We describe a model-based method of generating realistic mass lesions and placing them in appropriate locations on normal mammogram backgrounds. These lesions are highly irregular in shape and have no consistent internal landmarks. Thin plate splines are used both in modelling the background to each lesion, and to warp examples when training an appearance model.

Landmark-based thin-plate spline image registration is one of the most commonly used methods for non-rigid medical image registration and anatomical shape analysis. It is well known that this method does not produce a unique correspondence between two images away from the landmark locations because interchanging the role of source and target landmarks does not produce forward and reverse transf...

Thin plate splines provide smooth interpolation of the given data in two or more dimensions. These are analogous to cubic splines in one dimension. The main objectives of this report is to provide an implementation of the thin plate spine interpolation of data using various efficient methods and to investigate the possibility of using thin plate splines in adaptive optics. In this report, we co...

The thin plate spline (TPS) is an e ective tool for modeling coordinate transformations that has been applied successfully in several computer vision applications. Unfortunately the solution requires the inversion of a p p matrix, where p is the number of points in the data set, thus making it impractical for large scale applications. As it turns out, a surprisingly good approximate solution is...

We consider the problem of data interpolation or smoothing using a thin-plate spline approach. The thin-plate spline formulation involves the linear combination of logarithmic kernels. For problems with large data sets, current methods for evaluating the splines are too slow. Recently, Beatson and Newsam 1] have suggested using a form of multi-pole type technique for the rapid evaluation of thi...

This thesis investigates the employment of different deformable registration techniques to register pre-operative magnetic resonance and inter-operative ultrasound images during prostate biopsy. Accurate registration ensures appropriate biopsy sampling of malignant prostate tissues and reduces the rate of re-biopsies. Therefore, we provide comparisons and experimental results for some landmarka...

The qualitative and quantitative comparison of preand postoperative image data is an important possibility to validate computer assisted surgical procedures. Due to deformations after surgery a non-rigid registration scheme is a prerequisite for a precise comparison. Interactive landmark-based schemes are a suitable approach. Incorporation of a priori knowledge about the anatomical structures t...

We give some empirical comparisons between two nonparametric regression methods based on regularization: the elastic or thin-plate splines, and plastic splines based on total variation penalties. 1. Nonparametric regression with elastic and plastic splines We seek f to fit the dependence of response points zi ∈ R on two-dimensional covariates (xi, yi) living in the domain Ω. To this end, we emp...