A Kinematic Bottleneck Approach For Pose Regression of Flexible Surgical Instruments directly from Images

TL;DR

This paper introduces a self-supervised, real-time image-based method for 3-D pose estimation of flexible surgical instruments, leveraging a kinematic bottleneck and physical model to avoid manual annotations.

Contribution

It presents a novel auto-encoder framework that uses a kinematic bottleneck and physical model for self-supervised training, enabling real-time pose estimation without manual labels.

Findings

Validated on semi-synthetic, phantom, and in-vivo datasets

Achieved promising real-time 3-D pose estimation results

Demonstrated effectiveness for flexible robotic endoscopes

Abstract

3-D pose estimation of instruments is a crucial step towards automatic scene understanding in robotic minimally invasive surgery. Although robotic systems can potentially directly provide joint values, this information is not commonly exploited inside the operating room, due to its possible unreliability, limited access and the time-consuming calibration required, especially for continuum robots. For this reason, standard approaches for 3-D pose estimation involve the use of external tracking systems. Recently, image-based methods have emerged as promising, non-invasive alternatives. While many image-based approaches in the literature have shown accurate results, they generally require either a complex iterative optimization for each processed image, making them unsuitable for real-time applications, or a large number of manually-annotated images for efficient learning. In this paper we propose a self-supervised image-based method, exploiting, at training time only, the imprecise kinematic information provided by the robot. In order to avoid introducing time-consuming manual annotations, the problem is formulated as an auto-encoder, smartly bottlenecked by the presence of a physical model of the robotic instruments and surgical camera, forcing a separation between image background and kinematic content. Validation of the method was performed on semi-synthetic, phantom and in-vivo datasets, obtained using a flexible robotized endoscope, showing promising results for real-time image-based 3-D pose estimation of surgical instruments.