معرفی
Mehmet Akcakaya is an Associate Professor in the Department of Electrical and Computer Engineering at the University of Minnesota. His research focuses on machine learning for medical imaging, especially MRI reconstruction using deep learning and compressed sensing. He leads NIH-funded projects and is accepting PhD students.
His research interests include physics-informed deep learning, compressed sensing, and MRI reconstruction. He develops algorithms for high-resolution, accelerated MRI with applications in cardiac and brain imaging. Key challenges addressed are reference-free reconstruction and robustness in inverse problems.
Recent work shows a trend toward self-supervised and unsupervised deep learning for MRI, with innovations in cycle-consistent learning and diffusion models for inverse problems. There is increasing emphasis on few-shot and zero-shot adaptation to handle limited training data.
Dr. Akcakaya leads the NIH-funded project Robust and Efficient Learning of High-Resolution Brain MRI Reconstruction and collaborates on the Center for Mesoscale Connectomics. He is accepting PhD students and mentors research in medical imaging and machine learning.
His work is conducted at the University of Minnesota's Center for Magnetic Resonance Research, leveraging interdisciplinary collaborations in biomedical engineering and neuroscience.
