An Performance Analysis of Thyroid Tumor Detection using CANFES Classification with Optimized Features

14 Pages Posted: 1 Aug 2019

See all articles by Shankarlal B

Shankarlal B

Annamalai University

Sathya P.D.

Annamalai University

Date Written: August 1, 2019

Abstract

The human body metabolism is maintained by proper segregation of thyroid hormones from thyroid gland. The abnormal cells are formed in this thyroid gland due to improper food habits and genes from forefathers. These abnormal cells lead to the development of tumor regions in the thyroid gland. In this paper, a tumor region in thyroid ultra sound image is detected and segmented using Co-Active Adaptive Neuro Fuzzy Expert System (CANFES) classification framework with its improved performance. The proposed framework is split into three main modules as Enhancement, Gabor transform, CANFES classification trained by feature extraction process with tumor segmentation method. The extracted features are optimized using Genetic Algorithm (GA). The CANFES classification with GA improves approximately 2.5% of classification rate when compared with CANFES classification without GA This proposed system achieves 97.7% of sensitivity, 99.8% of specificity and 99.1% of accuracy.

Keywords: thyroid gland, tumor, classifications, features, optimization

Suggested Citation

B, Shankarlal and P.D., Sathya, An Performance Analysis of Thyroid Tumor Detection using CANFES Classification with Optimized Features (August 1, 2019). Proceedings of International Conference on Recent Trends in Computing, Communication & Networking Technologies (ICRTCCNT) 2019, Available at SSRN: https://ssrn.com/abstract=3430192 or http://dx.doi.org/10.2139/ssrn.3430192

Shankarlal B (Contact Author)

Annamalai University ( email )

Chidambaram
TN
India

Sathya P.D.

Annamalai University ( email )

TN
India

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
71
Abstract Views
439
Rank
589,797
PlumX Metrics