TY - JOUR
T1 - Quality of treatment planning evaluation for head and neck cancer using artificial neural networks intelligence system
AU - Lee, T.-F.
AU - Liao, T.-I.
AU - Chao, P.-J.
AU - Ting, H.-M.
AU - Wu, J.-M.
AU - Chen, W.-P.
AU - Shie, C.-S.
AU - Liao, Kuo-Chen
AU - Fang, F.-M.
PY - 2013
Y1 - 2013
N2 - Three types of artificial neural networks (ANNs) are instructed by three different training algorithms to effectively evaluate the quality of the Head and Neck cancer (HN) treatment plans. One hundred sets of HN treatment plans are collected to be the input data of the neural networks. Three ANNs including Elman (ANN), feed-forward (ANN), and pattern recognition (ANN) were trained by using three different models, i.e., leave-one-out (Train), random selection (Train), and user defined (Train) method. The conformal index (CI) and homogeneity index (HI) are used to be the feature values and to train the neurons. The networks with higher accuracy are ANN (93.65±3.60)%, ANN (88.05±5.84)%, and ANN (87.55±5.86)%, respectively. The ROC curves show that the ANN approach has the highest sensitivity, which is 99%. It can be concluded that ANN is a better choice for evaluating the quality of treatment plans for HN, this method reduces the amount of trail-and-error during the iterative process of generating inverse treatment plans. © 2013 American Scientific Publishers All rights reserved.
AB - Three types of artificial neural networks (ANNs) are instructed by three different training algorithms to effectively evaluate the quality of the Head and Neck cancer (HN) treatment plans. One hundred sets of HN treatment plans are collected to be the input data of the neural networks. Three ANNs including Elman (ANN), feed-forward (ANN), and pattern recognition (ANN) were trained by using three different models, i.e., leave-one-out (Train), random selection (Train), and user defined (Train) method. The conformal index (CI) and homogeneity index (HI) are used to be the feature values and to train the neurons. The networks with higher accuracy are ANN (93.65±3.60)%, ANN (88.05±5.84)%, and ANN (87.55±5.86)%, respectively. The ROC curves show that the ANN approach has the highest sensitivity, which is 99%. It can be concluded that ANN is a better choice for evaluating the quality of treatment plans for HN, this method reduces the amount of trail-and-error during the iterative process of generating inverse treatment plans. © 2013 American Scientific Publishers All rights reserved.
KW - Artificial neural networks (ANNs)
KW - Conformal index (CI)
KW - Dose-volume histogram (DVH)
KW - Head and neck cancer (HN)
KW - Homogeneity index (HI)
U2 - 10.1166/asl.2013.5140
DO - 10.1166/asl.2013.5140
M3 - Journal Article
SN - 1936-6612
VL - 19
SP - 3236
EP - 3243
JO - Advanced Science Letters
JF - Advanced Science Letters
IS - 11
ER -