A robust point matching algorithm for autoradiograph alignment

Anand Rangarajan, Eric Mjolsness, Suguna Pappu, Lila Davachi, Patricia S. Goldman-Rakic, James S. Duncan

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Neuroimaging of the human brain has opened the way for a genuine understanding of human cognition; but the circuitry and cellular basis of the extraordinary information processing capacity of humans can be addressed only in experimental animals such as nonhuman primates by using the 2-DG autoradiographic method. This method requires sacrifice of the animal and sectioning of the brain into serial sections followed by production of autoradiographs of individual brain sections which are not in register. We have developed a new automated alignment method to reconstitute the autoradiographs. Our alignment method automatically finds the 2-D spatial mapping and the homologies between the slices and robustly accounts for the natural and artifactual differences by applying the powerful mechanism of outlier rejection adapted from the robust statistics literature.

Original languageEnglish (US)
Title of host publicationVisualization in Biomedical Computing - 4th International Conference, VBC 1996, Proceedings
EditorsKarl Heinz Hohne, Ron Kikinis
PublisherSpringer Verlag
Number of pages10
ISBN (Print)3540616497, 9783540616498
StatePublished - 1996
Event4th International Conference on Visualization in Biomedical Computing, VBC 1996 - Hamburg, Germany
Duration: Sep 22 1996Sep 25 1996

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other4th International Conference on Visualization in Biomedical Computing, VBC 1996

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'A robust point matching algorithm for autoradiograph alignment'. Together they form a unique fingerprint.

Cite this