Michel Defrise has played a key role in the transition from 2D to 3D image reconstruction, helping advance the medical imaging capabilities of positron emission tomography (PET) and computed tomography (CT). He developed the Fourier Rebinning Algorithm, which allows fast 2D iterative reconstruction from 3D PET data, contributing to the widespread clinical application of PET scans in oncology. He also demonstrated that time-offlight PET data contain sufficient information for quantitative PET imaging when anatomical CT data do not allow a reliable correction of the photon attenuation by the patient. His work on filtered-backprojection algorithms for cone-beam CT with arbitrary acquisition geometries was the cornerstone for later developments of exact and approximate algorithms for 3D CT.
An IEEE Fellow, Defrise is an Emeritus Professor with the Department of Nuclear Medicine at Vrije Universiteit Brussel, Brussels, Belgium.