Gaussian Mixture Model for brain MRI Segmentation
In the last decades, Magnetic Resonance Imaging (MRI) has become a central tool in brain clinical studies. Most of these studies rely on accurate and
Stefano Cerri
Postdoctoral Researcher
Pioneer Centre for AI
stce at di dot ku dot dk
Menu
About
I’m a postdoctoral researcher at the Pioneer Centre for AI, developing generative models for analyzing brain MRI data. Previously, I was a research fellow at the Athinoula A. Martinos Center for Biomedical Imaging. I obtained my PhD at the Technical University of Denmark (DTU), supervised by Koen Van Leemput. I was also part of the 15 PhD students of the Translational Brain Imaging Train Network (TRABIT).
Past Experiences
- Research Fellow, Athinoula A. Martinos Center for Biomedical Imaging, 2021-2023
- PhD Degree, Medical Image Analysis, Technical University of Denmark, 2018-2021
- Machine Learning Intern, Horus Technology, 2017
- Master Degree, Computer Science, Politecnico di Milano, 2015-2017
- Bachelor Degree, Computer Science, Politecnico di Milano, 2012-2015
Publications
Selected publications, for a full list see Google scholar.
Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images
Sveinn Pálsson, Stefano Cerri, Hans Skovgaard Poulsen, Thomas Urup, Ian Law Koen Van Leemput
Scientific Reports, Vol. 12, 2022, 19744
Paper
Software
I’m actively maintaining the segmentation code of SAMSEG, included in FreeSurfer.
Modeling the Bias Field in Brain Segmentation
In a previous post, I described how to obtain brain segmentations using a Gaussian Mixture Model (GMM). One of the limitations of this brain segmentation