CoChat
Sign Up
Cited by 10Medical Imaging and AnalysisMedical Image Segmentation TechniquesRadiation Dose and Imaging

Machine learning for automated and real-time two-dimensional to three-dimensional registration of the spine using a single radiograph

Open access
Abumoussa, Andrew, Gopalakrishnan, Vivek, Succop, Benjamin, Galgano, Michael, Jaikumar, Sivakumar, Lee, Yueh Z., Bhowmick, Deb A.·2023·Neurosurgical Focus
DOI 10.3171/2023.3.focus2345
Open this paper in CoChat — free accountOpen PDF

Abstract

OBJECTIVE: The goal of this work was to methodically evaluate, optimize, and validate a self-supervised machine learning algorithm capable of real-time automatic registration and fluoroscopic localization of the spine using a single radiograph or fluoroscopic frame. METHODS: The authors propose a two-dimensional to three-dimensional (2D-3D) registration algorithm that maximizes an image similarity metric between radiographic images to identify the position of a C-arm relative to a 3D volume. This work utilizes digitally reconstructed radiographs (DRRs), which are synthetic radiographic images generated by simulating the x-ray projections as they would pass through a CT volume. To evaluate the algorithm, the authors used cone-beam CT data for 127 patients obtained from an open-source de-identified registry of cervical, thoracic, and lumbar scans. They systematically evaluated and tuned the algorithm, then quantified the convergence rate of the model by simulating C-arm registrations with 80 randomly simulated DRRs for each CT volume. The endpoints of this study were time to convergence, accuracy of convergence for each of the C-arm’s degrees of freedom, and overall registration accuracy based on a voxel-by-voxel measurement. RESULTS: A total of 10,160 unique radiographic images were simulated from 127 CT scans. The algorithm successfully converged to the correct solution 82% of the time with an average of 1.96 seconds of computation. The radiographic images for which the algorithm converged to the solution demonstrated 99.9% registration accuracy despite utilizing only single-precision computation for speed. The algorithm was found to be optimized for convergence when the search space was limited to a ± 45° offset in the right anterior oblique/left anterior oblique, cranial/caudal, and receiver rotation angles with the radiographic isocenter contained within 8000 cm 3 of the volumetric center of the CT volume. CONCLUSIONS: The investigated machine learning algorithm has the potential to aid surgeons in level localization, surgical planning, and intraoperative navigation through a completely automated 2D-3D registration process. Future work will focus on algorithmic optimizations to improve the convergence rate and speed profile.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3171/2023.3.focus2345

Cites

42 of 42 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Gradient-based optimization with DiffDRR recovers the true pose faster and more robustly than gradient-free methods on synthetic data.

    “On synthetic data, gradient-based optimization with DiffDRR, a differentiable renderer, has been shown to recover the true pose faster and more robustly than gradient-free methods [1, 22].”

    From Intraoperative 2D/3D Image Registration via Differentiable X-Ray Rendering (2024) · cites this paper as [1]

    The Experimental Design and Results sections describe using DiffDRR with gradient descent to recover the target pose from simulated radiographs. The Limitations and Future Directions section says this was not a systematic comparison with other methods, so the claimed speed and robustness advantage over gradient-free methods is unsupported. Page numbers were not supplied in our copy. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Registration of intraoperative 2D X-rays to 3D preoperative scans is a prerequisite for numerous advanced image guidance techniques, including vertebral level localization, surgical plan reprojection, instrument tracking, motion correction, and surgical robotics.

    “The registration of intraoperative 2D X-rays to 3D preoperative scans is a prerequisite for numerous advanced image guidance techniques, including vertebral level localization [11], surgical plan reprojection [12,13], instrument tracking [14], motion correction [15], and surgical robotics [16].”

    From Rapid patient-specific neural networks for intraoperative X-ray to volume registration (2025) · cites this paper as [11]

    Citation [16] is attached to the surgical-robotics portion of the sentence. The Discussion says 2D–3D registration is used for surgical planning and navigation, but the passages found do not establish that registering intraoperative X-rays to preoperative scans is a prerequisite for surgical robotics. A reference title mentions guiding a surgical robot, but that alone does not establish the prerequisite claim. (LLM, conf=0.70)

    Experimental — accuracy not yet measured

  • Gradient normalized cross correlation (gradNCC) is an image similarity metric that computes the correlation between Sobel-filtered versions of the two images.

    “gradient normalized cross correlation (gradNCC) [11], an image similarity metric that computes the correlation between Sobel-filtered versions of the two images.”

    From Rapid patient-specific neural networks for intraoperative X-ray to volume registration (2025) · cites this paper as [11]

    The Experimental Design section describes zero-normalized cross-correlation (ZNCC) as an image-similarity loss function, not gradNCC. Searches for gradNCC and Sobel filtering found no passage in our copy defining the claimed metric or describing correlation between Sobel-filtered images; the section labels provide no page numbers. (LLM, conf=0.73)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

2 of the 44 references couldn’t be matched to a known paper.

Research works printed but not linked: 3

  1. Longo UG, Loppini M, Romeo G, Maffulli N, Denaro V. Er- rors of level in spinal surgery: an evidence-based systematic review. J Bone Joint Surg Br. 2012;94(11):1546-1550.
  2. Munbodh R, Jaffray DA, Moseley DJ, et al. Automated 2D- 3D registration of a radiograph and a cone beam CT using line-segment enhancement. Med Phys. 2006;33(5):1398-1411.
  3. Briechle K, Hanebeck UD. Template matching using fast normalized cross correlation. In: Casasent DP, Chao TH, eds. SPIE Proceedings Volume 4387. Optical Pattern Recognition XII. SPIE; 2001: 95-102.

Cited by

Sorted by citations

Related

Authors
Abumoussa, A., Gopalakrishnan, V., Succop, B., Galgano, M., Jaikumar, S., Lee, Y. Z., & Bhowmick, D. A.
Affiliations
  • Departments of Neurosurgery and
  • Massachusetts Institute of Technology (US)
  • University of North Carolina at Chapel Hill (US)
  • Duke University Hospital (US)
ORCID
Venue
Neurosurgical Focus
Publisher
Journal of Neurosurgery Publishing Group (JNSPG)
Published
2023-06
Volume
54
Issue
6
Pages
E16
ISSN
1092-0684 (ISSN-L)
DOI
10.3171/2023.3.focus2345
OpenAlex
W4379056406
Cited by
9
References
44
Citations by year
2026: 3 · 2025: 3 · 2024: 3
FWCI
0.71
Type
journal article
Language
en
Open access
Locations
  • Neurosurgical FOCUS (publishedVersion)
  • PubMed (publishedVersion)
  • Carolina Digital Repository (University of North Carolina at Chapel Hill) (submittedVersion)
  • UNC Libraries
Topics
  • Physical Sciences › Engineering › Biomedical Engineering › Medical Imaging and Analysis
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiation Dose and Imaging
Keywords
2D-3D registration, spine, C-arm, digitally reconstructed radiographs, cone-beam computed tomography, self-supervised learning, real-time registration, surgical navigation, surgical planning, image similarity measures
MeSH terms
Machine Learning, Algorithms, Humans, Radiography, Spine, Tomography, X-Ray Computed, Imaging, Three-Dimensional

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review