Jump to: navigation, search

Harini Veeraraghavan

Harini Veeraraghavan is a researcher in the Biomedical Image Analysis Lab at General Electric Research since June 2008. At GE she applies her research for medical image analysis with focus on MRI, PET, and CT. She is interested in interactive learning approaches for segmentation and image analysis. Her current research interests include,

  • Multi-modal image analysis for volumetric image segmentation
  • Interactive and active learning approaches for robust segmentation and feature selection

Prior to joining GE, she was a postdoctoral researcher at Carnegie Mellon University where she developed learning by demonstration approaches applied to robotics with emphasis on learning from very small number of examples (in the order of one or few demonstrations). She got her Ph.D. from University of Minnesota, Twin Cities, where her research focused on image analysis and she applied machine learning techniques for learning and recognizing activities in videos specifically using very small number of labeled examples.

Under NAMIC, Harini has been working on developing segmentation tools in the Editor and developing image feature analysis libraries for 3D and 4D image analysis.

Current Work under NAMIC

  • Grow Cut image segmentation in Editor
  • Image Feature Libraries
    • Implemented features
      • Gabor
      • Image Entropy
      • FFT
      • Polynomial image features for dynamic time series 4D images