Stathis Hadjidemetriou received a B.Eng (Honors) in Electrical Engineering from McGill University in Montreal and an M.Sc. in Electrical Engineering from Columbia University in the City of New York. He received his Ph.D. in computer science also from Columbia University. His thesis was in the fields of image analysis and computer vision.
As post-doctoral fellow at Yale University and as research scientist at University hospitals he specialized in bio-medical image analysis. He has worked on image analysis for biological microscopy related to cancer. He has also worked on organ level biomedical MRI for neurological conditions of the brain as well as for extending the use of quantitative MRI for the whole body.
The methodologies he developed have been on the reconstruction and restoration of MRI data to remove imaging artifacts, the registration of multi-contrast MRI data, as well as on image analysis. The image analysis has included segmentation, motion tracking both for microscopy and torso MRI, as well as for MRI atlas computation.
He is currently associate professor at the Applied Information Technologies (AIT) program at CIIM. He continues the development and application of methodology both for image analysis as well as for more general aspects of artificial intelligence.
He has published numerous research papers in scientific journals and presented his work in numerous scientific conferences. He serves as a reviewer for various international conferences and journals.
He has extensive teaching experience in computer science, electrical engineering and biomedical engineering.
The research interests of Stathis Hadjidemetriou are on image understanding, mainly for biomedical imaging. This includes biological microscopy and organ level biomedical MRI. The methodology he uses is a combination of analytical image processing together with more recently developed methodology from the context of machine learning and neural networks for big data. The latter methodology he also applies in a more general context to the understanding of various types of social and financial data.
A research project in biomedical microscopy is the analysis of phase contrast cellular video sequences to track and parametrize the motion and the division of cells. The motion of cells is used to characterize migration for cancer metastasis. Similarly, cellular division is used to characterize cancer growth. The methodology first involves the reconstruction of the phase contrast data. Then, it involves the motion tracking of individual cells and the use of morphology to detect division.
Another work is the analysis of video sequences of fluorescence microscopy of frog embryos to computationally follow embryogenesis. The objective is to characterize and understand the causes of embryonic malformations. The tissue growth is represented with smooth vector fields. Some key features of these fields are then extracted for their biological significance.
His research also includes the analysis of organ level MRI data. The first step is the restoration of images for intensity non-uniformities that are very common in MRI. Then, is the intra-patient or inter-patient registration of different images. The registered images can be used to construct an atlas of anatomy or pathology. An example has been the computation of a whole-body MRI atlas for the detection of metastasis. These processing steps are implemented with non-parametric methodology.
He further develops the methodology to incorporate deep learning and neural networks in the context of image analysis for big data. This is necessary to address current needs to restore and analyze large amounts of data.
He also applies machine learning and neural networks to various general social problems. Some examples have been natural language understanding, financial data, and social data. Another objective of his work is to increase the accessibility of the processing tools by making them available on demand over the cloud.
Brain Immunoinformatics: A Symmetrical Link between Informatics, Wet Lab and the Clinic
I Papageorgiou, D Bittner, MN Psychogios, S Hadjidemetriou
Symmetry, 2021, 13 (11), 2168
Detection and tracking of cell divisions in phase contrast video microscopy
S Hadjidemetriou, B Gabrielli, T Pike, F Stevens, K Mele, P Vallotton
Proceeding of the Third MICCAI International Workshop on Microscopic Image Analysis with Applications in Biology, 2008
Motion tracking of the outer tips of microtubules
S Hadjidemetriou, D Toomre, J Duncan
Medical image analysis, 2008, 12 (6), 689-702
Non-Parametric Bayesian Estimation of Rigid Registration for Multi-Contrast Data in Big Data Analysis
S Hadjidemetriou, I Papageorgiou
Big Data in Multimodal Medical Imaging, 2019, 169-191
Non-parametric Bayesian registration (NParBR) of body tumors in DCE-MRI data
D Pilutti, M Strumia, M Büchert, S Hadjidemetriou
IEEE Transactions on Medical Imaging, 2015, 35 (4), 1025-1035
Volumetric analysis of MRI data monitoring the treatment of polycystic kidney disease in a mouse model
S Hadjidemetriou, W Reichardt, J Hennig, M Buechert, D von Elverfeldt
Magnetic Resonance Materials in Physics, Biology and Medicine, 2011, 24 (2), 109-119
Image processing and analysis
Multiresolution histograms and their use for recognition
E Hadjidemetriou, MD Grossberg, SK Nayar
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26 (7), 831-847
Histogram preserving image transformations
E Hadjidemetriou, MD Grossberg, SK Nayar
International Journal of Computer Vision, 2001, 45 (1), 5-23
Appearance matching with partial data
E Hadjidemetriou, SK Nayar
Proceedings of DARPA Image Understanding Workshop, Monterey, CA, 1998