|
KMID : 1005420060080040241
|
|
Journal of Cerebrovascular and Endovascular Neurosurgery
2006 Volume.8 No. 4 p.241 ~ p.247
|
|
|
A New Method for Quantification of Ischemic Brain Lesion Using a Personal Computer and a Flatbed Scanner
|
|
Kim Tae-Yeon
Lee Hyung-Jin
Lee Jin-Seok
Yang Ji-Ho
Lee Il-Woo
|
|
|
Abstract
|
|
|
Objective: Quantification of infarct size has been frequently used as an useful index to assess the focal cerebral ischemia models and therapeutic strategies. Currently-used image analysis systems require dedicated and expensive equipment to carry out this task. The planimeter method is time-consuming and complex. The first purpose of this study was to develop a new, low-cost and simple method to perform the image acquisition and analysis using flatbed color scanner and a commercial image processing software (Adobe Photoshop 6.0). The second purpose was to evaluate the accuracy and reproducibility of the new method.
Methods: Fifty-nine adult male Sprague-Dawley rats were sacrificed 24 hours following the focal cerebral ischemia. Hematoxylin-Eosin stain was used as a conventional staining method to differentiate ischemic damage from healthy brain tissue. Digital images were captured from the stained coronal sections using a flatbed color scanner and analyzed with a commercial image processing software. To evaluate the accuracy of the new method, the data obtained from the new procedure was correlated with those from a dedicated standard image analysis system and planimeter method. To evaluate the reproducibility, interobservor correlation coefficients were estimated. The correlation in the outcome of infarct size measurements between two different examiners and three different methods were evaluated using the Pearson’s correlation coefficient and regression analysis. In each examiners, the data from three different methods were analyzed using nonparametric Freedman ANOVA. In each methods, the data from two different examiners were compared using Wilcoxon signed-rank test.
Results: The following results were obtained. 1. From the data taken by the two examiners, there was a close correlation in the outcome of infarct size measurement anong the new method, standard system method and planimeter method (P=0.0001). 2. There were no significant differences in the means of infarct size measurement among the three different methods in the results of examiner 1 and 2 (Freedman ANOVA, P>0.05). 3. A high agreement was obtained in measurement of the infarct size between two different examiners with the same source of samples. Using the Photoshop method, the correlation coefficient was 0.991 (P=0.0001). 4. In each methods, there were no significant differences in the means of infarct size measurement between two different examiners. Using the Photoshop method, results from examiner 1 and 2 were 27.33+/-4.18 mm2 and 27.35+/-4.14 mm2 respectively. Also there was no significant difference found by using Wilcoxon signed-rank test (P=0.95).
Conclusions: These data demonstrated that ischemic lesion of focal cerebral ischemia in rats can be accurately and reproducibly quantified using this method. The low-cost and simplicity may facilitate the application of this method in measurement of ischemic brain damage.
|
|
|
KEYWORD
|
|
|
Infarct size, Image analysis, Flatbed scanner, Image processing software
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
 
|
|