Limits...
Alternative parametric boundary reconstruction method for biomedical imaging.

Kolibal J, Howard D - J. Biomed. Biotechnol. (2008)

Bottom Line: Determining the outline or boundary contour of a two-dimensional object, or the surface of a three-dimensional object poses difficulties particularly when there is substantial measurement noise or uncertainty.The technique is applied to parametric boundary data and has potential applications in biomedical imaging.It should be considered as one of several techniques to improve the visualization of images.

View Article: PubMed Central - PubMed

Affiliation: Department of Mathematics, College of Science and Technology, The University of Southern Mississippi, Hattiesburg, MS 39406-0001, USA. joseph.kolibal@usm.edu

ABSTRACT
Determining the outline or boundary contour of a two-dimensional object, or the surface of a three-dimensional object poses difficulties particularly when there is substantial measurement noise or uncertainty. By adapting the mathematical approach of stochastic function recovery to this task, it is possible to obtain usable estimates for these boundaries, even in the presence of large amounts of noise. The technique is applied to parametric boundary data and has potential applications in biomedical imaging. It should be considered as one of several techniques to improve the visualization of images.

Show MeSH

Related in: MedlinePlus

Recovery of the boundary of disk specified by 96 points with Gaussiannoise of ν = 3, starting at (1, 0) on the x-axis and moving counterclockwise around thecircle with the boundary recovered using approximate interpolation, with σA = 1 × 10−6 and σB = 1 × 10−4. The recovered curve is shown as a thicker line.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2366076&req=5

fig2: Recovery of the boundary of disk specified by 96 points with Gaussiannoise of ν = 3, starting at (1, 0) on the x-axis and moving counterclockwise around thecircle with the boundary recovered using approximate interpolation, with σA = 1 × 10−6 and σB = 1 × 10−4. The recovered curve is shown as a thicker line.

Mentions: Recovering the boundary of a disk becomes moredifficult, as shown in Figure 2, particularly if the amount of noise is quitelarge. In this example, the Gaussian noise for a disk of radius 5 is specified as ν = 3 yielding an extensively scattered data set. While the level of noise in this illustrative example is much higher than wouldbe expected in any realistic imaging situation, the example serves a two-foldpurpose: (1) on the one had it shows the robustness of the method at recoveringa reasonable representation of the surface from the data that is more consistentwith noise than data, and (2) it shows that the effects of errors inconstructing the parameterization are less of an issue than might be presumed.In the figure, the connectivity of the boundary data is not ordered in θ moving counterclockwise around the circle, and so it is quite likely that any minor errors in parameterization, for example,using even the simplest unconstrained nearest neighbor search of the data,would cause unrecoverable errors in the surface representation that isrecovered.


Alternative parametric boundary reconstruction method for biomedical imaging.

Kolibal J, Howard D - J. Biomed. Biotechnol. (2008)

Recovery of the boundary of disk specified by 96 points with Gaussiannoise of ν = 3, starting at (1, 0) on the x-axis and moving counterclockwise around thecircle with the boundary recovered using approximate interpolation, with σA = 1 × 10−6 and σB = 1 × 10−4. The recovered curve is shown as a thicker line.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2366076&req=5

fig2: Recovery of the boundary of disk specified by 96 points with Gaussiannoise of ν = 3, starting at (1, 0) on the x-axis and moving counterclockwise around thecircle with the boundary recovered using approximate interpolation, with σA = 1 × 10−6 and σB = 1 × 10−4. The recovered curve is shown as a thicker line.
Mentions: Recovering the boundary of a disk becomes moredifficult, as shown in Figure 2, particularly if the amount of noise is quitelarge. In this example, the Gaussian noise for a disk of radius 5 is specified as ν = 3 yielding an extensively scattered data set. While the level of noise in this illustrative example is much higher than wouldbe expected in any realistic imaging situation, the example serves a two-foldpurpose: (1) on the one had it shows the robustness of the method at recoveringa reasonable representation of the surface from the data that is more consistentwith noise than data, and (2) it shows that the effects of errors inconstructing the parameterization are less of an issue than might be presumed.In the figure, the connectivity of the boundary data is not ordered in θ moving counterclockwise around the circle, and so it is quite likely that any minor errors in parameterization, for example,using even the simplest unconstrained nearest neighbor search of the data,would cause unrecoverable errors in the surface representation that isrecovered.

Bottom Line: Determining the outline or boundary contour of a two-dimensional object, or the surface of a three-dimensional object poses difficulties particularly when there is substantial measurement noise or uncertainty.The technique is applied to parametric boundary data and has potential applications in biomedical imaging.It should be considered as one of several techniques to improve the visualization of images.

View Article: PubMed Central - PubMed

Affiliation: Department of Mathematics, College of Science and Technology, The University of Southern Mississippi, Hattiesburg, MS 39406-0001, USA. joseph.kolibal@usm.edu

ABSTRACT
Determining the outline or boundary contour of a two-dimensional object, or the surface of a three-dimensional object poses difficulties particularly when there is substantial measurement noise or uncertainty. By adapting the mathematical approach of stochastic function recovery to this task, it is possible to obtain usable estimates for these boundaries, even in the presence of large amounts of noise. The technique is applied to parametric boundary data and has potential applications in biomedical imaging. It should be considered as one of several techniques to improve the visualization of images.

Show MeSH
Related in: MedlinePlus