Computer-aided diagnosis of lung nodule using gradient tree boosting and Bayesian optimization

• Published in: PloS one Vol. 13; no. 4; p. e0195875
• Main Authors: Nishio, Mizuho, Nishizawa, Mitsuo, Sugiyama, Osamu, Kojima, Ryosuke, Yakami, Masahiro, Kuroda, Tomohiro, Togashi, Kaori
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 19.04.2018; Public Library of Science (PLoS)
• Subjects: Algorithms; Archives & records; Artificial intelligence; Bayesian analysis; Biology and Life Sciences; Care and treatment; Classification; Computed tomography; Computer and Information Sciences; Data search; Diagnosis; Diagnostic systems; Drug dosages; Gene expression; Hospitals; Information processing; Learning algorithms; Lifestyles; Lung cancer; Lung nodules; Machine learning; Mathematical analysis; Medical diagnosis; Medical imaging; Medical imaging equipment; Medical screening; Medicine and Health Sciences; Methods; Nodules; Nuclear medicine; Optimization; Parameter estimation; Parameters; People and Places; Physical Sciences; Research and Analysis Methods; Support vector machines; University graduates; Japan
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0195875

We aimed to evaluate a computer-aided diagnosis (CADx) system for lung nodule classification focussing on (i) usefulness of the conventional CADx system (hand-crafted imaging feature + machine learning algorithm), (ii) comparison between support vector machine (SVM) and gradient tree boosting (XGBoost) as machine learning algorithms, and (iii) effectiveness of parameter optimization using Bayesian optimization and random search. Data on 99 lung nodules (62 lung cancers and 37 benign lung nodules) were included from public databases of CT images. A variant of the local binary pattern was used for calculating a feature vector. SVM or XGBoost was trained using the feature vector and its corresponding label. Tree Parzen Estimator (TPE) was used as Bayesian optimization for parameters of SVM and XGBoost. Random search was done for comparison with TPE. Leave-one-out cross-validation was used for optimizing and evaluating the performance of our CADx system. Performance was evaluated using area under the curve (AUC) of receiver operating characteristic analysis. AUC was calculated 10 times, and its average was obtained. The best averaged AUC of SVM and XGBoost was 0.850 and 0.896, respectively; both were obtained using TPE. XGBoost was generally superior to SVM. Optimal parameters for achieving high AUC were obtained with fewer numbers of trials when using TPE, compared with random search. Bayesian optimization of SVM and XGBoost parameters was more efficient than random search. Based on observer study, AUC values of two board-certified radiologists were 0.898 and 0.822. The results show that diagnostic accuracy of our CADx system was comparable to that of radiologists with respect to classifying lung nodules.