|University of Copenhagen|
|2100 København Ø|
I enjoy research in medical image analysis, computer vision, and machine learning.
I received a Ph.D degree in Computer Science (CS) from the University of Copenhagen (DIKU), Denmark in 2012, advised by Mads Nielsen and Sami Brandt. During my PhD, I spent half a year in the CS Department at Stanford University under the supervision of Andrew Ng. Previously, I graduated with a Diploma (equivalent to MSc) in CS from the University of Freiburg in Germany. During my Diploma, I was a visiting scholar for half a year at the University of Western Australia and for one year at the University of South Australia.
I am particularly interested in
- developing biomarkers for breast cancer, cardiovascular diseases, and osteoporosis
- segmentation methods, deep learning, shape modeling, Bayesian methods, probabilistic graphical models, and domain adaptation.
Adhish Prasoon, Kersten Petersen, Christian Igel, Francois Lauze, Erik Dam, and Mads Nielsen, Deep Feature Learning for Knee Cartilage Segmentation Using a Triplanar Convo- lutional Neural Network, In MICCAI(2), pages 246-253, 2013
Kersten Petersen, Martin Lillholm, Peng Fei Diao, Joselene Marques, Nico Karssemeijer, and Mads Nielsen, Automated Density Risk Scoring with Learned Hierarchical Features, In International Workshop for Breast Densitometry and Breast Cancer Risk Assessment 2013.
Joselene Marquez, Dan Richter Jørgensen, Peng Fei Diao, Kersten Petersen, Nico Karsse- meijer, and Mads Nielsen, Several mammographic texture measures, including MTR, add simultaneously to percentage density risk segregation, In International Workshop for Breast Densitometry and Breast Cancer Risk Assessment 2013.
Kersten Petersen, Konstantin Chernoff, Gopal Karemore, and Mads Nielsen, Fully automatic estimation of film-based breast percentage density separate out postmenopausal hormone replacement treatment effects as well as expert’s estimation, In ECR 2013.
Peter Mysling, Kersten Petersen, Mads Nielsen, and Martin Lillholm, A unifying frame- work for automatic and semi-automatic segmentation of vertebrae from radiographs using sample-driven active shape models, Machine Vision and Applications, 24(7):1–14, 2012.
- Kersten Petersen, Konstantin Chernoff, Mads Nielsen, and Andrew Ng, Multiscale Denois- ing Autoencoders for Breast Density Scoring, In STMI 2012, in conjunction with MICCAI 2012.
Kersten Petersen, Bayesian Image Segmentation with Multiscale Feature Learning, PhD thesis, University of Copenhagen, January 2012.
Kersten Petersen, Melanie Ganz, Peter Mysling, Mads Nielsen, Lene Lillemark, Alessan- dro Crimi, and Sami S. Brandt, A Bayesian Framework for Automated Cardiovascular Risk Scoring on Standard Lumbar Radiographs, IEEE Transactions on Medical Imaging, 31(3):663–676, 2012.
Peter Mysling, Kersten Petersen, Mads Nielsen, Lene Lillemark, Alessandro Crimi, and Sami S. Brandt, Automatic Segmentation of Vertebrae from Radiographs: A sample-driven active shape model approach, In Machine Learning in Medical Imaging, volume 7009 of LNCS, pages 10-17. Springer Berlin, Heidelberg, 2011.
Kersten Petersen, Mads Nielsen, and Sami S. Brandt, Conditional Point Distribution Mod- els, In Proc. MCV 2010, in conjunction with MICCAI 2010, Springer LNCS.
Kersten Petersen, Mads Nielsen, and Sami S. Brandt, A Static SMC Sampler on Shapes for the Automated Segmentation of Aortic Calcifications, In Proc. ECCV 2010, Springer LNCS.
Kersten Petersen, Janis Fehr, and Hans Burkhardt, Fast Generalized Belief Propagation for MAP estimation on 2D and 3D grid-like Markov Random Fields, In Proc. DAGM 2008, Springer LNCS, pp. 41-50. [DAGM Award]