Vazim Ibrahim

vazim PhD Scholar

  • Contact Details :
    • Email Address : vazim.i@iiitmk.ac.in
    • Phone no. (M) : +91-9497171540

  • Research Focus : Super-Resolution Recontruction in parallel MRI


Super-Resolution applications for echo-planar image reconstruction in parallel MRI.

Echo-planar imaging (EPI) is an ultrafast imaging technique in magnetic resonance imaging (MRI) capable of acquiring the entire k-space with at most one or two excitations. This requires long echo time, and therefore limits the number of samples. Although this leads to an increase in the voxel dimensions, the application of gradient reversals during acquisition of each line introduces inhomogeneities and intravoxel dephasing. The consequent loss of SNR and low resolution leads to the partial voluming effect. Also, the presence of field inhomogeneities creates echo shifts in k-space resulting in phase dispersion, additional signal losses and geometric distortions.

My primary research focus is to investigate the application of super-resolution variational frameworks in MR image reconstruction. The variational frameworks include reconstruction based models, convolutional neural networks, adversarial learning, variational networks and algorithm unrolling that contribute to the textural and edge details of the reconstructed image. This post processing method can actually reduce the MR aquisition time, which eventually reduces the burden in MR aquisition procedure.

Publication:

Conferences:
  • Vazim Ibrahim, Sumit Datta, Alex P. James and Joseph Suresh Paul."Interleaved Hybrid Domain Learning for Super-Resolution MRI." In 2022 IEEE International Symposium on Circuits and Systems (ISCAS) (pp. 3433-3437). 2022.

  • Sreekanth Madhusoodhanan, Vazim Ibrahim, Chandrasekharan Kesavadas and Joseph Suresh Paul."Enhancing GRE magnitude image using low-rank and sparsity analysis of structured matrix." Proceeding Proc GC Intl Soc Mag Reson Med. 2020.

  • Vazim Ibrahim and Joseph Suresh Paul. "Super-Resolution MRI Using Fractional order Kernel Regression and Total Variation." In the Proceeding on International Conference on Computational Intelligence, 2020.

Back