http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Saeed Heidary,Saeed Setayeshi,Mohammad Ghannadi-Maragheh 한국물리학회 2014 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.65 No.5
The aim of this study is to compare the adaptive neuro-fuzzy inference system (ANFIS) andthe artificial neural network (ANN) to estimate the cross-talk contamination of 99mTc / 201Tlimage acquisition in the 201Tl energy window (77 ± 15% keV). GATE (Geant4 Application inEmission and Tomography) is employed due to its ability to simulate multiple radioactive sourcesconcurrently. Two kinds of phantoms, including two digital and one physical phantom, are used. Inthe real and the simulation studies, data acquisition is carried out using eight energy windows. TheANN and the ANFIS are prepared in MATLAB, and the GATE results are used as a training dataset. Three indications are evaluated and compared. The ANFIS method yields better outcomes fortwo indications (Spearman’s rank correlation coefficient and contrast) and the two phantom resultsin each category. The maximum image biasing, which is the third indication, is found to be 6%more than that for the ANN.