Meeting Banner
Abstract #3067

Automatic Selection of Arterial Input Function Using K-mean Cluster Algorithm

Tian-Yu Su 1 , Sheng-Min Huang 1 , Cheng-He Li 1 , Kung-Chu Ho 2 , and Fu-Nien Wang 1

1 Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan, 2 Nuclear Medicine, Chang Gung Memorial Hospital, Taoyuan, Taiwan

D2O perfusion has been first realized on rat model through 1H signal. However, a standard AIF of D2O has not been directly determined from MR image to date. In this study, k-means cluster analysis was employed as an automatic algorithm for determining the AIF of D2O on mice. The AIF curves we found present characteristics including curve shape similarity, high in peak, narrow width and regular shape, which implies that k-means provide a reliable AIF estimations. In summary, D2O AIF curves were successfully extracted from mouse brain images in this study through automatic algorithm.

This abstract and the presentation materials are available to members only; a login is required.

Join Here