Original Contribution
Toward Registration of 3D Ultrasound and CT Images of the Spine in Clinical Praxis: Design and Evaluation of a Data Acquisition Protocol

https://doi.org/10.1016/j.ultrasmedbio.2009.06.1089Get rights and content

Abstract

Recent work has demonstrated the accuracy and operational viability of an algorithm proposed by the authors that successfully registers 3-D ultrasound data with CT or MRI data. The successful application of this method to intraoperative navigation, however, depends critically on the quality of the acquired ultrasound data. This gives rise to two questions concerning the usability of the algorithm in clinical praxis. First, how can one guarantee high-quality, user-independent ultrasound registration data with this procedure? Second, can this approach work reliably in clinical practice, namely within the operating theater? To address both of these questions, we present an ultrasound data acquisition protocol that leads the user through the data acquisition process and also provides the criteria to adjust the relevant ultrasound parameters. We also evaluated criteria for the visual inspection of the suitability of the ultrasound data for the registration process. Results for this evaluation show that these visual criteria can be used to decide preoperatively if an ultrasound registration will be successful in a patient. The intraoperative evaluation of the protocol showed that high-quality registrations can be achieved under realistic conditions. This protocol and the visual inspection criteria, together with the ultrasound registration algorithm, provide a surgical team with a means of performing precise, cost-effective navigation in patients for whom a navigated intervention was previously impossible. We evaluated the proposed procedure in clinical practice. (E-mail: [email protected])

Section snippets

Introduction and Literature

Pedicle screw fixation for spinal column fusion is an established procedure for the treatment of spinal instability (Roy-Camille et al. 1986; Krismer et al. 1997). The screws have to be inserted through the spinal pedicles into the vertebrae, and the surgeon has to decide where to place the screws and how they should be angled. For purposes of orientation, the surgeon needs a good anatomical overview. This can usually be achieved only at the cost of considerable tissue disruption. Image-guided

Materials and Methods

If diagnostic ultrasound is used as intraoperative imaging modality for registration, the ultrasound data acquisition has to be integrated into the surgical procedure.

The concept of image-guided surgery with intraoperative ultrasound imaging is presented in Fig. 1. The 3-D CT data obtained preoperatively for diagnosis and operation planning have to be matched with the intraoperatively recorded 3-D ultrasound data as the operation progresses. Because ultrasound data acquisition, preprocessing of

Results

For the design and evaluation of our registration algorithm and the registration protocol, we considered the first eight datasets from Winter et al. (2008). For evaluation of the clinical protocol we considered these eight and the four new data sets. We show that it is possible to decide before surgery whether ultrasound can be used for intraoperative registration. In addition, we present an evaluation of our registration algorithm, focusing on the clinical relevant aspects.

Discussion

There are only a few studies on the registration of bony structures in ultrasound and CT or MRI data. Existing volume-based techniques such as gray-value correlation or maximization of mutual information (Pluim et al. 2003; Studholme et al. 1996; Wells et al. 1996), which belong to the standard procedures in registration of CT or MRI data, cannot be used for ultrasound registration of the spine because it is a prerequirement of volume-based approaches that the same structures be visualized in

Summary

Few attempts have been made to register bony structures of the spine with ultrasound and CT or MRI data. Automatic methods that require segmentation of the bone surface in the ultrasound datasets have not yet been successfully applied to the spinal column in vivo. The method we used, however, can register the vertebrae of patients in ultrasound and CT data.

Because our method depends on the quality of the acquired ultrasound data, two questions remain. The first is how to guarantee a

Acknowledgment

This work was an activity of the Kompetenzzentrum Medizintechnik Ruhr (KMR), Bochum. It was supported by the Bundesministerium für Bildung und Forschung (Az. 13N8079).

References (29)

  • F. Gebhard et al.

    Grenzen der CT-basierten Computernavigation in der Wirbelsäulenchirurgie

    Der Unfallchirurg

    (2000)
  • Hansen N, Kern S. Evaluating the CMA evolution strategy on multimodal test functions. In: Yao X, Burke E, Lozano JA,...
  • N. Hansen et al.

    Completely derandomized self-adaptation in evolution strategies

    Evol Comput

    (2001)
  • Igel C, Heidrich-Meisner V, Glasmachers T. Shark. J Machine Learn Res...
  • Cited by (20)

    • The state-of-the-art in ultrasound-guided spine interventions

      2020, Medical Image Analysis
      Citation Excerpt :

      On the CT image, the density of bone tissues correlates with the CT intensity which lies between 140 to 260 Hounsfield units (HU) (Yan et al., 2011; Sugano, 2003; Winter et al., 2008). The posterior surface of the vertebrae can be obtained by ray tracing a line going from posterior to anterior, in which the surface location corresponds to the first occurrence of voxels above a given threshold, typically 150 HU (Brendel et al., 2002; 2005; Winter et al., 2008; 2009; Yan et al., 2011; 2012b; 2012a; Gueziri et al., 2019). The approach yields satisfactory results to extract the posterior vertebral surface.

    • Navigated 3-Dimensional Intraoperative Ultrasound for Spine Surgery

      2019, World Neurosurgery
      Citation Excerpt :

      The change of spinal alignment due to patient positioning, which is completely different in preoperative imaging in supine position compared with the intraoperative situation with the patient in prone position, can only be accounted for by intraoperative imaging, as we had implemented it with iCT for registration. iUS itself might be an alternative for registration, however, iUS-based registration of spinal structures is still in a very experimental state and not yet accurate enough for clinical application.29-32 Improved surface-based registration techniques of CT and reconstructed bony structures in intraoperative 3D-US, as demonstrated in a tibia phantom, might be helpful for registration in spine surgery as well.33

    • In Vivo Measurement of the Human Lumbar Spine Using Magnetic Resonance Imaging to Ultrasound Registration

      2019, Journal of Manipulative and Physiological Therapeutics
      Citation Excerpt :

      Although excluding the TP would reduce the area of available posterior vertebral surface for registration use, which might have a negative effect on the registration accuracy, our results implied that the spatial noise and artifacts in the surrounding soft tissues was a more critical factor that could affect the robustness of our MRI-ultrasound registration algorithm. Given that the success of the MRI-ultrasound registration depends heavily on the quality of acquired ultrasound images,22 to avoid the waste of time, effort, and money, it is imperative to check for ultrasound image quality before proceeding with any image processing work and MRI scan. In this study, a dated ultrasound scanner (Ultramark 400c, ATL Ultrasound Inc) was used.

    • Fast and Accurate Data Extraction for Near Real-Time Registration of 3-D Ultrasound and Computed Tomography in Orthopedic Surgery

      2015, Ultrasound in Medicine and Biology
      Citation Excerpt :

      Many groups have tried eliminating the need for of bone localization by registering the US and CT volumes using intensity-based registration methods. Brendel et al. (2002, 2005); Dekomien et al. (2007) and Winter et al. (2002, 2009) fused CT and US volumes of the spine and developed an intensity-based algorithm that required minimal US pre-processing. They proposed several registration methods that took advantage of the segmentation of CT volumes and the ridgelike structures of bone in US volumes.

    View all citing articles on Scopus
    View full text