MD Anderson Cancer Center, (AUG 2011 -- DEC 2014)

  • Inverse problems in signals and image processing, computer vision. Sparse decomposition of signals.
  • Implement PET daily Quality Control with Java
  • Compressive sensing in PET imaging:   Most positron emission tomography/computed tomography (PET/CT) scanners consist of tightly packed, discrete detector rings to improve scanner efficiency. Our aim was to decrease the detector elements per ring (introducing gaps) while maintaining image quality by employing compressive sensing (CS) techniques in PET imaging. This makes scanner be cheaper, more accessible to physicians and patients and make scan time shorter

RICE, (2009--2014)

  • A Compact Five Neuron Model of the Mammalian Respiratory Central Pattern Generator: Model Validity and Phasic Sensitivity
  • Hyperspectral data reconstruction combing spatial and spectral sparsity

IHDM - Information and Signal Processing Lab, Polytechnic University (2004--2006)

    • Incorporated image processing to quality control of cylinder block dies
    • Support vector machines (SVM) and related methods,  detection of faces in images, dimensionality reduction