Search Contact information
University of Cambridge Home Department of Engineering
University of Cambridge > Engineering Department > Machine Intelligence Lab

Abstract for rohling_thesis

PhD Thesis, University of Cambridge


Robert N Rohling

September 1998

In 3D freehand ultrasound, a position sensor is attached to a conventional ultrasound probe so that a set of 2D B-scans are acquired along with their relative locations. This allows the set of B-scans to be reconstructed into a regular 3D voxel array for visualisation and analysis. This thesis is concerned with two topics: reconstruction and spatial compounding of 3D freehand ultrasound data.

A key requirement for reconstruction is calibration of the freehand system. Four different calibration techniques are described using a common mathematical framework, then evaluated according to their repeatability, precision and accuracy. The new Cambridge phantom and the conventional cross-wire phantom are shown to offer the best performance.

Following calibration, interpolation of the set of irregularly spaced B-scans is required to reconstruct a regular voxel array. A radial basis function interpolation method is adapted for reconstruction of ultrasound data and compared with three standard reconstruction methods. The new functional method exploits the underlying shape of the ultrasound data to fill gaps and average overlapping data, without introducing the kind of artifacts common with the standard methods.

The second topic, spatial compounding, aims to improve the visualisation and volumetric analysis of the reconstructed data, typically inhibited by artifacts such as speckle. The basic principle of spatial compounding is to average multiple images of the same object taken from different viewpoints. With averaging, the signal to noise ratio increases, but accurate image registration is required to avoid blurring. Although a number of 2D studies can be found in the literature, this is the first to fully investigate 3D spatial compounding.

An initial series of tests are performed in vitro. The statistical theory to predict the increase in signal to noise ratio is developed and verified empirically. Using a novel registration method based on image landmarks, spatial compounding is shown to improve visualisation and volume estimation. Having proven the benefits of spatial compounding in vitro, tests with higher levels of compounding are performed in vivo. A more robust correlation-based registration technique is used on an examination of a human gall bladder and, again, visualisation and volume estimation are clearly improved.

Key words: 3D ultrasound, freehand, reconstruction, calibration, interpolation, radial basis function, spatial compounding, registration.

[7.8 MB compressed PostScript, 184 pages]

(ftp:) (http:)
PDF (automatically generated from original PostScript document - may be badly aliased on screen):
  (ftp:) rohling_thesis.pdf | (http:) rohling_thesis.pdf

If you have difficulty viewing files that end '.gz', which are gzip compressed, then you may be able to find tools to uncompress them at the gzip web site.

If you have difficulty viewing files that are in PostScript, (ending '.ps' or '.ps.gz'), then you may be able to find tools to view them at the gsview web site.

We have attempted to provide automatically generated PDF copies of documents for which only PostScript versions have previously been available. These are clearly marked in the database - due to the nature of the automatic conversion process, they are likely to be badly aliased when viewed at default resolution on screen by acroread.

© 2005 Cambridge University Engineering Dept
Information provided by milab-maintainer