Twisting compensation of Optical Multicore Fiber Shape Sensors for Flexible Medical Instruments

Proceedings of SPIE(2020)

引用 6|浏览2
暂无评分
摘要
Shape sensing has become an area of great interest for many medical applications, such as epidural administration, colonoscopy, biopsies, and cardiac procedures, where real-time data of a dynamic object is required and visual contact is absent. Fiber Optic Shape Sensors (FOSS) consist of optical multi-fiber cables or Multicore Fibers (MCF) with embedded strain sensors, which can reconstruct the sensor shape from its multidimensional bending. Regrettably, the accuracy of three-dimensional shape sensing is remarkably restricted because of twisting, which makes impossible to correctly detect the bending direction. This paper reports an experimental study aimed at investigating the accuracy of optical shape sensors based on spun multicore fibers in sensing twisting, employing one of the most used multicore fiber geometry for sensing applications, the seven-core fiber. Firstly, a theoretical approach to model the mechanical behavior of multicore fiber was developed. Secondly, a pre-twisted fiber optic shape sensor was fabricated in the Institute for Telecommunications and Multimedia Applications (iTEAM), by inscribing four Fiber Bragg Gratings (FBG) in a Spun Multicore Fiber (diameter of 125.1 mu m) with a pre-twisting of 64.9 rotation/meter, manufactured and provided by FIBERCORE. To conclude, a series of experiments were performed to corroborate the theoretical approach and evaluate the sensor performance. The proposed Spun-MCF-based Shape Sensor was able to sense twisting with a sensitivity of 0.23 pm/degrees and accuracy of 4.81 degrees within a wide dynamic range of +/- 270 degrees, maintaining a perfectly elastic behavior at high level of twisting deformation.
更多
查看译文
关键词
Fiber Optic Sensor,Spun Multicore Fiber,Twisted Multicore Fiber,Optical Shape Sensing,Twisting Sensing,Distributed Sensing,Fiber Bragg Grating,Fiber Optic Shape Sensor
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要