HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Bruyant, P. P. Articles by Mallet, J.-J. Search for Related Content PubMed PubMed Citation Articles by Bruyant, P. P. Articles by Mallet, J.-J.

Noise Removal Using Factor Analysis of Dynamic Structures: Application to Cardiac Gated Studies

Nuclear Spectroscopy and Image Processing Research Group, Biophysics Laboratory, Claude Bernard University, Lyon, France

Correspondence: For correspondence or reprints contact: P. Bruyant, PhD, Biophysics Laboratory 10, avenue Rockefeller 69373, Lyon cedex 08, France.

ABSTRACT

Factor analysis of dynamic structures (FADS) facilitates the extraction of relevant data, usually with physiologic meaning, from a dynamic set of images. The result of this process is a set of factor images and curves plus some residual activity. The set of factor images and curves can be used to retrieve the original data with reduced noise using an inverse factor analysis process (iFADS). This improvement in image quality is expected because the inverse process does not use the residual activity, assumed to be made of noise. The goal of this work is to quantitate and assess the efficiency of this method on gated cardiac images. Methods: A computer simulation of a planar cardiac gated study was performed. The simulated images were added with noise and processed by the FADS-iFADS program. The signal-to-noise ratios (SNRs) were compared between original and processed data. Planar gated cardiac studies from 10 patients were tested. The data processed by FADS-iFADS were subtracted to the original data. The result of the substraction was studied to evaluate its noisy nature. Results: The SNR is about five times greater after the FADS-iFADS process. The difference between original and processe data is noise only, i.e., processed data equals original data minus some white noise. Conclusion: The FADS-iFADS process is successful in the removal of an important part of the noise and therefore is a tool to improve the image quality of cardiac images. This tool does not decrease the spatial resolution (compared with smoothing filters) and does not lose details (compared with frequential filters). Once the number of factors is chosen, this method is not operator dependent.

Key Words: factor analysis of dynamic structures • image processing • noise • gated cardiac images