Title: Performance comparison of various machine learning classifiers using fusion of LBP, intensity and GLCM feature extraction techniques for thyroid nodules classification

Authors: Rajshree Srivastava; Pardeep Kumar

Addresses: Department of Computer Science and Engineering, Jaypee University of Information Technology, Waknaghat, Solan, Himachal Pradesh, India ' Department of Computer Science and Engineering, Jaypee University of Information Technology, Waknaghat, Solan, Himachal Pradesh, India

Abstract: Machine Learning (ML) and feature extraction techniques have shown a great potential in medical imaging field. This work presents an effective approach for the identification and classification of thyroid nodules. In the proposed model, various features are extracted using Grey Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP) and intensity-based matrix. These features are fed to various ML classifiers like K-Nearest Neighbour (KNN), Decision-Tree (DT), Artificial Neural Network (ANN), Naïve Bayes, Extreme Gradient Boosting (XGBoost), Random Forest (RF), Linear Regression (LR) and Support Vector Machine (SVM). From the result analysis, it can be observed that proposed Model-4 has performed better in comparison with the rest of seven proposed models with the reported literature. An improvement of 4% to 5% is seen in performance evaluation of model in comparison with reported literature.

Keywords: machine learning; LBP; GLCM; intensity; noise removal; feature extraction.

DOI: 10.1504/IJGUC.2024.136708

International Journal of Grid and Utility Computing, 2024 Vol.15 No.1, pp.84 - 96

Received: 12 Dec 2022
Accepted: 25 Feb 2023
Published online: 19 Feb 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article