-
应变灵敏度标定系统如图7所示,主要由两台手动位移平台组成,两台位移平台间距230 mm,位移平台采用微纳光科公司的WN103TM13H位移台,行程13 mm,最小刻度10 µm。光纤传感形状测量系统如图8所示,主要包括运动控制系统、光纤传感系统、探针模型和计算机等。通过固定探针一端,利用运动控制系统使探针实现不同方向、不同偏移量的弯曲。运动控制系统主要由运动控制器和电动多维位移台组成:运动控制器采用微纳光科仪器公司WNMC400控制器,多维位移台采用微纳光科公司WN303ZA位移台。光纤传感系统(见图8)主要由ASE光源、光谱仪、耦合器等组成。ASE光源选用接口类型为FC/APC的Lightpromotech M1043-1,其输出光谱为1529~1605 nm,功率为13 dBm,光平坦度小于2 dB。光谱仪采用波长为1200~2400 nm的YOKOGAWA AQ6375,其功率为−70~+20 dBm,快速测量时间为0.2 s,跨度为100 nm。光纤传感系统的反射信号被传输到光谱仪中并被转换成数字信号。利用上位机中Matlab软件对信号进行寻峰处理和差值运算,利用形状重构算法完成对探针形状参数的计算。
-
首先对光纤封装前后的状态进行对比。封装前,从光栅点1到光栅点4,FBG的波长逐渐变大。实验测得三根光纤上的十二个光栅点在植入探针前后中心波长如表1所示。分析表1可以看出,使用3M公司DP2216高性能粘合剂对光纤进行封装对其初始波长并无明显影响,其中每个测量点数据测量5次,每次测量光谱呈现相同的变化规律,最大偏差不超过0.0196 nm,具有较好的一致性。其次对三根无封装光纤光栅串传感器进行标定。将裸光纤光栅夹在标定台上后,对光纤光栅的拉伸基本上是均匀的,而对光纤光栅的压缩大多数情况下并不均匀。光纤光栅栅区长度本身有一定的尺寸,进行压缩时,存在栅距不均匀、光谱质量变差的问题,会增大解调难度,降低应变测量精度。在文中的应用环境中,光纤光栅都是工作在预拉状态,因此,只需要测量出无封装FBG传感器拉应变的应变灵敏度系数就能满足应用需求。
Fiber FBG Wavelength
before embedded/nmWavelength
after embedded/nma 1 1525.8163 1525.7967 2 1529.6933 1529.6742 3 1533.7671 1533.7475 4 1538.1146 1538.1146 b 1 1525.7575 1525.7575 2 1529.8308 1529.8308 3 1533.9042 1533.9042 4 1537.8208 1537.8013 c 1 1525.9753 1525.9925 2 1529.9092 1529.9092 3 1533.7671 1533.7867 4 1537.8600 1537.8600 Table 1. Wavlength shift of different FBGs before and after integrating needle
实验过程中,先测量光纤光栅在自由状态下的中心波长,然后将光纤光栅安装到标定板上,每拉伸460 µm (相当于200 µε)记录一组数据,直至拉伸6900 µm (3000 µm),重复实验5次。同样的,将光纤光栅封装到探针后重复上述实验,测得光纤光栅的应变灵敏度如表2,封装后的应变灵敏度如图9所示。
Fiber FBG FBG strain sensitivity/
$\mathrm{p}\mathrm{m}\cdot {\mathrm{{\text{µ}} }\mathrm{\varepsilon } }^{-1}$FBG strain sensitivity
after embedded/
$\mathrm{p}\mathrm{m}\cdot {\mathrm{{\text{µ}} }\mathrm{\varepsilon } }^{-1}$a 1 1.11 0.44 2 1.12 0.79 3 1.12 0.84 4 1.12 0.84 b 1 1.13 0.56 2 1.15 0.81 3 1.14 0.83 4 1.15 0.84 c 1 1.13 0.38 2 1.14 0.77 3 1.14 0.79 4 1.15 0.78 Table 2. Strain sensitivity of FBGs
-
文中实验通过将介入手术探针尖端固定,利用运动控制系统控制探针在X轴上进行移动。当介入手术探针的末端位置偏移量变化范围是0~−15 mm时,实验测得光纤光栅的光谱变化如图10所示。用寻峰算法计算得到每个FBG中心波长的漂移量,实验分析得出,当介入手术探针的末端位置偏移量(
$ \mathrm{\Delta }x $ )由0变化到−15 mm时,光纤a和光纤b的每个FBG光谱的峰值逐渐向波长减小的方向移动,光纤c的峰值逐渐向波长增大的方向移动。利用FBG中心波长漂移量和插值算法可以建立介入探针变形测量的数据库,即不同的偏移量与FBG的不同中心波长漂移量呈现一一对应的关系。分析图10可以得出不同光纤的中心波长漂移量的绝对值随着介入手术探针末端偏移量的增加而增大,并且呈现线性关系。分析图10可得,位于探针中部的FBG所承受的剪切力最大,FBG中心波长漂移对末端位置的响应程度与传感位置有着密切关系。并且由于三根光纤呈120°分布,探针弯曲时不同位置的光纤受到的拉伸、压缩作用不同。我们设定拉伸状态是正应变,压缩状态是负应变,因此光纤a、光纤b上受到压缩作用,FBG中心波长漂移和末端位置的关系为单调递增,光纤c收到拉伸作用,FBG中心波长漂移和末端位置的关系为单调递减。
加入应变灵敏度矩阵前后的测量结果如图11所示。把使用应变灵敏度理论值的测量结果与加入实测应变灵敏度矩阵的测量结果进行比较,比较结果如表3所示。分析表格得出,随着末端偏移量的增加,形状测量的误差也在逐渐递增。文中使用绝对误差
${{r}}_{e}(k)$ 、相对误差${{r}}_{em}(k)$ α对视觉测量系统和光纤传感系统进行性能评估,计算公式如下所示:式中:
$ \Upsilon (k)$ 为形状测量系统测量结果;$ \gamma \left(k\right) $ 为末端偏移量理论值。在不同弯曲状态下,加入应变灵敏度矩阵的介入手术探针末端偏移量的平均误差为0.415 mm,最大误差为0.601 mm,最大误差百分比为4.01%。相对于没有进行传感器应变灵敏度标定的系统,利用文中提出的方法测量介入手术探针末端偏移量与算法改进前相比。平均误差降低了1.385 mm,最大误差降低了3.317 mm,说明该方法可有效提高测量精度。Deformation theoretical value/mm −3 −6 −9 −12 −15 Shape reconstruction without sensitivity matrix/mm −2.181 −5.920 −10.684 −14.498 −18.918 Absolute error/mm 0.819 0.08 1.684 2.498 3.918 Relative error 5.46% 0.53% 11.23% 16.65% 26.12% Shape reconstruction with sensitivity matrix/mm −2.896 −5.667 −8.538 −11.425 −14.399 Absolute error/mm 0.104 0.333 0.462 0.575 0.601 Relative error 0.69% 2.22% 3.08% 3.83% 4.01% Table 3. Deformation measured error analysis of X-axis
Needle shape optical fiber measurement method introducing strain sensitivity matrix
doi: 10.3788/IRLA20210623
- Received Date: 2021-08-30
- Rev Recd Date: 2021-10-08
- Publish Date: 2021-12-31
-
Key words:
- fiber Bragg grating /
- strain sensitivity /
- shape sensing /
- interventional needle
Abstract: Interventional needle shape monitoring method can provide doctors with important information during surgery and is a necessary means to ensure the safety of surgery. In order to improve interventional needle shape measurement accuracy, a fiber grating sensor (FBG) array was presented, which introduced a strain sensitivity matrix to measure the shape of the probe. Firstly, based on the theory of FBG, the relationship between the wavelength shift of the fiber grating implanted in the probe and its strain was analyzed, and the strain sensitivity matrix was introduced, the relationship between the FBG center wavelength shift and its bending curvature was studied. Then the geometric parameter relations and coordinate transformation equations of the needle's local elements were deduced, and the needle shape reconstruction model based on fiber grating was established. Finally, in order to verify the influence of introducing the strain sensitivity matrix on the shape measurement accuracy, the shape measurement experiment of the intervention needle implanted with the FBGs array under different bending states was carried out, and the error of the shape measurement before and after the strain sensitivity matrix was compared and analyzed. The experimental results show that the introduction of the FBG’s strain sensitivity matrix can effectively improve the measurement accuracy of the needle. Under different bending conditions, the average error of the end of the interventional surgical needle is reduced by 1.385 mm, and the maximum error is reduced by 3.317 mm. The shape measurement method based on FBGs array proposed in this paper has broad application prospects in the direction of shape measurement of flexible medical instruments in interventional surgery.