About Me


I am a 3rd year PhD student at the University of Southern California working at the Magnetic Engineering Research Laboratory MREL and the Dynamic Imaging Science Center DISC. My hobbies include music production YouTube and playing tennis.

Research Interests

My research interests are in fast, low-latency real-time MRI reconstruction, both with and without data-driven (Machine Learning) methods. My work involves fast image reconstruction to support bio-feedback in the scanner in real-time. This includes interventional imaging and real-time speech therapy.

Our lab’s work is dedicated to dynamic imaging at “mid-field”, or 0.55T with high performance gradient systems. 0.55T loses in signal-to-noise ratio (SNR) compared to industy standard 1.5T and 3T systems, but benefits from reduced off-resonance, shorter T1, longer T2/T2*, reduced specific absorption rate (SAR), lower acoustic noise, and lower manufacture costs, allowing increased MRI accessibility.

I am also very interested in reproducible research, open science, and strive to make all my work and data as public as possible. To read more about MRI open science, click here.


In chronological order:

  1. S Ponrartana, HN Nguyen, SX Cui, Y Tian, P Kumar, JC Wood, KS Nayak. “Low-Field 0.55T MRI evaluation of the fetus.” Pediatric Radiology (2023). https://doi.org/10.1007/s00247-023-05604-x. PMID: 36882594.
  2. Y Lim, P Kumar, KS Nayak. “Speech Production RT-MRI at 0.55 Tesla.” Magnetic Resonance in Medicine (2023). Accepted.

Conference Abstracts (refereed)

In chronological order:

  1. P Kumar, Y Lim, KS Nayak. “Feasibility of Super Resolution Speech RT-MRI using Deep Learning.” International Society for Magnetic Resonance in Medicine 2021. https://archive.ismrm.org/2021/1782.html
  2. P Kumar, KS Nayak. “Low latency real-time mri at 0.55T using self-calibrating Through-Time GRAPPA.” International Society of Magnetic Resonance in Medicine 2022. https://archive.ismrm.org/2022/2348.html
  3. P Kumar, KS Nayak. “Real-time deep learning non-Cartesian image reconstruction using a causal variational network.”. Proc. ISMRM 31st Scientific Session, Toronto, June 2023, accepted.
  4. P Kumar, B Tasdelen, KS Nayak. “Open-source dynamic MRI workflow for reproducible research”. Proc. ISMRM 31st Scientific Session, Toronto, June 2023, accepted.
  5. J Wong, P Kumar, KS Nayak, Y Tian. “Artifact reduction for real-time spiral MRI using out-in sampling at 0.55T”. Proc. ISMRM 31st Scientific Session, Toronto, June 2023, accepted.
  6. C Hagedorn, P Kumar, B Villegas, M OuYoung, S Cui, M Sheth, S Narayanan, KS Nayak, U Sinha. “Role Of High-Performance Low Field Magnetic Resonance Imaging In Management Of Tongue Cancer”. Proc. AHNS 11th International Conference, Montreal, July 2023. https://ahns.jnabstracts.com/Detail?ID=129269


The IEEE Signal Processing society has partnered with USC Viterbi’s K-12 outreach program, and I have had the pleasure of leading the Medical Imaging portion of the creative team, where I taught several groups of high schoolers about signal processing, MRI physics, and showed them a guided tour of DISC’s MRI lab including a live human scan and some basic anatomy discussions. More details here.

I am a mentor for Minds Matter Southern California, a program that offers mentoring, support, group instruction, and social-emotional learning to low-income high school students in the Southern California region. The brilliant students and volunteers at Minds Matter create an environment of respect that allows the students to thrive and learn more about the college application process and go on high-school summer programs across the United States.