A ℓ2, 1 norm regularized multi-kernel learning for false positive reduction in Lung nodule CAD. (March 2017)

• Record Type: Journal Article
• Authors: Cao, Peng; Liu, Xiaoli; Zhang, Jian; Li, Wei; Zhao, Dazhe; Huang, Min; Zaiane, Osmar
• Is Part Of: Computer methods and programs in biomedicine. Volume 140(2017)
• Page Start: 211
• Page End: 231
• Publication Date: 2017-03
• Subjects: Lung nodule detection -- False positive reduction -- Classification -- Heterogeneous feature fusion -- Multi-kernel learning; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2016.12.007
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: A novel l 2, 1 norm regularized multi-kernel learning is proposed for classification in lung nodule detection. We developed two efficient strategies to optimize the method to achieve the heterogeneous features subsets fusion and obtain the optimal kernel matrix. We have conducted extensive experiments to investigate our proposed methods. The proposed methods could be applied for the detection of many other potential lesions, such as mass, polyp, and microcalcification. Abstract: Objective: The aim of this paper is to describe a novel algorithm for False Positive Reduction in lung nodule Computer Aided Detection(CAD). Methods: In this paper, we describes a new CT lung CAD method which aims to detect solid nodules. Specially, we proposed a multi-kernel classifier with a ℓ2, 1 norm regularizer for heterogeneous feature fusion and selection from the feature subset level, and designed two efficient strategies to optimize the parameters of kernel weights in non-smooth ℓ2, 1 regularized multiple kernel learning algorithm. The first optimization algorithm adapts a proximal gradient method for solving the ℓ2, 1 norm of kernel weights, and use an accelerated method based on FISTA; the second one employs an iterative scheme based on an approximate gradient descent method. Results: The results demonstrates that the FISTA-style accelerated proximal descent method is efficient for the ℓ2, 1 norm formulation of multiple kernel learning with the theoretical guarantee of the convergence rate. Moreover, the experimental results demonstrate the effectiveness of the proposed methods in terms of Geometric mean (G-mean) and Area under the ROC curve (AUC), and significantly outperforms the competing methods. Conclusions: The proposed approach exhibits some remarkable advantages both in heterogeneous feature subsets fusion and classification phases. Compared with the fusion strategies of feature-level and decision level, the proposed ℓ2, 1 norm multi-kernel learning algorithm is able to accurately fuse the complementary and heterogeneous feature sets, and automatically prune the irrelevant and redundant feature subsets to form a more discriminative feature set, leading a promising classification performance. Moreover, the proposed algorithm consistently outperforms the comparable classification approaches in the literature.