Institut Denis Poisson, UMR CNRS, Université d'Orléans & Université de Tours, France.

Title: Statistics Methods for Medical Image Processing and Reconstruction

Abstract :

In this talk we will see how statistical methods, from the simplest to the most advanced ones, can be used to address various problems in medical image processing and reconstruction for different imaging modalities. Image reconstruction allows to obtain the images in question, while image processing (on the already reconstructed images) aims at extracting some information of interest. We will review several statistical methods (manely Bayesian) to address various problems of this type.

Keywords: Image processing, image reconstruction, Statistics, frequentist, Bayesian, Parametrics, Nonparametrics.

References

1. D. Fall, N. Dobigeon, P. Auzou, "A Bayesian Estimation Formulation to Voxel-Based Lesion Symptom Mapping", Proc. European Signal Processing Conf. (EUSIPCO), Belgrade, Serbia, Sept. 2022.
2. D. Fall, "Bayesian Nonparametrics and Biostatistics: the case of PET Imaging", International Journal of Biostatistics, 2019.
3. D. Fall, E. Lavau and P. Auzou, "Voxel-Based Lesion-Symptom Mapping: a Nonparametric Bayesian Approach", Proc. IEEE International Conference on Acoustics, Speech and Signl Processing (ICASSP) 2018.