Premium
FACTORS INFLUENCING THE DISPERSION OF INDICATOR IN DYE‐DILUTION CURVES IN NORMAL MAN
Author(s) -
Thorburn Geoffrey D
Publication year - 1964
Publication title -
australian journal of experimental biology and medical science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.999
H-Index - 104
eISSN - 1440-1711
pISSN - 0004-945X
DOI - 10.1038/icb.1964.51
Subject(s) - dispersion (optics) , statistics , asymmetry , residual , indicator dilution , regression analysis , volume (thermodynamics) , linear regression , dilution , regression , mathematics , sampling (signal processing) , coefficient of variation , optics , thermodynamics , physics , algorithm , quantum mechanics , cardiac output , detector , blood pressure , medicine , radiology
Summary The main factors influencing the dispersion of indicator in man have been found to be the flow through the system and the volume between the injection and sampling sites. For a given flow and volume, alteration in pathlength of a system leads to systematic differences in the dispersion of indicator and to variation in the degree of asymmetry of the resultant curves. A considerable proportion of the residual variation in the dispersion of indicator after fitting the regression function for flow and volume can be explained by differences in pathlength either in the one individual or between individuals. By using a fixed anatomical injection and sampling site it was possible to reduce the variation in pathlength within an individual. In addition, the variation in pathlength between individuals was reduced by the introduction of a measure of asymmetry as a third independent variable in the multiple regression equations. The probit ratio b 1 /b 2 , which was used for this purpose, was shown to be related to the pathlength of the system. These results are discussed in relation to the application of the Korner‐Shillingford technique to the estimation of valvular incompetence in man.