近红外光学导航系统模拟实时CT导航对肝内小病灶穿刺的研究

发布时间：2018-01-25 15:28

  本文关键词： 近红外 光学导航 计算机断层扫描成像 精度 穿刺　出处：《南方医科大学》2017年博士论文　论文类型：学位论文

【摘要】：研究目的:1.对比迭代模型重建(iterative model reconstruction,IMR)技术与滤波反投影(filteredbackprojection,FBP)技术在薄层(1mm)CT平扫图像上对肝内小转移瘤(直径≤1Omm)的检出效能差异,探索IMR技术是否具有为肝脏小病灶穿刺提供更好的穿刺背景图像的潜力。2.评估一种新型近红外光学导航系统与CT图像融合在理想化穿刺模型条件下对静态模型内小靶点引导穿刺的准确性及可行性,并在试验进程中对出现的或可能出现的关键性问题找到相应的解决策略。3.探讨近红外导航技术与CT图像融合对模拟屏气状态及自由呼吸状态下离体猪肝模型内小靶点穿刺的准确性及可行性,解决静态及动态条件下穿刺过程中如何进行准确定位并引导穿刺等关键性问题。4.探讨近红外导航技术与CT图像融合对模拟屏气状态及自由呼吸状态下活体小型猪肝脏内小靶点穿刺的准确性及可行性,发现并尝试解决活体动物穿刺过程中可能出现的相关问题。第一部分迭代模型重建(IMR)与滤波反投影(FBP)技术在薄层CT平扫图像上对肝脏小转移瘤成像对比研究1方法常规CT扫描89例肝转移瘤患者,重建为层厚1mm的FBP及IMR图像。病灶总数(total number,TN):增强IMR图像上肝内小病灶(≤10mm)的检出量;病灶检出数(detected number,DN):1mm薄层CT平扫图像上小病灶检出量;检出率(detected rate,DR)= DN/TN;以5分法评估图像质量;计算门静脉及下腔静脉对比噪声比(contrast-to-noise ratio,CNR)。2结果TN 值为 239 个,IMR 组、FBP 组 DR 值分别为 43.93%(105/239)、17.57%(42/239)(p0.05);主观评分分别为 4.17±0.51、3.28±0.54(p0.05)。IMR组门静脉及下腔静脉CNR值均高于FBP组(p均0.05)。3结论IMR技术较FBP技术明显提高薄层CT平扫图像上肝内小转移瘤检出率及主要血管分支显示质量,研究结果证实IMR技术较FBP技术能为部分肝内小病灶穿刺提供更好的薄层CT平扫图像作为背景参考图像。第二部分近红外光学导航系统对琼脂模型内小靶点穿刺精度及可行性评估1方法1.1分组常规组与导航组,60靶点/组。1.2设备256层螺旋CT,管电压/管电流:120kv/300mAs;矩阵:512×512;扫描野:350mm×350mm;层厚/层距:1mm/0.5mm。近红外光学导航系统由合作单位提供。1.3穿刺方法常规组(1)CT扫描模型,确定靶点位置。(2)定位标记贴于模型表面,再次CT扫描,据定位工具进针。(3)CT扫描评估结果。导航组(1)CT扫描附带表面标记点的模型;(2)匹配标记点图像位置及空间位置;(3)注册融合活检针/模型的位置信息;(4)在导航系统图像上进针接触靶点。(5)CT扫描评估结果。1.4数据靶点定位误差(TPE):针尖与靶点距离,≤10mm穿刺成功;侧向误差(LE):穿刺针延长线与靶点最短距离;角度误差(AE):穿刺针实际路径与计划路径的夹角;深度误差(DE):实际路径与计划路径的深度差异绝对值。首次成功率(FSR):首次TPE值≤10mm的靶点数与总靶点的比值。1.5统计组间FSR值差异用卡方检验评估,余用t检验评估,p0.05认为差异有统计学意义。2结果常规/导航组TPE值分别为12.19±4.19mm、3.18±1.70mm,LE值分别为6.80±2.86mm、2.28±1.49mm,AE值分别为 4.91±1.50°、2.96±1.29°,DE 值分别为 2.66±1.07mm、1.37±0.99mm(p 均0.05)。常规/导航组 FSR 分别为 35%、100%(p0.05)。3结论在理想化模型条件下,该近红外光学导航系统的穿刺精度及临床可行性参数均优于常规组,研究结果证明该光学导航的穿刺精度及临床可行性参数在静态刚性模型/器官的临床穿刺应用方面具有较大潜力,但对于静态或动态非刚性模型/器官的穿刺应用仍然需要进一步研究。第三部分近红外光学导航系统对屏气及呼吸模型内靶点穿刺的关键问题研究1材料与方法1.1分组屏气模型:常规/导航组,45靶点/组。呼吸模型:常规/导航组,45靶点/组。1.2设备硬件同第二部分。用标记点注册误差(fiducial registration error,FRE)曲线监控标记点实时位置误差,曲线波谷区术前与术中标记点差异最小。2结果屏气模型:常规组/导航组TPE值分别为10.69±3.38mm、5.61±3.23mm(p0.05),LE分别为8.31±3.66 mm、3.60±2.22 mm(p0.05),AE分别为6.16±2.84°、3.57±2.38°(p0.05),DE 分别为 3.3±1.28mm、4.13±1.94mm(p0.05),FSR 分别为 33%(15/45)、87%(39/45)(p0.05)。呼吸模型：常规组/导航组TPE值分别为13.47±3.50mm、9.17±3.77mm(p0.05);LE 分别为 11.56±3.45mm、8.13±3.52mm(p0.05);AE 分别为 7.87±4.05、6.38±3.61(p0.05);DE 分别为 4.26±2.20mm、3.75±1.61mm(p0.05)。FSR 分别约 11%(5/45)、62%(39/45)(p0.05)。3 结论导航组对运动与静止肝脏模型的穿刺TPE、LE、FSR值均优于常规组，证明该光学导航系统同时具有提升静止及运动的非刚性模型/器官内靶点穿刺精度的价值。第四部分近红外光学导航系统对屏气及呼吸动物肝内靶点的穿刺应用研究1 材料与方法1.1 分组屏气动物：常规组与导航组，，45靶点/组。呼吸动物：常规组与导航组，45靶点/组。1.2 设备1.3 扫描及穿刺方法 同第三部分。1.4 数据收集1.5 统计2 结果屏气动物：常规/导航组TPE值分别为：13.86±3.45mm、5.22±3.02mm(p0.05);LE 分别为 9.39±3.20mm、4.01±2.43mm(p0.05);DE 分别为1.85±1.09mm、3.90±2.53mm(p0.05);AE 分别为 5.33±2.48°、4.37±2.82°(p0.05);FSR 分别为 16%(7/45)、91%(41/45)(p0.05)。呼吸动物:常规/导航组TPE值分别为16.35±3.07mm、7.42±3.56mm(p0.05),LE 分别为 10.49±4.62mm、5.56±3.36mm(p0.05),AE 分别为 7.25±4.29°、5.15±3.33°(p0.05),DE 值分别约2.89±1.56mm、3.90±3.28mm(p0.05),FSR值分别为 9%(4/45)、73%(33/45)(p0.05)。3结论导航组在运动与静止动物的穿刺TPE、LE、FSR值均优于常规组,证明该导航系统能够提升活体屏气或呼吸动物肝内小靶点的穿刺精度,但对于呼吸动物穿刺操作仍较困难。
[Abstract]:Objective: 1. contrast iterated model reconstruction (iterative model reconstruction, IMR) and filtered backprojection (filteredbackprojection, FBP) technology in thin (1mm) CT scan images of small hepatic metastases (less than 1Omm) detection performance differences, explore whether IMR technology has provided the background image better puncture the potential.2. evaluated a novel near infrared optical navigation system and CT image fusion model in the ideal condition of puncture under the accuracy and feasibility of small target guided static model of small lesions of liver biopsy, and in the process of key problems or possible to find the corresponding solution strategy of nearly.3. infrared navigation technology and CT image fusion for simulation and breath holding state free breathing from the accuracy and feasibility of the model of small target body liver puncture, solve the static and dynamic. How to accurately locate the key problems and guided.4. to investigate the accuracy and feasibility of fusion simulation and breath holding state free breathing in pig liver small target puncture near infrared navigation technology and CT image under the puncture process, and attempt to resolve related problems that may arise in the process of living animal puncture the first part. Iterative model reconstruction (IMR) and filter back projection (FBP) technique research on CT images of small liver metastases compared to 1 conventional methods of CT scan imaging of 89 cases of liver metastases in patients with thin CT flat, FBP and IMR image reconstruction thickness 1mm. The total number of lesions (total number, TN). IMR image enhancement on intrahepatic small lesion (10mm) detection; number of lesions (detected number, DN): 1mm CT scan images of small lesions detected; the detection rate (detected = DN/TN; rate, DR) with 5 points method to evaluate image quality The amount of calculation of the portal vein and inferior vena cava; contrast to noise ratio (contrast-to-noise ratio, CNR.2) the TN value was 239, IMR group, FBP group, DR = 43.93% (105/239), 17.57% (42/239) (P0.05); subjective scores were 4.17 + 0.51,3.28 + 0.54 (P0.05).IMR group and portal vein inferior vena cava CNR values were higher than that of group FBP (P 0.05) conclusion.3 technology IMR is FBP technology significantly improved the CT scan images of small hepatic metastasis detection rate and main branch vessels display quality, the results proved that IMR technology is FBP technology for small lesions in the liver puncture part provide thin flat CT better the scan image as the reference image. The second part of the near infrared optical navigation system evaluation method and the conventional group 1.1 group 1 group on the agar model small navigation target puncture accuracy and feasibility of 60 target group /.1.2 equipment of 256 slice spiral CT, tube voltage / current tube: 120kv/300mAs; 鐭╅樀:512脳512;鎵�鎻忛�

本文编号：1463126