Image Understanding by Adaptive Networks of Goal Seeking Neurons

Abstract

We interfaced a two-dimensional (2-D) digitizing tablet to one of the serial ports and wrote suitable programs to allow digitization and inputting of brain sections to the computer and disk. Each brain was stained with horseradish peroxidase (HRP) and consisted of about 150 sections. We also wrote programs to allow the computer to generate three-dimensional reconstructions of each brain (3-D). The SUN workstation software package called SUNCORE allows us to view brains reconstructed in 3-D from any point of view and perspective. That is, on the color monitor we can rotate, translate, and scale on any of the x, y, z axes the reconstructed brain in minutes. These newly acquired capabilities allow us to evidence cortical and subcortical regions that feed information to the areas we analyze with microelectrodes to study the functional responses of single neurons. At this writing we have reconstructed three brains. A first impression is that numerous cortical areas send terminals to a recording site; that is, there seem to be substantive interconnections between somato-sensory, Claire-Bishop, insular, and visual 1, 2, 3. We have not done any recordings or HRP injections in the insular cortex areas 18 and 19. All our recordings were from somato-sensory, area 17, and Claire-Bishop in the cortex. Subcortically we recorded only from the hypothalamus. In all these regions we found adaptation to varying degrees. Keywords: Brain function.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 20, 1985
Accession Number
ADA217197

Entities

People

  • D. N. Spinelli

Organizations

  • University of Massachusetts Amherst

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Anatomy
  • Brain
  • Cells
  • Computer Architecture
  • Computer Programs
  • Computer Simulations
  • Computers
  • Education
  • Information Processing
  • Learning
  • Neurons
  • Neurosciences
  • Serial Ports
  • Three Dimensional
  • Training
  • Two Dimensional

Readers

  • Computer Science/Computer Engineering/Data Science/Digital Signal Processing.
  • Electrochemical Surface Science
  • Neuroscience