Phd. Student in Deep Learning for Medical Images

Phd. Student in Deep Learning for Medical Images

Dr. İlkay Öksüz

We offer a funded PhD. scholarships on Medical Image Acquisition and Reconstruction at the National Magnetic Resonance Center. The position will be based in Bilkent University National Magnetic Resonance Research Center (UMRAM) under supervision of Prof. Dr. Tolga Çukur. The position holder will be in close collaboration with Dr. İlkay Öksüz and Istanbul Technical University.

Medical Image Analysis
Several recent advances have led to more efficient ways in which MRI can be reconstructed. The most prominent example of these is compressed sensing MRI, which allows the reconstruction of MRI from under-sampled k-space data. The reconstruction of the MRI requires solving a non-linear optimization problem. For medical image reconstruction, machine learning can play a key role. The project aims to exploit recent advances in machine learning, in particular in the area of representation learning to learn representations and priors directly from raw data that are optimal for recovery of images (or even for the reconstruction of clinically useful information). The key tasks for the project are:
1. Acquisition of raw cardiac MR data
2. Reconstruction of cardiac MRI from under-sampled k-space data.
3. Development of deep learning based reconstruction algorithms leveraging temporal dependencies in cardiac MRI

Funding & Details
PhD positions are fully funded (4500TL/month) by The Scientific and Technological Research Council of Turkey (TÜBİTAK). The prospective students will be able to attend to prestigious international conferences and spend some time in top international institutes. The candidates must be enrolled in a graduate program in Turkey and can start to work immediately. We encourage applications from underrepresented groups.

Required Skills
We are looking for highly motivated students with an academic background in Physics, Computer Science or Engineering. The following skills are required:
• Experience with Python and deep learning libraries (e.g. Pytorch, Tensorflow or Keras).
• Motivation in machine learning and computer vision problems/algorithms.
• Enthusiasm to work on clinical challenges and knowledge on MR Physics.
• Good academic writing skills and English communication skills.


How to Apply
Please send an email to by attaching your CV, transcript, and a brief description explaining why you want to be considered (max 1 page).