Thyroid Volume Estimation Analysis with Neural and Fuzzy-Neuro Techniques - A Comparative Study

Abstract

This paper presents a comparative study of various methods that used to identifies the thyroid volume from the ultrasound images. The Radial basis function neural network method, and by the hybrid structure of neural network and the fuzzy logic method. The performance of the algorithms Active Contour without edges, Localized region Based active contour and Distance Regularized Level Set will improve the results by 80%, the hybrid structure of the neurofuzzy method was improved from 81% to 92.9% by using Fuzzy-RBF, Fuzzy-MLP, and Fuzzy-CSFNN methods.The result shows that the Fuzzy Neuro method gives the highest accuracy level the by using the neural network method for detecting the thyroid volume from the segmentation process.

Automated segmentation using PCA and area estimation of thyroid gland using ultrasound images

“…The segmentation process will be applied for the input image for the detection of gland region [13]. The ROI based clustering is used here for effective segmentation to diagnose the tumour part.…”

Thyroid disease diagnosis: A survey

“…A comparative study of various methods to identify the thyroid volume from the US images -the Radial Basis Function (RBF) NN method; the hybrid structure of NN and the fuzzy logic method is presented in [12]. A neural network using Back Propagation (BP) was developed in [15] to diagnose thyroid problems.…”