Prediction of aortic diameter values in healthy Turkish infants, children and adolescents via Adaptive Network Based Fuzzy Inference System

The aorta diameter size one of the cardiac value is very important to guess for child before adult age, due to growing up body. In conventional method, the experts use curve charts to decide whether their measured aortic diameter size is normal or not. Our proposed method presents a valid virtual aortic diameter result related to age, weight and sex. The proposed method comprises of two stages: (i) data normalization using a normalization method called Line Base Normalization Method (LBNM) that is firstly proposed by us, (ii) normalized aortic diameter prediction using Adaptive Network Based Fuzzy Inference Systems (ANFIS). Data set includes real Turkish infants, children and adolescents values and divided into two groups as 50% training -50% testing split of whole dataset to show performance of ANFIS. LBNM compared to three normalization methods including Min-Max normalization, Z-score, and decimal scaling methods. The results were compared to real aortic diameters values by expert with nine year experiences in medical area.