Automated Pixel Shifting in Digital Subtraction Angiography — an Application of Cepstral Filtering

We present a method for digital subtraction angiography based on phase-based nonrigid deformation with specific consideration for changes between the image pairs. Input images are transformed into a scale-space representation using complex-valued filter responses. We apply a novel selection criterion to discern object motion and actual change by comparing the magnitudes in the responses. By manipulating phase, we directly generate a deformed image without explicit calculation of motion vectors. Our method is particularly useful in angiographic imaging where subtle changes between the image pairs should be preserved within the deformation and subtraction process. The experiments show that the proposed method preserves contrast for vessels and tumor stains while reducing motion artifact, which is clinically meaningful. INDEX TERMSImage registration, change detection, biomedical imaging, angiography.

show abstract

“…An early method for motion compensation in DSA, pixelshift [4], tracks a region in the mask image and compensates…”

Section: A Related Workmentioning

confidence: 99%

Phase-Based Nonrigid Deformation for Digital Subtraction Angiography

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The phase-correlation measure was originally proposed by Kuglin & Hines [86] and has since then been applied to DSA by Leclerc & Benchimol [87] and Wu et al [166]. A related approach, based on the power-cepstrum 26 representations of the images, was used by Englmeier et al [38].…”

Section: Pp-11mentioning

confidence: 99%

“…There are a number of algorithms that made use of real multidimensional optimization strategies. 38 For example, Yanagisawa et al [167] used the downhill simplex method of Nelder & Mead [111], with which the optimal 2D displacement vector is found by defining three candidate vectors, the end-points of which constitute a triangle, and by proceeding downhill by means of successive reflections and contractions of this triangle. Venot et al [154] proposed to use a stochastic search approach in which, for each iteration, the displacement vector to be evaluated is computed from the vector of the previous iteration by adding a random vector according to some suitably chosen probability density function.…”

Section: Ivd Optimization and Accelerationmentioning

confidence: 99%

See 1 more Smart Citation

Retrospective motion correction in digital subtraction angiography: a review

Meijering

Niessen

Viegever³

1999

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Digital subtraction angiography (DSA) is a well-established modality for the visualization of blood vessels in the human body. A serious disadvantage of this technique, inherent to the subtraction operation, is its sensitivity to patient motion. The resulting artifacts frequently reduce the diagnostic value of the images. Over the past two decades, many solutions to this problem have been put forward. In this paper, we give an overview of the possible types of motion artifacts and the techniques that have been proposed to avoid them. The main purpose of this paper is to provide a detailed review and discussion of retrospective motion correction techniques that have been described in the literature, to summarize the conclusions that can be drawn from these studies, and to provide suggestions for future research.

show abstract

“…or by automated procedures. Computer-based comparisons may be carried out image-or template-oriented (crosscorrelation [18,22,23], squared differences [8][9][10][11][12]), edge-oriented (gradients [7,24], shape analysis [25], Hough transform, chamfer matching) or by using other features (Fourier analysis [26], moments of inertia [13], surface fitting [17,19,27], optical flow [28], etc. ).…”

Section: Introductionmentioning

confidence: 99%

Registered image subtraction for CT-, MR- and coronary angiography

et al. 1997

View full text Add to dashboard Cite

Conventional post-processing of digital subtraction angiography (DSA) by 3D spiral CT, 3D MRI and 2D DSA is often hampered by extended artefacts due to patient movements during examination. In this paper an image registration procedure prior to the digital subtraction is introduced allowing an enhanced visualization of the contrast agent. The object displacement is detected by analysis of image deformations in small local regions. The motion pattern is used to compute a new synthetic mask of maximum congruence with the contrast medium image. This new mask image is then used in the subsequent subtraction. The algorithm works fully automatically and does not need any interactive placement of landmarks. Results obtained from subtraction of uncorrected and corrected sequences were compared with each other. The registration procedure provided good results in the suppression of subtraction artefacts and in the enhancement of vascular structures. Results are presented from subtraction of 2D and 3D data from CTA, MR mammography and coronary angiography.

show abstract

Automated Pixel Shifting in Digital Subtraction Angiography — an Application of Cepstral Filtering

Cited by 4 publications

References 1 publication

Phase-Based Nonrigid Deformation for Digital Subtraction Angiography

Phase-Based Nonrigid Deformation for Digital Subtraction Angiography

Retrospective motion correction in digital subtraction angiography: a review

Registered image subtraction for CT-, MR- and coronary angiography

Contact Info

Product

Resources

About