·1370 工程科学学报,第42卷,第10期 表1尿道特征点定位结果对比 MRI/TRUS融合成像引导靶向前列腺穿刺中的诊断价值.中华 男科学杂志,2016,22(9):782) Table 1 Comparison of location results of urethral points [5]Schlenker B.Apfelbeck M,Buchner A,et al.MRI-TRUS fusion Method proposed Literature method Sample biopsy of the prostate:quality of image fusion in a clinical setting. DSC TE/mm DSC TE/mm Clin Hemorheol Microcirculat,2018,70(4):433 1 0.9938 1.76 0.9939 1.93 [6] Mitra J,Marti R,Oliver A,et al.Prostate multimodality image 3 0.9873 1.24 0.9816 3.84 registration based on B-splines and quadrature local energy.Int/ 0.9897 1.92 0.9893 1.27 Comput Assisted Radiol Surg,2012,7(3):445 [7] 0.9906 1.58 Sun Y,Yuan J,Qiu W,et al.Three-dimensional nonrigid MR 0.9872 2.55 TRUS registration using dual optimization.IEEE Trans Med Imag, 0.9871 1.47 0.9884 1.42 2015,34(5):1085 s 0.9897 1.59 0.9880 2.20 [8] Moradi M,Janoos F,Fedorov A,et al.Two solutions for d 0.0024 0.23 0.0039 0.93 registration of ultrasound to MRI for image-guided prostate interventions /2012 Annual International Conference of the IEEE 4结论 Engineering in Medicine and Biology Society.San Diego,2012: 1129 (1)针对前列腺核磁超声融合引导穿刺手术 [9]Fedorov A,Khallaghi S.Sanchez C A,et al.Open-source image 中前列腺的变形配准问题,提出一种基于监督学 registration for MRI-TRUS fusion-guided prostate interventions. 习的超声图像自动分割方法,并与术前核磁图像 Int J Comput Assisted Radiol Surg,2015,10(6):925 进行图像配准 [10]Ni D,Wu H L.MRI-TRUS multi-modality image fusion for (2)新方法通过引入随机森林分类器建立了 targeted prostate biopsy.J Shenhen Univ Sci Eng,2016,33(2): 边界驱动的姿态估计模型,实现前列腺超声图像 111 的自动分割,与专家轮廓对比具有高的精度 (倪东,吴海浪.基于核磁-超声融合的前列腺靶向穿刺系统.深 圳大学学报:理工版,2016,33(2):111) (3)在图像配准方面,使用了形状矢量来构建 [11]Wang WR.Research on Prostate Puncture Assisted by MR and 薄板样条的配对特征点同时引入各向异性误差作 TRUS Image[Dissertation].Harbin:Harbin Institute of 为正则因子,配准结果表明,与传统方法相比,新 Technology,2018 方法在图像的配准方面具有较高的精度,在前列 (王炜荣.MR与TRUS图像辅助前列腺穿刺技术研究学位论 腺核磁超声图像融合引导方面具有临床应用价 文]哈尔滨:哈尔滨工业大学,2018) 值,是一种准确、稳定的图像配准方法 [12]Du C.MR and TRUS Image Denoising and Segmentation Methods in Prostate Puncture Guidance[Dissertation].Harbin:Harbin 参考文献 Institute of Technology,2019 (杜超.前列腺穿刺引导中的MR和TRUS图像去噪与分割方法 [1]Deng Y S,He Y H,Zhou X F.Development of prostate targeted [学位论文].哈尔滨:哈尔滨工业大学,2019) puncture technology.J Mimimally Imvasive Urology,2018,7(6): [13]Cootes T F,Edwards G J,Taylor C J.Active appearance models. 428 IEEE Trans Pattern Anal Mach Intellig,2001,23(6):681 (邓益森,何宇辉,周晓峰.前列腺靶向穿刺技术发展概况.微创 [14]Bookstein F L.Principal warps:thin-plate splines and the 泌尿外科杂志,2018,7(6):428) decomposition of deformations.IEEE Trans Pattern Anal Mach [2]Guichard G,Larre,Gallina A,et al.Extended 21-sample needle Intellig,1989,11(6):567 biopsy protocol for diagnosis of prostate cancer in 1000 [15]Svetnik V,Liaw A,Tong C,et al.Random forest:a classification consecutive patients.Eur Urol,2007.52(2):430 and regression tool for compound classification and QSAR [3]Zhou Z E,Yan W G,Zhou Y,et al.Recent progress in MRI- modeling.JChem Inf Comput Sci,003,43(6):1947 ultrasound fusion for guidance of targeted prostate biopsy.ChinJ [16]Rohr K,Forefett M,Stiehl H S.Spline-based elastic image Srg,2016,54(10):792 registration:integration of landmark errors and orientation (周智恩,严维刚,周毅,等.MRI超声融合引导下前列腺靶向穿 attributes.Comput Vision Image Understand,2003,9(2):153 刺活检的最新进展.中华外科杂志,2016,54(10):792) [17]Evangelidis G D,Psarakis E Z.Parametric image alignment using [4]Qu H W,Liu H,Cui Z L,et al.Focusing on MRI-suspected lesions enhanced correlation coefficient maximization./EEE Trans in targeted transrectal prostate biopsy guided by MRI-TRUS Pattern Anal Mach Intellig,2008,30(10):1858 fusion imaging for the diagnosis of prostate cancer.Natl JAndrol, [18]Ghosh P,Mitchell M,Tanyi J A,et al.Incorporating priors for 2016,22(9):782 medical image segmentation using a genetic algorithm. (曲华伟,刘辉,崔子连,等.重点穿刺MR可疑病灶区域在 Neurocomputing,2016,195:1814 结论 (1)针对前列腺核磁超声融合引导穿刺手术 中前列腺的变形配准问题,提出一种基于监督学 习的超声图像自动分割方法,并与术前核磁图像 进行图像配准. (2)新方法通过引入随机森林分类器建立了 边界驱动的姿态估计模型,实现前列腺超声图像 的自动分割,与专家轮廓对比具有高的精度. (3)在图像配准方面,使用了形状矢量来构建 薄板样条的配对特征点同时引入各向异性误差作 为正则因子,配准结果表明,与传统方法相比,新 方法在图像的配准方面具有较高的精度,在前列 腺核磁超声图像融合引导方面具有临床应用价 值,是一种准确、稳定的图像配准方法. 参 考 文 献 Deng Y S, He Y H, Zhou X F. Development of prostate targeted puncture technology. J Mimimally Invasive Urology, 2018, 7(6): 428 (邓益森, 何宇辉, 周晓峰. 前列腺靶向穿刺技术发展概况. 微创 泌尿外科杂志, 2018, 7(6):428) [1] Guichard G, Larré, Gallina A, et al. Extended 21-sample needle biopsy protocol for diagnosis of prostate cancer in 1000 consecutive patients. Eur Urol, 2007, 52(2): 430 [2] Zhou Z E, Yan W G, Zhou Y, et al. Recent progress in MRIultrasound fusion for guidance of targeted prostate biopsy. Chin J Surg, 2016, 54(10): 792 (周智恩, 严维刚, 周毅, 等. MRI-超声融合引导下前列腺靶向穿 刺活检的最新进展. 中华外科杂志, 2016, 54(10):792) [3] Qu H W, Liu H, Cui Z L, et al. Focusing on MRI-suspected lesions in targeted transrectal prostate biopsy guided by MRI-TRUS fusion imaging for the diagnosis of prostate cancer. Natl J Androl, 2016, 22(9): 782 (曲华伟, 刘辉, 崔子连, 等. 重点穿刺MRI可疑病灶区域在 [4] MRI/TRUS融合成像引导靶向前列腺穿刺中的诊断价值. 中华 男科学杂志, 2016, 22(9):782) Schlenker B, Apfelbeck M, Buchner A, et al. MRI-TRUS fusion biopsy of the prostate: quality of image fusion in a clinical setting. Clin Hemorheol Microcirculat, 2018, 70(4): 433 [5] Mitra J, Martí R, Oliver A, et al. Prostate multimodality image registration based on B-splines and quadrature local energy. Int J Comput Assisted Radiol Surg, 2012, 7(3): 445 [6] Sun Y, Yuan J, Qiu W, et al. Three-dimensional nonrigid MRTRUS registration using dual optimization. IEEE Trans Med Imag, 2015, 34(5): 1085 [7] Moradi M, Janoos F, Fedorov A, et al. Two solutions for registration of ultrasound to MRI for image-guided prostate interventions // 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. San Diego, 2012: 1129 [8] Fedorov A, Khallaghi S, Sánchez C A, et al. Open-source image registration for MRI-TRUS fusion-guided prostate interventions. Int J Comput Assisted Radiol Surg, 2015, 10(6): 925 [9] Ni D, Wu H L. MRI-TRUS multi-modality image fusion for targeted prostate biopsy. J Shenzhen Univ Sci Eng, 2016, 33(2): 111 (倪东, 吴海浪. 基于核磁-超声融合的前列腺靶向穿刺系统. 深 圳大学学报: 理工版, 2016, 33(2):111) [10] Wang W R. Research on Prostate Puncture Assisted by MR and TRUS Image[Dissertation]. Harbin: Harbin Institute of Technology, 2018 (王炜荣. MR与TRUS图像辅助前列腺穿刺技术研究[学位论 文]. 哈尔滨: 哈尔滨工业大学, 2018) [11] Du C. MR and TRUS Image Denoising and Segmentation Methods in Prostate Puncture Guidance[Dissertation]. Harbin: Harbin Institute of Technology, 2019 (杜超. 前列腺穿刺引导中的MR和TRUS图像去噪与分割方法 [学位论文]. 哈尔滨: 哈尔滨工业大学, 2019) [12] Cootes T F, Edwards G J, Taylor C J. Active appearance models. IEEE Trans Pattern Anal Mach Intellig, 2001, 23(6): 681 [13] Bookstein F L. Principal warps: thin-plate splines and the decomposition of deformations. IEEE Trans Pattern Anal Mach Intellig, 1989, 11(6): 567 [14] Svetnik V, Liaw A, Tong C, et al. Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci, 2003, 43(6): 1947 [15] Rohr K, Fornefett M, Stiehl H S. Spline-based elastic image registration: integration of landmark errors and orientation attributes. Comput Vision Image Understand, 2003, 90(2): 153 [16] Evangelidis G D, Psarakis E Z. Parametric image alignment using enhanced correlation coefficient maximization. IEEE Trans Pattern Anal Mach Intellig, 2008, 30(10): 1858 [17] Ghosh P, Mitchell M, Tanyi J A, et al. Incorporating priors for medical image segmentation using a genetic algorithm. Neurocomputing, 2016, 195: 181 [18] 表 1 尿道特征点定位结果对比 Table 1 Comparison of location results of urethral points Sample Method proposed Literature method DSC TE/mm DSC TE/mm 1 0.9938 1.76 0.9939 1.93 2 0.9873 1.24 0.9816 3.84 3 0.9897 1.92 0.9893 1.27 4 0.9906 1.58 0.9872 2.55 5 0.9871 1.47 0.9884 1.42 AP 0.9897 1.59 0.9880 2.20 d2 0.0024 0.23 0.0039 0.93 · 1370 · 工程科学学报,第 42 卷,第 10 期