T.F. Cootes

Publications (3)0 Total impact

• Source

  Available from: tina-vision.net

  Article: Bayesian and non-Bayesian probabilistic models for medical image analysis

  P A Bromiley, N A Thacker, M L J Scott, M Pokri, A J Lacey, T F Cootes
  [show abstract] [hide abstract]
  ABSTRACT: Bayesian approaches to data analysis are popular in machine vision, and yet the main advantage of Bayes theory, the ability to incorporate prior knowledge in the form of the prior probabilities, may lead to problems in some quantitative tasks. In this paper we demonstrate examples of Bayesian and non-Bayesian techniques from the area of magnetic resonance image (MRI) analysis. Issues raised by these examples are used to illustrate difficulties in Bayesian methods and to motivate an approach based on frequentist methods. We believe this approach to be more suited to quantitative data analysis, and provide a general theory for the use of these methods in learning (Bayes risk) systems and for data fusion. Proofs are given for the more novel aspects of the theory. We conclude with a discussion of the strengths and weaknesses, and the fundamental suitability, of Bayesian and non-Bayesian approaches for MRI analysis in particular, and for machine vision systems in general. q 2003 Published by Elsevier Science B.V.
  Image Vision Comput. 01/2003; 21.
• Source

  Available from: man.ac.uk

  Article: Derivation of the renormalisation formula for the product of uniform probability distributions and extension to non-integer dimensionality

  P A Bromiley, T F Cootes, N A Thacker
  [show abstract] [hide abstract]
  ABSTRACT: A standard formula, given below, exists for renormalising (reflattening) a quantity that is the product of several quantities having uniform probability distributions. However, extensive literature searches have failed to reveal an acceptable derivation for the equation. Therefore, we present derivations using a variety of techniques: derivations for problems of fixed dimensionality using integration in the sample space defined by the problem, derivations in log space leading to a derivation for problems of arbitrary dimensionality, and a short derivation building on the ideas in the previous sections which is suitable for use in publications. Finally, we extend the original renormalisation formula to cope with non-integer values for the dimensionality.
• Source

  Available from: tina-vision.net

  Article: Bayesian and non-Bayesian probabilistic models for medical image analysis

  P.A. Bromiley, N.A. Thacker, M.L.J. Scott, M. Pokrić, A.J. Lacey, T.F. Cootes
  [show abstract] [hide abstract]
  ABSTRACT: Bayesian approaches to data analysis are popular in machine vision, and yet the main advantage of Bayes theory, the ability to incorporate prior knowledge in the form of the prior probabilities, may lead to problems in some quantitative tasks. In this paper we demonstrate examples of Bayesian and non-Bayesian techniques from the area of magnetic resonance image (MRI) analysis. Issues raised by these examples are used to illustrate difficulties in Bayesian methods and to motivate an approach based on frequentist methods. We believe this approach to be more suited to quantitative data analysis, and provide a general theory for the use of these methods in learning (Bayes risk) systems and for data fusion. Proofs are given for the more novel aspects of the theory. We conclude with a discussion of the strengths and weaknesses, and the fundamental suitability, of Bayesian and non-Bayesian approaches for MRI analysis in particular, and for machine vision systems in general.
  Image and Vision Computing.