KMID : 0861420020070020179
|
|
Korean Journal of Nuclear Medicine Technology 2002 Volume.7 No. 2 p.179 ~ p.188
|
|
Reliability Estimation of Thyroid Uptake Probe using Regression Analysis
|
|
Kim Byung-Jin
Cho Seong-Wook Kim Jin-Eui
|
|
Abstract
|
|
|
Purpose : The aim of this study is to correct the count loss of Thyroid uptake probe with activity increase and estimate the reliability on the linearity using regression analysis.
Materials & Methods : We measured count with altered activity using thyroid uptake probe and used regression analysis to display the distribution of measured count statistically. We described functionally the expected true count that would be distributed the linearity and the measured count that was distributed the non-linearity using trend line. Regression analysis using the bias range of count that was taken place between the trend lines made the correction factor. To confirm the accuracy of correction factor, we also estimated the reliability of the linearity using regression analysis. We used the residual analysis to determine the compatibility of regression analysis.
Results : Measured count (0~1000 ¥ìCi) had the count loss with the activity increase. When we applied the correction factor to the measured count. We couldn¡¯t use the correction factor in the below 100 ¥ìCi and above 900 ¥ìCi. Because the corrected count was out of expected true count. Being corrected, however, measured count (100~900 ¥ìCi) had equal slope with that of expected true count. Also the rectilinear equation was indicated a high reliability (99.96%) in the result of regression analysis and was only displayed the variable error below ¡¾1 in the result of residual analysis.
Conclusion : As we can expect more correct count statically, correction factor that was applied to regression analysis can improve the reliability of the linearity.
|
|
KEYWORD
|
|
Regression Analysis, Residual Analysis, Trend Line, Reliability
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|