Efrat Shimron

Postdoctoral Fellow

Department of Electrical Engineering and Computer Sciences (EECS)

UC Berkeley

Efrat Shimron

About

I am a postdoc in the Department of Electrical Engineering and Computer Sciences (EECS) at UC Berkeley. I have the pleasure of being co-advised by Prof. Michael (Miki) Lustig (UC Berkeley) and Prof. Shreyas Vasanawala (Stanford). I work with their collaborative groups on developing MRI reconstruction methods for rapid dynamic MRI scans. My research leverages Machine Learning, physics-based modeling and computational imaging.

In 2015-2019 I was a PhD student at the Technion - Israel Institute of Technology. I spent joyful years in the lab of my PhD advisor Prof. Haim Azhari, where I developed sparsity-based methods for MR image reconstruction.

I am an advocate of reproducible research - all my papers have open-source code packages (see links below). I'm also the 2021-2023 Trainee Representative of the ISMRM Reproducible Research Study Group (RRSG) Governing Committee.

Research

New! This work was highlighted in a PNAS commentary article.

The paper was published in PNAS and covered in the Berkeley News and UT Austin News.

This work received a Magna cum Laude Award at the ISMRM 2021 Annual Meeting. Here is the git repo.

Here's a 20-min talk from the MRI Together international workshop (Dec 2021) and a 5-min talk from ISMRM'21 conference (this link requires registration to conference).

BladeNet: Dynamic Abdominal MRI using PROPELLER and Deep Learning

ISMRM 2022 abstract: #0684

New! My abstract was accepted for an oral presentation at the ISMRM conference in London: Wed, May 11, at 15:20, in room S11 (Breakout B) - New/Deep Machine Learning Techniques session.

This work introduces a new method for rapid dynamic (temporal) abdominal MRI with high spatio-temporal resolution and built-in motion correction.

Efrat Shimron - TED Compressed Sensing

BladeNet


Efrat Shimron - CORE-PI method

CORE-PI

CORE-PI is a parallel imaging method for MRI. It enables reconstruction from subsampled k-space data acquired with multi-coil arrays in 2D Cartesian scans. CORE-PI is parameter-free, i.e. you there's no to calibrate any hyperparamer in order to use it. Check out our code!

Efrat Shimron - CORE-Deblur

CORE-Deblur

CORE-Deblur is a Parallel Imaging Compressed Sensing (PI-CS) method.

Our paper introduced the concept of MR image reconstruction by deblurring using Compressed Sensing (CS), and showed that CORE-Deblur can expedite CS computations: the number of required iterations is reduced by a factor of 10.

News

Contact & social media: Email (efrat.s at berkeley dot edu), LinkedIn, Twitter

Software: GitHub

Collaborations


Students (mentees)

  • Adham Elarabawy

  • Han Cui

  • Frederic Wang

  • Jerry (Boyuan) Ma

  • Han Qi

  • Max Lister

  • Alma Harlangen

(c) website copyright Efrat Barak Shimron, 2022