Registration of Untracked 2D Laparoscopic Ultrasound Liver Images to CT Using Content-Based Retrieval and Kinematic Priors

Abstract

Laparoscopic Ultrasound (LUS) can enhance the safety of laparoscopic liver resection by providing information on the location of major blood vessels and tumours. Since many tumours are not visible in ultrasound, registration to a pre-operative CT has been proposed as a guidance method. In addition to being multi-modal, this registration problem is greatly affected by the differences in field of view between CT and LUS, and thus requires an accurate initialisation. We propose a novel method of registering smaller field of view slices to a larger volume globally using a Content-based retrieval framework. This problem is under-constrained for a single slice registration, resulting in non-unique solutions. Therefore, we introduce kinematic priors in a Bayesian framework in order to jointly register groups of ultrasound images. Our method then produces an estimate of the most likely sequence of CT images to represent the ultrasound acquisition and does not require tracking information nor an accurate initialisation. We demonstrate the feasibility of this approach in multiple LUS acquisitions taken from three sets of clinical data.

Registration of Untracked 2D Laparoscopic Ultrasound to CT Images of the Liver Using Multi-Labelled Content-Based Image Retrieval

“…This enables the registration problem to be accurately initialised without tracking data. Previously, we presented preliminary results on a limited sample of synthetic and real LUS sequences, demonstrating the feasibility of this framework without addressing the CBIR system performance comprehensively [11]. In this work, we generalise our CBIR system to include multiple labels in the vessel feature encoding which increases registration performance.…”

“…In this paper, we extend a novel registration method that provides an accurate initialisation to the problem without requiring tracking information nor a manual interaction with the images. Preliminary results have been previously presented [11].…”

Vessel segmentation for automatic registration of untracked laparoscopic ultrasound to CT of the liver

“…To further validate our segmentation, we integrate our results in our previously developed untracked LUS to CT registration framework [8,9]. In summary, this method comprises two main steps, an image retrieval step and a probabilistic optimisation.…”

“…However, the manual identification of anatomical landmarks for an initialisation disrupts the clinical workflow, and the introduction of tracking devices in the operating room is costly and logistically challenging. To enable clinical translation, we have previously proposed an untracked registration method based on Content-based Image Retrieval (CBIR) that replaces the need for a manual initialisation [8,9], and presented results using manually segmented vessels. In this paper, we perform the first automatic vessel segmentation of 2D untracked LUS images using deep learning (DL), and show its potential in the automation of our registration pipeline.…”

Fan-Slicer: A Pycuda Package for Fast Reslicing of Ultrasound Shaped Planes

“…Fan-slicer is a package designed for the sampling/ simulation of ultrasound shaped planes from a preoperative scan such as Computed Tomography (CT) or Magnetic Resonance Imaging (MRI). This software has been initially implemented as part of an imaging pipeline to aid the development of ultrasound guidance algorithms for laparoscopic liver surgery [1,2] and endoscopic interventions [3]. Given a set of Laparoscopic Ultrasound (LUS) images and a pre-operative 3D scan, the resampling of LUS planes in pre-operative space enables both the implementation of image registration pipelines and visualisation of the corresponding results.…”