Format

Send to

Choose Destination
See comment in PubMed Commons below
Neuroinformatics. 2007 Summer;5(2):105-14.

OdorMapComparer: an application for quantitative analyses and comparisons of fMRI brain odor maps.

Author information

1
Center for Medical Informatics, Yale University School of Medicine, New Haven, CT 06520-8009, USA. nian.liu@yale.edu

Abstract

Brain odor maps are reconstructed flat images that describe the spatial activity patterns in the glomerular layer of the olfactory bulbs in animals exposed to different odor stimuli. We have developed a software application, OdorMapComparer, to carry out quantitative analyses and comparisons of the fMRI odor maps. This application is an open-source window program that first loads two odor map images being compared. It allows image transformations including scaling, flipping, rotating, and warping so that the two images can be appropriately aligned to each other. It performs simple subtraction, addition, and average of signals in the two images. It also provides comparative statistics including the normalized correlation (NC) and spatial correlation coefficient. Experimental studies showed that the rodent fMRI odor maps for aliphatic aldehydes displayed spatial activity patterns that are similar in gross outlines but somewhat different in specific subregions. Analyses with OdorMapComparer indicate that the similarity between odor maps decreases with increasing difference in the length of carbon chains. For example, the map of butanal is more closely related to that of pentanal (with a NC = 0.617) than to that of octanal (NC = 0.082), which is consistent with animal behavioral studies. The study also indicates that fMRI odor maps are statistically odor-specific and repeatable across both the intra- and intersubject trials. OdorMapComparer thus provides a tool for quantitative, statistical analyses and comparisons of fMRI odor maps in a fashion that is integrated with the overall odor mapping techniques.

PMID:
17873372
[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Loading ...
    Support Center