| Title | Histologic subtype classification of non-small cell lung cancer using PET/CT images |
| Authors | Han, Yong Ma, Yuan Wu, Zhiyuan Zhang, Feng Zheng, Deqiang Liu, Xiangtong Tao, Lixin Liang, Zhigang Yang, Zhi Li, Xia Huang, Jian Guo, Xiuhua |
| Affiliation | Capital Med Univ, Sch Publ Hlth, Dept Epidemiol & Hlth Stat, Beijing, Peoples R China Capital Med Univ, Beijing Municipal Key Lab Clin Epidemiol, Beijing, Peoples R China Capital Med Univ, Dept Nucl Med, Xuanwu Hosp, Beijing, Peoples R China Peking Univ, Dept Nucl Med, Key Lab Carcinogenesis & Translat Res, Canc Hosp, Beijing, Peoples R China La Trobe Univ, Dept Math & Stat, Melbourne, Vic, Australia Univ Coll Cork, Sch Math Sci, Cork, Ireland |
| Keywords | F-18-FDG PET/CT MANAGEMENT CLASSIFIERS RADIOMICS |
| Issue Date | Aug-2020 |
| Publisher | EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING |
| Abstract | Purposes To evaluate the capability of PET/CT images for differentiating the histologic subtypes of non-small cell lung cancer (NSCLC) and to identify the optimal model from radiomics-based machine learning/deep learning algorithms. Methods In this study, 867 patients with adenocarcinoma (ADC) and 552 patients with squamous cell carcinoma (SCC) were retrospectively analysed. A stratified random sample of 283 patients (20%) was used as the testing set (173 ADC and 110 SCC); the remaining data were used as the training set. A total of 688 features were extracted from each outlined tumour region. Ten feature selection techniques, ten machine learning (ML) models and the VGG16 deep learning (DL) algorithm were evaluated to construct an optimal classification model for the differential diagnosis of ADC and SCC. Tenfold cross-validation and grid search technique were employed to evaluate and optimize the model hyperparameters on the training dataset. The area under the receiver operating characteristic curve (AUROC), accuracy, precision, sensitivity and specificity was used to evaluate the performance of the models on the test dataset. Results Fifty top-ranked subset features were selected by each feature selection technique for classification. The linear discriminant analysis (LDA) (AUROC, 0.863; accuracy, 0.794) and support vector machine (SVM) (AUROC, 0.863; accuracy, 0.792) classifiers, both of which coupled with thel(2,1)NR feature selection method, achieved optimal performance. The random forest (RF) classifier (AUROC, 0.824; accuracy, 0.775) andl(2,1)NR feature selection method (AUROC, 0.815; accuracy, 0.764) showed excellent average performance among the classifiers and feature selection methods employed in our study, respectively. Furthermore, the VGG16 DL algorithm (AUROC, 0.903; accuracy, 0.841) outperformed all conventional machine learning methods in combination with radiomics. Conclusion Employing radiomic machine learning/deep learning algorithms could help radiologists to differentiate the histologic subtypes of NSCLC via PET/CT images. |
| URI | http://hdl.handle.net/20.500.11897/607396 |
| ISSN | 1619-7070 |
| DOI | 10.1007/s00259-020-04771-5 |
| Indexed | SCI(E) |
| Appears in Collections: | 北京肿瘤医院 |
