Automating Vascular Shunt Insertion with the dVRK Robotic Surgical Assistant

(Under Review)

(Coming Soon)

Abstract

Vascular shunt insertion is a fundamental surgical procedure used to temporarily restore blood flow to tissues. It is often performed in the field after major trauma. We formulate a problem of automated vascular shunt insertion and propose a pipeline to perform Automated Vascular Shunt Insertion (AVSI) using a da Vinci Research Kit. The pipeline uses a learned visual model to estimate the locus of the vessel rim, plans a grasp on the rim, and moves to grasp at that point. The first robot gripper then pulls the rim to stretch open the vessel with a dilation motion. The second robot gripper then proceeds to insert a shunt into the vessel phantom (a model of the blood vessel) with a chamfer tilt followed by a screw motion. Results suggest that AVSI achieves a high success rate even with tight tolerances and varying vessel orientations up to 30 degrees.

Overview

Our insertion pipeline is generally broken down into three phases: detection, grasping, and insertion.

A: Detection. Given a stereo RGB input from the camera, the pipeline uses an asymmetrical U-Net trained with LUV to produce a segmentation mask of the predicted location of the vessel rim.

B: Grasping. Using the predicted mask, we fit a circle to the rim of the vessel phantom in 3D space. The robot then computes an maximally distant grasp point from the two fixed points on the rim, moves the gripper to the rim, and pulls the rim outward to dilate it.

C: Insertion. After the dilation, the second robot gripper inserts a shunt into the vessel phantom using a chamfer tilt insertion, and secures it with a screw motion.

Full insertion step, including the grasp planning, servoing chamfer tilt insertion, and screw correction

Videos

2023-ICRA-Automated-Shunt-Video.mp4

Submission Overview Video

IMG_3716.MOV

Small Shunt Insertion Example

CroppedVideo-225.mp4

Large Shunt Insertion Example

Results

The three trested vessel ablation angle cases

We perform 20 trials per ablation stage, insertion condition, and shunt size. For the 8mm shunt, the success rates for the 15º and 30º vessel phantom rim angles are 90% and 85% with average trial times of 15.0s and 13.8s, respectively.

For the larger, 14mm outer diameter shunt, the success rates for the 15º and 30º vessel phantom rim angles are 90% and 50%, respectively. The negative impact on the success rate between the low-angle (0º, 15º) and high-angle cases (30º), results from a decrease in the effectiveness of the initial chamfer tilt insertion; at high angles, the initial chamfer tilt insertion results in an orientation that is not fully seated within the vessel rim, and even with the screw motion, one part of the shunt ends up outside of the vessel phantom. On the other hand, in both the 0º and 15º cases, the larger shunt was initially inserted at a well-grounded position for the following screw motion.

We consider two failure modes: a dilation failure (D), in which the robot fails to grasp the vessel phantom rim or fails to pull outward, and a shunt insertion failure (S), in which the full rim of the shunt is not enclosed by the vessel phantom after completion of the pipeline.

insertion failure.MOV

Example of a Large Shunt Insertion Failure