Neural networks for volumetric MR imaging of the brain

Published

Journal Article

Multimedia, including imaging and videoconferencing, is becoming the predominant mode of professional and technical communication. In the medical arena Multimedia Systems will have to deal with images of different types, including live videos of the participants in a video-conference, and still images or videos of radiological or MR (Magnetic Resonance) information. In the last thirty years there has been an explosion in our knowledge of the biochemical machinery of the nervous system. This knowledge has been primarily developed from invitro and invivo experiments in invertebrates and mammals. In the last few years with the advent of new imaging technologies such as Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) it has become possible to explore the integrated central nervous system (both biochemically and biophysically) in living humans. A major limitation in utilizing these techniques in an optimal fashion has been the lack of sophisticated image analysis systems which can extract the relevant information from the images in an automated or semiautomated manner. Brain MR images contain massive information requiring lengthy and complex interpretation (as in the identification of significant portions of the image), quantitative evaluation (as in the determination of the size of certain significant regions), and sophisticated interpretation (as in determining any image portions which indicate signs of lesions or of disease). In this paper we first survey the clinical and research needs for brain imaging. We discuss our recent work on the use of novel artificial neural networks which have a recurrent structure to extract precise morphometric information from MRI scans of the human brain. Experimental data using our novel approach is presented and suggestions are made for future research.

Duke Authors

Cited Authors

  • Gelenbe, E; Feng, Y; Ranga, K; Krishnan, R

Published Date

  • January 1, 1996

Published In

  • Proceedings of International Workshop on Neural Networks for Identification, Control, Robotics, and Signal/Image Processing, Nicrosp

Start / End Page

  • 194 - 202

Citation Source

  • Scopus