Non-Rigid Volume to Surface Registration using a Data-Driven Biomechanical Model

TL;DR

This paper introduces a neural network-based method for real-time non-rigid registration of preoperative 3D organ models to intraoperative surface data in laparoscopic liver surgery, handling noisy and sparse data without re-training.

Contribution

It presents a data-driven biomechanical model trained on simulations that generalizes to new patients and robustly handles noisy intraoperative data for accurate registration.

Findings

High accuracy in real data registration

Robust to noisy intraoperative surfaces

Fast inference suitable for real-time applications

Abstract

Non-rigid registration is a key component in soft-tissue navigation. We focus on laparoscopic liver surgery, where we register the organ model obtained from a preoperative CT scan to the intraoperative partial organ surface, reconstructed from the laparoscopic video. This is a challenging task due to sparse and noisy intraoperative data, real-time requirements and many unknowns - such as tissue properties and boundary conditions. Furthermore, establishing correspondences between pre- and intraoperative data can be extremely difficult since the liver usually lacks distinct surface features and the used imaging modalities suffer from very different types of noise. In this work, we train a convolutional neural network to perform both the search for surface correspondences as well as the non-rigid registration in one step. The network is trained on physically accurate biomechanical simulations of randomly generated, deforming organ-like structures. This enables the network to immediately generalize to a new patient organ without the need to re-train. We add various amounts of noise to the intraoperative surfaces during training, making the network robust to noisy intraoperative data. During inference, the network outputs the displacement field which matches the preoperative volume to the partial intraoperative surface. In multiple experiments, we show that the network translates well to real data while maintaining a high inference speed. Our code is made available online.