Ender Mete Ekşioğlu

Professor Ender Mete Ekşioğlu serves in the Department of Electrical and Electronics Engineering at Istanbul Technical University, where he has held academic positions since 1999. His current role as Professor began in 2019 following progression from Associate Professor (2013-2019) and Assistant Professor (2005-2012) ranks.

• PhD in Electronics and Communication Engineering, Istanbul Technical University (2000-2005)
• Degree from University of Michigan (1994-1997)
• Additional studies at University of Michigan (1997-1999)

Professor Ekşioğlu's research spans Machine Learning, Deep Learning, and advanced Signal Processing with emphasis on medical imaging applications. His work bridges traditional signal processing techniques with modern deep learning approaches, particularly in Magnetic Resonance Image Reconstruction where he applies sparsity principles and neural network architectures. Key focus areas include image denoising, segmentation, and enhancement across medical, underwater, and atmospheric imaging contexts.

His publication trends demonstrate consistent innovation in imaging inverse problems, with recent work integrating pixel-level processing with physical modeling in underwater imaging and developing topological awareness for medical image segmentation. The research shows strong continuity between traditional signal processing methods (like DCT) and cutting-edge deep learning architectures.

Professor Ekşioğlu has led six significant research projects funded by TUBITAK and BAP, including 'Deep Learning in Image Processing Inverse Problems' and 'Parallel magnetic resonance imaging techniques and applications'. These projects have resulted in substantial research output with applications spanning medical diagnostics to remote sensing.

• h-index of 13 based on Scopus citations
• 2000+ total research outputs
• Active supervision of 27 theses in progress

His laboratory work focuses on imaging inverse problems, with particular strength in MRI reconstruction techniques that combine deep learning with recursive algorithms. Current research directions include topological awareness in segmentation networks and hybrid approaches that integrate traditional signal processing with neural networks for improved robustness.