您的位置：山东大学 -> 科技期刊社 -> 《山东大学学报（医学版）》

山东大学学报 (医学版) ›› 2023, Vol. 61 ›› Issue (3): 80-89.doi: 10.6040/j.issn.1671-7554.0.2022.1280

• 临床医学 • 上一篇    

130例锥形束CT影像腰椎椎弓根螺钉自动规划的初步分析

刘亚军1,2,袁强1,吴静晔1,韩晓光1,郎昭1,张勇3   

  1. 1.北京积水潭医院脊柱外科, 北京 100035;2.北京市创伤骨科研究所, 北京 100035;3.北京天智航医疗科技股份有限公司天玑实验室, 北京 100192
  • 发布日期:2023-03-24
  • 通讯作者: 张勇. E-mail:zhangyong@tinavi.com
  • 基金资助:
    北京市自然科学基金-海淀原始创新联合基金(L202011);青年北京学者计划(2020-025);北京市医院管理中心登峰人才培养计划(DFL20220401);北京市自然科学基金-海淀原始创新联合基金(L192048)

Preliminary exploration of automatic planning of lumbar pedicle screws based on cone-beam CT in 130 cases

LIU Yajun1,2, YUAN Qiang1, WU Jingye1, HAN Xiaoguang1, LANG Zhao1, ZHANG Yong3   

  1. 1. Department of Spine Surgery, Beijing Jishuitan Hospital, Beijing 100035, China;
    2. Beijing Research Institute of Traumatology and Orthopaedics, Beijing 100035, China;
    3. Tianji Laboratory, Beijing Tinavi Medical Technologies Co., Ltd, Beijing 100192, China
  • Published:2023-03-24

PDF (PC)

71

摘要： 目的 利用深度学习技术开发一种基于锥形束CT(CBCT)影像的椎弓根螺钉自动规划系统,验证和评估其在腰椎椎弓根螺钉规划中的有效性和效率。 方法 回顾性分析2017年1月至2019年12月行单中心机器人辅助腰椎椎弓根螺钉内固定术中所用CBCT引导影像和椎弓根螺钉规划信息患者130例,其中脊椎分割和椎弓根螺钉自动规划训练100例,算法验证30例。通过Dice系数和定性评估来评价分割结果;基于Ravi分级和 Zdichavsky分级评价椎弓根螺钉自动规划结果;基于Babu方法评估椎弓根螺钉上关节突侵犯情况;利用Kappa分析评估观察者间一致性。 结果 10例分割定量评估中,Dice值为0.99。30例分割定性评估中,专家对于分割结果的平均可接受率为94.9%,平均满意率为80.5%。共计自动规划了300枚椎弓根螺钉,298枚(99%)为Ravi 1级,300枚(100%)为Zdichavsky IA级。2枚(1%)为Ravi 2级(突破皮质<2 mm),均为向外侧偏出皮质。评估椎弓根螺钉侵犯关节突的情况,300枚螺钉中,294枚(98%)为0级,共计6枚侵犯上关节突,其中3枚(1%)为1级,2枚(0.67%)为2级,1枚(0.33%)为3级。Kappa分析显示,观察者间具有良好一致性(K=0.71~0.94; 一致率 90%~96%)。运行速度评估时,自动分割完成平均所需时间为3.52 s,特征点提取所需时间为3.81 s,总的平均运行时间为7.33 s。 结论 基于CBCT影像的椎弓根螺钉自动规划系统可实现良好的自动分割和规划效果,对于辅助医生手术规划和改善脊柱外科手术机器人导航流程具有潜在临床应用价值。

关键词: 椎弓根螺钉, 深度学习, 锥形束CT, 自动规划, 图像分割, 脊柱外科手术机器人

Abstract: Objective To develop an automatic pedicle screw planning system based on cone-beam CT(CBCT)images using deep learning technology, and to evaluate its efficiency and planning efficiency. Methods The CBCT and screw planning data of 130 cases who underwent robot-assisted lumbar pedicle screw fixation from a single academic institution during Jan. 2017 and Dec. 2019 were retrospectively analyzed, including 100 cases for network training for spinal segmentation and automatic planning, and 30 cases for validation. The segmentation results were evaluated using Dice coefficient and qualitative validation. The pedicle screw placement was evaluated using the Ravi and Zdichavsky grading systems. The viocation of superior facet joint was evaluated with Babus method. Interobserver reliability was assessed with Kappa statistics. Results In the quantitative assessment of segmentation in 10 cases, the Dice value was 0.99. In the qualitative evaluation of segmentation in 30 cases, the average acceptance rate of the segmentation results by experts was 94.9%, and the average satisfaction rate was 80.5%. In total, the automatic placed 300 pedicle screws between the L1 and L5 spinal levels, including 300(100%)Zdichavsky 1A, 298(99.0%)Ravi 1, and 2(1.0%)Ravi 2(<2 mm breech), both of which were laterally deviated from the cortex. Of the 300 screws, 294(98%)were Grade 0, 6(2%, all L5)of which had superior facet joint violation, including 3(1%)Grade 1, 2(0.67%)Grade 2, and 1(0.33%)Grade 3. Kappa statistics showed excellent overall agreement between raters(K=0.71-0.94; 90%-96% agreement). In running speed evaluation, the average time required for automatic segmentation was 3.52 seconds, the time required for feature point extraction was 3.81 seconds, and the total average running time was 7.33 seconds. Conclusion The automatic planning system of pedicle screws based on CBCT images shows excellent automatic segmentation and screw planning, which has the potential to aid surgeons in screw planning and improve the navigation workflow of spinal surgical robots.

Key words: Pedicle screws, Deep learning, Cone-beam CT, Automatic planning, Image segmentation, Spinal surgical robot

中图分类号: 

  • R681.5
[1] Roy-Camille R, Roy-Camille M, Demeulenaere C. Osteosynthesis of dorsal, lumbar, and lumbosacral spine with metallic plates screwed into vertebral pedicles and articular apophyses[J]. Presse Med, 1970, 78(22): 1447-1448.
[2] Vaccaro AR, Garfin SR. Pedicle screw fixation in the lumbar spine[J]. J Am Acad Orthop Surg, 1995, 3(5): 263-274.
[3] Puvanesarajah V, Liauw JA, Lo SF, et al. Techniques and accuracy of thoracolumbar pedicle screw placement[J]. World J Orthop, 2014, 5(2): 112-123.
[4] Mason A, Paulsen R, Babuska JM, et al. The accuracy of pedicle screw placement using intraoperative image guidance systems[J]. J Neurosurg Spine, 2014, 20(2): 196-203.
[5] DSouza M, Gendreau J, Feng A, et al. Robotic-assisted spine surgery: History, efficacy, cost, and future trends[J]. Robot Surg, 2019, 6: 9-23.
[6] Han X, Tian W, Liu Y, et al. Safety and accuracy of robot-assisted versus fluoroscopy-assisted pedicle screw insertion in thoracolumbar spinal surgery: a prospective randomized controlled trial[J]. J Neurosurg Spine, 2019:1-8. doi:10.3171/2018.10.spine18487.
[7] Huntsman KT, Riggleman JR, Ahrendtsen LA, et al. Navigated robot-guided pedicle screws placed successfully in single-position lateral lumbar interbody fusion[J]. J Robot Surg, 2020, 14(4): 643-647.
[8] Mao G, Gigliotti MJ, Myers D, et al. Single-surgeon direct comparison of O-arm neuronavigation versus mazor X robotic-guided posterior spinal instrumentation[J]. World Neurosurg, 2020, 137: e278-e285. doi:10.1016/j.wneu.2020.01.175.
[9] Nolte LP, Slomczykowski MA, Berlemann U, et al. A new approach to computer-aided spine surgery: fluoroscopy-based surgical navigation[J]. Eur Spine J, 2000, 9(Suppl 1): S78-S88.
[10] Liu YJ, Tian W, Liu B, et al. Accuracy of CT-based navigation of pedicle screws implantation in the cervical spine compared with X-ray fluoroscopy technique[J]. Chinese Journal of Surgery, 2005, 43(20): 1328-1330.
[11] Tkatschenko D, Kendlbacher P, Czabanka M, et al. Navigated percutaneous versus open pedicle screw implantation using intraoperative CT and robotic cone-beam CT imaging[J]. Eur Spine J, 2020, 29(4): 803-812.
[12] Kendlbacher P, Tkatschenko D, Czabanka M, et al. Workflow and performance of intraoperative CT, cone-beam CT, and robotic cone-beam CT for spinal navigation in 503 consecutive patients[J]. Neurosurg Focus, 2022, 52(1): E7.
[13] Siddiqui MI, Wallace DJ, Salazar LM, et al. Robot-assisted pedicle screw placement: learning curve experience[J]. World Neurosurg, 2019, 130: e417-e422.
[14] Klinder T, Ostermann J, Ehm M, et al. Automated model-based vertebra detection, identification, and segmentation in CT images[J]. Med Image Anal, 2009, 13(3): 471-482.
[15] Korez R, Ibragimov B, Likar B, et al. A framework for automated spine and vertebrae interpolation-based detection and model-based segmentation[J]. IEEE Trans Med Imaging, 2015, 34(8): 1649-1662.
[16] Knez D, Likar B, Pernus F, et al. Computer-assisted screw size and insertion trajectory planning for pedicle screw placement surgery[J]. IEEE Trans Med Imaging, 2016, 35(6): 1420-1430.
[17] Yao JH, Burns JE, Getty S, et al. Automated extraction of anatomic landmarks on vertebrae based on anatomic knowledge and geometrical constraints[J]. 2014 IEEE 11th Int Symp Biomed Imaging ISBI 2014, 2014: 397-400.
[18] Pereañez M, Lekadir K, Castro-Mateos I, et al. Accurate segmentation of vertebral bodies and processes using statistical shape decomposition and conditional models[J]. IEEE Trans Med Imaging, 2015, 34(8): 1627-1639.
[19] Siemionow K, Luciano C, Forsthoefel C, et al. Autonomous image segmentation and identification of anatomical landmarks from lumbar spine intraoperative computed tomography scans using machine learning: a validation study[J]. J Craniovertebr Junction Spine, 2020, 11(2): 99-103.
[20] Scherer M, Kausch L, Ishak B, et al. Development and validation of an automated planning tool for navigated lumbosacral pedicle screws using a convolutional neural network[J]. Spine J, 2022, 22(10): 1666-1676.
[21] Siemionow KB, Forsthoefel CW, Foy MP, et al. Autonomous lumbar spine pedicle screw planning using machine learning: a validation study[J]. J Craniovertebr Junction Spine, 2021, 12(3): 223-227.
[22] Lyu YY, Liao HF, Zhu HQ, et al. A3DSegNet: Anatomy-aware artifact disentanglement and segmentation network for unpaired segmentation, artifact reduction, and modality translation[EB/OL].(2020)[2022-10-15]. arXiv: 2001.00339. https://arxiv.org/abs/2001.00339.
[23] Zou KH, Warfield SK, Bharatha A, et al. Statistical validation of image segmentation quality based on a spatial overlap index[J]. Acad Radiol, 2004, 11(2): 178-189.
[24] Ravi B, Zahrai A, Rampersaud R. Clinical accuracy of computer-assisted two-dimensional fluoroscopy for the percutaneous placement of lumbosacral pedicle screws[J]. Spine(Phila Pa 1976), 2011, 36(1): 84-91.
[25] Zdichavsky M, Blauth M, Knop C, et al. Accuracy of pedicle screw placement in thoracic spine fractures: part I: inter-and intraobserver reliability of the scoring system[J]. Eur J Trauma. 2004(30): 234-240.
[26] Babu R, Park JG, Mehta AI, et al. Comparison of superior-level facet joint violations during open and percutaneous pedicle screw placement[J]. Neurosurgery, 2012, 71(5): 962-970.
[27] 刘亚军, 韩晓光, 田伟. 我国医用机器人的研究现状及展望[J]. 骨科临床与研究杂志, 2018, 3(4): 193-194.
[28] Karandikar P, Massaad E, Hadzipasic M, et al. Machine learning applications of surgical imaging for the diagnosis and treatment of spine disorders: current state of the art[J]. Neurosurgery, 2022, 90(4): 372-382.
[29] Watanabe K, Aoki Y, Matsumoto M. An application of artificial intelligence to diagnostic imaging of spine disease: estimating spinal alignment from moiré images[J]. Neurospine, 2019, 16(4): 697-702.
[30] Pan Y, Chen Q, Chen T, et al. Evaluation of a computer-aided method for measuring the Cobb angle on chest X-rays[J]. Eur Spine J, 2019, 28(12): 3035-3043.
[31] Lu JT, Pedemonte S, Bizzo B, et al. DeepSPINE: automated lumbar vertebral segmentation, disc-level designation, and spinal Stenosis grading using deep learning[EB/OL].(2018)[2022-10-15]. arXiv: 1807.10215. http://arxiv.org/abs/1807.10215.
[32] Ravindra VM, Senglaub SS, Rattani A, et al. Degenerative lumbar spine disease: estimating global incidence and worldwide volume[J]. Global Spine J, 2018, 8(8): 784-794.
[33] Cho W, Cho SK, Wu C. The biomechanics of pedicle screw-based instrumentation[J]. J Bone Joint Surg Br, 2010, 92(8): 1061-1065.
[34] Jarvers JS, Schleifenbaum S, Pfeifle C, et al. Comparison of three different screw trajectories in osteoporotic vertebrae: a biomechanical investigation[J]. BMC Musculoskelet Disord, 2021, 22(1): 418.
[35] Weidling M, Heilemann M, Schoenfelder S, et al. Influence of thread design on anchorage of pedicle screws in cancellous bone: an experimental and analytical analysis[J]. Sci Rep, 2022, 12(1): 8051.
[36] Scorza D, El Hadji S, Cortés C, et al. Surgical planning assistance in keyhole and percutaneous surgery: a systematic review[J]. Med Image Anal, 2021, 67: 101820. doi:10.1016/j.media.202.101820.
[37] Wicker R, Tedla B. Automatic determination of pedicle screw size, length, and trajectory from patient data[J]. Conf Proc IEEE Eng Med Biol Soc, 2004, 2004: 1487-1490. doi:10.1109/iembs.2004.1403457.
[38] Goerres J, Uneri A, De Silva T, et al. Spinal pedicle screw planning using deformable atlas registration[J]. Phys Med Biol, 2017, 62(7): 2871-2891.
[39] Vijayan R, De Silva T, Han R, et al. Automatic pedicle screw planning using atlas-based registration of anatomy and reference trajectories[J]. Phys Med Biol, 2019, 64(16): 165020.
[40] Caprara S, Fasser MR, Spirig JM, et al. Bone density optimized pedicle screw instrumentation improves screw pull-out force in lumbar vertebrae[J]. Comput Methods Biomech Biomed Eng, 2022, 25(4): 464-474.
[41] Esfandiari H, Newell R, Anglin C, et al. A deep learning framework for segmentation and pose estimation of pedicle screw implants based on C-arm fluoroscopy[J]. Int J Comput Assist Radiol Surg, 2018, 13(8): 1269-1282.
[42] Burström G, Buerger C, Hoppenbrouwers J, et al. Machine learning for automated 3-dimensional segmentation of the spine and suggested placement of pedicle screws based on intraoperative cone-beam computer tomography[J]. J Neurosurg Spine, 2019, 31(1): 147-154.
[43] Kausch L, Thomas S, Scherer M, et al. Automatic image-based pedicle screw planning[C] , Cristian A, Linte, Jeffrey H, et al. Medical Imaging 2021: Image-Guided Procedures, Robotic Interventions, and Modeling. 2021.
[44] Ma C, Zou D, Qi H, et al. A novel surgical planning system using an AI model to optimize planning of pedicle screw trajectories with highest bone mineral density and strongest pull-out force[J]. Neurosurg Focus, 2022, 52(4): E10.
[45] Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation[C] //Navab N, Hornegger J, Wells W, et al. Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015. Lecture Notes in Computer Science. Berlin: Springer, 2015.
[46] Proietti L, Scaramuzzo L, Schirò GR, et al. Degenerative facet joint changes in lumbar percutaneous pedicle screw fixation without fusion[J]. Orthop Traumatol Surg Res, 2015, 101(3): 375-379.
[47] Varoquaux G, Cheplygina V. Machine learning for medical imaging: methodological failures and recommendations for the future[J]. NPJ Digit Med, 2022, 5(1): 48.
[48] Caprara S, Fasser MR, Spirig JM, et al. Bone density optimized pedicle screw instrumentation improves screw pull-out force in lumbar vertebrae[J]. Comput Methods Biomech Biomed Engin, 2022, 25(4): 464-474.
[49] Santoni BG, Hynes RA, McGilvray KC, et al. Cortical bone trajectory for lumbar pedicle screws[J]. Spine J, 2009, 9(5): 366-373.
[50] Jang JS, Lee SH, Lim SR. Guide device for percutaneous placement of translaminar facet screws after anterior lumbar interbody fusion. Technical note[J]. J Neurosurg, 2003, 98(1 suppl): 100-103.
[51] Zhuang XM, Yu BS, Zheng ZM, et al. Effect of the degree of osteoporosis on the biomechanical anchoring strength of the sacral pedicle screws: an in vitro comparison between unaugmented bicortical screws and polymethylmethacrylate augmented unicortical screws[J]. Spine(Phila Pa 1976), 2010, 35(19): E925-E931.
[52] Klinder T, Ostermann J, Ehm M, et al. Automated model-based vertebra detection, identification, and segmentation in CT images[J]. Med Image Anal, 2009, 13(3): 471-482.
[53] Di Angelo L, Di Stefano P, Guardiani E. An automatic method for feature segmentation of human thoracic and lumbar vertebrae[J]. Comput Methods Programs Biomed, 2021, 210: 106360. doi:10.1016/j.cmpb.2021.106360.
[54] Cheng P, Cao X, Yang Y, et al. Automatically recognize and segment morphological features of the 3D vertebra based on topological data analysis[J]. Comput Biol Med, 2022, 149: 106031. doi:10.1016/j.compbiomed.2022.106031.
[55] Dexu Wang, Zhikai Yang, Ziyan Huang, et al. Spine Segmentation with Multi-view GCN and Boundary Constraint[J]. Annu Int Conf IEEE Eng Med Biol Soc, 2022: 2136-2139.
[56] Augustus J Rush, Nicholas Shepard, Michael Nolte, et al. Augmented Reality in Spine Surgery: Current State of the Art[J]. Int J Spine Surg, 2022, 16(S2): S22-S27.
[1] 冯世庆. 计算机视觉与腰椎退行性疾病[J]. 山东大学学报 (医学版), 2023, 61(3): 1-6.
[2] 王煜,尹增正,聂晓琨,马志德,熊世江. 1 125颗下颌第一恒磨牙近中中央根管的CBCT检出率及相关因素分析[J]. 山东大学学报 (医学版), 2022, 60(3): 100-109.
[3] 王琳琳,孙玉萍. 从临床医生角度,看人工智能在癌症精准诊疗中的应用及思考[J]. 山东大学学报 (医学版), 2021, 59(9): 89-96.
[4] 尹增正,聂晓琨,王煜,马志德,熊世江. 800颗上颌第一前磨牙颊根腭侧面沟和根管的CBCT影像形态学观察[J]. 山东大学学报 (医学版), 2021, 59(8): 74-79.
[5] 刘学业,李齐明,唐弘毅,徐秋平,陈文倩,郭泾. 年轻成人颞下颌关节髁突体积、表面积与关节盘矢向位置的关系[J]. 山东大学学报 (医学版), 2021, 59(6): 117-121.
[6] 刘琚,吴强,于璐跃,林枫茗. 基于深度学习的脑肿瘤图像分割[J]. 山东大学学报 (医学版), 2020, 1(8): 42-49, 73.
[7] 林浩添,李龙辉,陈睛晶. 儿童眼病的人工智能研究进展[J]. 山东大学学报 (医学版), 2020, 58(11): 11-16.
[8] 曲毅,张焕开,宋先,初宝睿. 人工智能诊断系统在视网膜疾病的研究进展[J]. 山东大学学报 (医学版), 2020, 58(11): 39-44.
[9] CheungCarol Y.,冉安然. 青光眼影像人工智能深度学习研究现状与展望[J]. 山东大学学报 (医学版), 2020, 58(11): 24-32, 38.
[10] 聂晓琨,郝新宇,张淑存,葛堂娜,熊世江. 下颌第一恒磨牙远舌根的锥形束CT研究[J]. 山东大学学报 (医学版), 2019, 57(3): 85-90.
[11] 沈倍勇,王晓飞,李军心,周琦. 基于灰度值差的全冠修复体CBCT图像伪影评价方法[J]. 山东大学学报 (医学版), 2019, 57(12): 52-56.
[12] 郝新宇,李建华,张静,聂晓琨,熊世江. 上颌第一磨牙根管解剖结构的锥形束CT研究[J]. 山东大学学报 (医学版), 2018, 56(4): 87-91.
[13] 张静,杨兴华,李建华,葛堂娜,张淑存,熊世江. 下颌后牙C形根管的锥形束CT研究[J]. 山东大学学报 (医学版), 2018, 56(2): 56-61.
[14] 张冰,贾凌璐,贾婷婷,张云鹏,文勇,徐欣. 上颌窦外侧壁牙槽管解剖的锥形束CT研究[J]. 山东大学学报（医学版）, 2016, 54(5): 92-96.
[15] 李德水, 刘盼盼, 史晓昕, 郭泾. 正畸美学区的CBCT测量分析[J]. 山东大学学报（医学版）, 2015, 53(2): 75-80.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2018 《山东大学学报(医学版)》编辑部　
地址:山东大学科技期刊社（济南市历城区山大南路27号山东大学中心校区明德楼B座721室） 电话: 0531-88366918 E-mail: xbyxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发