Performance testing of several classifiers for differentiating obstructive lung diseases based on texture analysis at high-resolution computerized tomography (HRCT)

This paper presents a study on the performance of several machine classifiers for differentiating obstructive lung diseases using texture analysis on various region of interest (ROI) sizes in 265 high-resolution computerized tomography (HRCT) images taken from 92 subjects. Four machine classifiers were implemented: naïve Bayesian classifier, Bayesian classifier, artificial neural net (ANN) and support vector machine (SVM). The SVM showed the best performance in overall accuracy for classifying obstructive lung diseases in ROI sizes of 32x32 and 64x64, while the naïve Bayesian method performed significantly worse than the other classifiers. The study highlights the importance of selecting an appropriate classification scheme for improving performance based on the characteristics of the data set. This research has implications for the development of AVIEW Lung Texture, as it demonstrates the effectiveness of machine classifiers for automating quantitative analysis and avoiding intra-inter-reader variability in the diagnosis of obstructive lung diseases using texture analysis.