Abdallah Zaid Alkilani

I am a PhD student in the Department of Electrical and Electronics Engineering and the National Magnetic Resonance Research Center (UMRAM) at Bilkent University, working under the supervision of Prof. Emine Ülkü Sarıtaş and Prof. Tolga Çukur. My research interests lie at the intersection of image processing and machine/deep learning, with a focus on medical imaging applications.

During my MS, my research centered on developing novel unsupervised deep learning methods to correct susceptibility artifacts in Echo Planar Imaging (EPI). I introduced FD-Net, an innovative forward-distortion network that integrates physics-based constraints to correct these artifacts without requiring supervised training data. I further extended this work by incorporating complex-valued processing via phase-injection techniques to better address signal dropout and pileup artifacts. Additionally, I adapted FD-Net for multi-echo blip-reversed EPI acquisitions, enhancing its robustness across different imaging conditions.

Beyond MRI artifact correction, I am also interested in broader applications of deep learning in medical imaging, particularly diffusion MRI. My research aims to develop practical, high-fidelity solutions that enhance clinical workflows while maintaining accuracy and efficiency in medical image processing.