Abstract:The diagnosis of medical conditions and subsequent treatment often involves radionuclide imaging techniques. To refine localisation accuracy and improve diagnostic confidence, compared with the use of a single scanning technique, a combination of two (or more) techniques can be used but with a higher risk of misalignment. For this to be reliable and accurate, recorded data undergo processing to suppress noise and enhance resolution. A step in image processing techniques for such inverse problems is the inclusion of smoothing. Standard approaches, however, are usually limited to applying identical models globally. In this study, we propose a novel Laplace and Gaussian mixture prior distribution that incorporates different smoothing strategies with the automatic model-based estimation of mixture component weightings creating a locally adaptive model. A fully Bayesian approach is presented using multi-level hierarchical modelling and Markov chain Monte Carlo (MCMC) estimation methods to sample from the posterior distribution and hence perform estimation. The proposed methods are assessed using simulated γ -eye TM camera images and demonstrate greater noise reduction than existing methods but without compromising resolution. As well as image estimates, the MCMC methods also provide posterior variance estimates and hence uncertainty quantification takes into consideration any potential sources of variability. The use of mixture prior models, part Laplace random field and part Gaussian random field, within a Bayesian modelling approach is not limited to medical imaging applications but provides a more general framework for analysing other spatial inverse problems. Locally adaptive prior distributions provide a more realistic model, which leads to robust results and hence more reliable decision-making, especially in nuclear medicine. They can become a standard part of the toolkit of everyone working in image processing applications.

Bayesian Separation of Document Images with Hidden Markov Model

A Hierarchical Markov Random Field Model for Bayesian Blind Image Separation

Bayesian Blind Separation Of Mixed Text Patterns

Background Subtraction Based on Nonparametric Bayesian Estimation.

A Hidden Markov model for Bayesian data fusion of multivariate signals

Hierarchical Markovian models for hyperspectral image segmentation

Bayesian mixtures of common factor analyzers: Model, variational inference, and applications

A Bayesian Approach to Clustering Matting Components in Spectral Matting

A Hierarchical Bayesian Model Accounting for Endmember Variability and Abrupt Spectral Changes to Unmix Multitemporal Hyperspectral Images

Spatial-based Bayesian Hidden Markov Models with Dirichlet Mixtures for Video Anomaly Detection

Bayesian Inference for Layer Representation with Mixed Markov Random Field

Multi-Channel Adaptive Mixture Background Model for Real-time Tracking.

Mixture Bayesian Regularization of PCR Model and Soft Sensing Application

Blind Separation of Superimposed Moving Images Using Image Statistics

Mixture prior distributions and Bayesian models for robust radionuclide image processing

Decomposition Of Mixed Pixels Using Bayesian Self-Organizing Map (Bsom) Neural Networks

Enhancing hyperspectral image unmixing with spatial correlations

Bayesian Hyperspectral and Multispectral Image Fusions via Double Matrix Factorization.

A Hierarchical Gamma Mixture Model Toward Hidden Markov Random Field for High-Resolution SAR Image Segmentation

Decomposition of Mixed Pixels Based on Bayesian Self-Organizing Map and Gaussian Mixture Model.

Towards interpretability of Mixtures of Hidden Markov Models