmentation in magnetic resonance. Acta Oncologica 2013; 52 (7): 1374-1377. doi: 10.3109/0284186X.2013.822099
[8] Chandra SS, Dowling JA, Shen KK, Raniga P, Pluim JP et al. Patient specific prostate segmentation in 3-d magnetic resonance images. IEEE Transactions on Medical Imaging 2012; 31 (10): 1955-1964. doi: 10.1109/TMI.2012.2211377
[9] Vincent G, Guillard G, Bowes M. Fully automatic segmentation of the prostate using active appearance models. In: MICCAI 2012 Grand Challenge: Prostate MR Image Segmentation; USA; 2012. pp. 35-43. doi: 10.1371/journal.pone.0076645
[10] Ling M, Guo R, Tian Z, Venkataraman R, Sarkar S et al. A random walk-based segmentation framework for 3d ultrasound images of the prostate. Medical Physics 2017; 44 (10): 5128-5142. doi: 10.1002/mp.12396
[11] Egger J. Pcg-cut: graph driven segmentation of the prostate central gland. PLoS One 2013; 8 (10): e76645. doi: 10.1371/journal.pone.0076645
[12] Egger J, Mostarkic Z, Grosskopf S, Freisleben B. A fast vessel center line extraction algorithm for catheter simulation. In: CBMS 2007 Twentieth IEEE International Symposium on Computer Based Medical Systems; Maribor, Slovenia; 2007. pp.177-182. doi: 10.1109 /CBMS.2007.5
[13] Tian Z, Liu L, Zhang Z, Fei B. Superpixel-based segmentation for 3d prostate mr images. IEEE Transactions on Medical Imaging 2016; 35 (3): 791-801. doi: 10.1109/TMI.2015.2496296
[14] Mahapatra D. Graph cut based automatic prostate segmentation using learned semantic information. In: ISBI 2013 IEEE 10th International Symposium on Biomedical Imaging; San Francisco, CA, USA; 2013. pp. 1316-1319. doi: 10.1109/ISBI.2013.6556774
[15] Gao Y, Wang L, Shao Y, Shen D. Learning distance transform for boundary detection and deformable segmentation in ct prostate images. In: MLMI 2014 International Workshop on Machine Learning in Medical Imaging; Cham, Switzerland; 2014. pp. 93-100. doi: 10.1007/978-3-319-10581-9_12.
[16] Liao S, Gao Y, Shi Y, Yousuf A, Karademir I et al. Automatic prostate MR image segmentation with sparse label propagation and domain-specic manifold regularization. In: IPMI 2013 International Conference on Information Processing in Medical Imaging; Berlin, Germany; 2013. pp. 511-523. doi: 10.1007/978-3-642-38868-2_43
[17] Padgett K, Swallen A, Nelson A, Pollack A, Stoyanova R. Sufj171: Robust atlas based segmentation of the prostate and peripheral zone regions on mri utilizing multiple MRI system vendors. Medical Physics 2016; 43 (6): 3447-3447. doi: 10.1118/1.495607
[18] Khurd P, Grady L, Gajera K, Diallo M, Gall P et al. Facilitating 3d spectroscopic imaging through automatic prostate localization in MR images using random walker segmentation initialized via boosted classiers. In: International Workshop on Prostate Cancer Imaging; Berlin, Germany; 2011. pp. 47-56. doi: 10.1007/978-3-642-23944-1_5
[19] Ghose S, Mitra J, Oliver A, Marti R, Llado X et al. A supervised learning framework for automatic prostate segmentation in trans rectal ultrasound images. In: ACIVS 2012 International Conference on Advanced Concepts for Intelligent Vision Systems; Berlin, Germany; 2012. pp. 190-200. doi: 10.1007/ 978-3-642-33140-4_17
[20] Qiu W, Yuan J, Ukwatta E, Sun Y, Rajchl M et al. Fast globally optimal segmentation of 3d prostate MRI with axial symmetry prior. In: MICCAI 2013 International Conference on Medical Image Computing and Computer-Assisted Intervention; Berlin, Germany; 2013. pp. 198-205. doi: 10.1007/978-3-642-40763-5_25
[21] Yu L, Yang X, Chen H, Qin J, Heng PA. Volumetric convents with mixed residual connections for automated prostate segmentation from 3d MR images. In: AAAI 2017 Thirty-First Conference on Artificial Intelligence; San Francisco, CA, USA; 2017. pp. 66-72.
[22] Milletari F, Navab N, Ahmadi SA. V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: 3DV 2016 Fourth International Conference on 3D Vision; Stanford, CA, USA; 2016. pp. 565-571. doi: 10.1109/3DV.2016.79
[23] He B, Xiao D, Hu Q, Jia F. Automatic magnetic resonance image prostate segmentation based on adaptive feature learning probability boosting tree initialization and CNN-ASM refinement. IEEE Access 2018; 6: 2005-2015. doi: 10.1109/ACCESS.2017.2781278
[24] Ling M, Guo R, Zhang G, Schuster DM, Fei B. A combined learning algorithm for prostate segmentation on 3d CT images. Medical Physics 2017; 44 (11): 5768-5781. doi: 10.1002/mp.12528
[25] Martin S, Troccaz J, Daanen V. Automated segmentation of the prostate in 3d MR images using a probabilistic atlas and a spatially constrained deformable model. Medical Physics 2010; 37 (4): 1579-1590. doi: 10.1118/1.3315367
[26] Drozdzal M, Chartrand G, Vorontsov E, Shakeri M, Di Jorio L et al. Learning normalized inputs for iterative estimation in medical image segmentation. Medical Image Analysis 2018; 44: 1-13. doi: 10.1016/j.media.2017.11.005
[27] Cootes TF, Edwards GJ, Taylor CJ. Active appearance models. IEEE Transactions on Pattern Analysis and Machine Intelligence 2001; 23: 681-685. doi: 10.1109/34.927467
[28] Bookstein FL. Principal warps: Thin-plate splines and the decomposition of deformations. IEEE Transactions on Pattern Analysis and Machine Intelligence 1989; 11 (6): 567-585. doi: 10.1109/34.24792.
[29] Awad J, Abdel-Galil T, Salama M, Tizhoosh H, Fenster A et al. Prostate’s boundary detection in transrectal ultrasound images using scanning technique. In: IEEE CCECE 2003 Canadian Conference on Electrical and Computer Engineering; Montreal, Canada; 2003. pp. 1199-1202. doi: 10.1109/CCECE.2003.1226113
[30] Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence 1990; 12 (7): 629-639. doi: 10.1109/34.56205
[31] Xu C, Prince JL. Snakes, shapes, and gradient vector flow. IEEE Transactions on Image Processing 1998; 7 (3): 359-369. doi: 10.1109/83.661186
[32] Li B, Acton ST. Active contour external force using vector field convolution for image segmentation. IEEE Transactions on Image Processing 2007; 16 (8): 2096-2106. doi: 10.1109/TIP.2007.899601
[33] Salimi A, Pourmina A, Moien MS. Fully automatic prostate segmentation in MR images using a new hybrid active contour-based approach. Signal, Image and Video Processing 2018; 12 (8): 1629-1637. doi: 10.1007/s11760-018- 1320-y