A fuzzy shape representation of a segmented vessel tree and kernel-induced random forest classifier for the efficient prediction of lung cancer

• Published in: The Journal of supercomputing Vol. 76; no. 8; pp. 5801 - 5824
• Main Authors: Deepa, P., Suganthi, M.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2020; Springer Nature B.V
• Subjects: Classifiers; Compilers; Computer Science; Decision support systems; Decision trees; Fuzzy logic; Image filters; Image segmentation; Interpreters; Kernels; Lung cancer; Nodules; Noise reduction; Processor Architectures; Programming Languages; Regularization; Vessels; Contourlet filter; Kernel-induced random forest technique; Total variation regularisation; Fuzzy shape representation; Lung nodules; EM algorithm
• ISSN: 0920-8542; 1573-0484
• DOI: 10.1007/s11227-019-03002-5

An intelligent clinical decision support system is proposed classifying lung nodules for lung cancer prediction using a kernel-induced random forest classifier. A contourlet filter is used for image denoising. Fuzzy logic is used to represent the segmented shape of a vessel tree. The fuzzy shape of the vessel tree is then given to a classifier as a feature for learning. A hybridization of expected maximization and total variation regularisation is proposed for the vessel tree segmentation. The proposed use of a fuzzy shape vessel tree and kernel-induced random forest classifier promises to be an efficient method of detecting lung nodules for cancer diagnosis. The proposed system is evaluated for precision, recall and accuracy in comparison with many previously available techniques.