Shape and Flow Sensing in Arterial Image Guidance From UV Exposed Optical Fibers Based on Spatio-Temporal Networks

Minimally invasive revascularization procedures such as percutaneous transluminal angioplasty seek to treat occlusions in peripheral arteries. However their ability to treat long occlusions are hampered by difficulties to monitor the location of intravascular devices such as guidewires using fluoroscopy which requires continuous radiation, and lack the capacity to measure physiological characteristics such as laminar blood flow close to occlusions. Fiber optic technologies provide means of tracking by measuring fibers under strain, however they are limited to known geometrical models and are not used to measure external variations.

We present a navigation framework based on optical frequency domain reflectometry (OFDR) using fully-distributed optical sensor gratings enhanced with ultraviolet exposure to track the three-dimensional shape and surrounding blood flow of intra-vascular guidewires. To process the strain information provided by the continuous gratings, a dual-branch model learning spatio-temporal features allows to predict the output measures based on scattered wavelength distributions. The first network determines the 3D shape appearance of the guidewire using the input backscattered wavelength shift data in combination with prior segmentations, while a second network (graph temporal convolution network) produces estimates of vascular flow velocities using ground-truth 4D-flow MRI acquisitions.

Experiments performed on synthetic and animal models, as well as in a preliminary human trial shows the capability of the model to generate accurate 3D shape tracking and blood flow velocities differences below 2 cm/s, thus providing realistic physiologic and anatomical properties for intravascular techniques.

The study demonstrates the feasibility of using the device clinically, and could be integrated within revascularization workflows for treating occlusions in arteries, since the navigation framework involves minimal manual intervention.