Title | Multi-view learning for lymph node metastasis prediction using tumor and nodal radiomics in gastric cancer |
Authors | Yang, Jing Wang, Li Qin, Jiale Du, Jichen Ding, Mingchao Niu, Tianye Li, Rencang |
Affiliation | Peking Univ, Aerosp Ctr Hosp, Aerosp Sch Clin Med, Beijing 100049, Peoples R China Shenzhen Bay Lab, Shenzhen 518118, Peoples R China Zhejiang Univ, Womens Hosp, Sch Med, Hangzhou 310006, Zhejiang, Peoples R China Univ Texas Arlington, Dept Math, Arlington, TX 76019 USA Univ Texas Arlington, Dept Comp Sci & Engn, Arlington, TX 76019 USA |
Keywords | MULTIDETECTOR ROW CT PROGNOSTIC IMPACT IMAGES CLASSIFICATION INVOLVEMENT DISEASE SYSTEM |
Issue Date | 7-Mar-2022 |
Publisher | PHYSICS IN MEDICINE AND BIOLOGY |
Abstract | Purpose. This study aims to develop and validate a multi-view learning method by the combination of primary tumor radiomics and lymph node (LN) radiomics for the preoperative prediction of LN status in gastric cancer (GC). Methods. A total of 170 contrast-enhanced abdominal CT images from GC patients were enrolled in this retrospective study. After data preprocessing, two-step feature selection approach including Pearson correlation analysis and supervised feature selection method based on test-time budget (FSBudget) was performed to remove redundance of tumor and LN radiomics features respectively. Two types of discriminative features were then learned by an unsupervised multi-view partial least squares (UMvPLS) for a latent common space on which a logistic regression classifier is trained. Five repeated random hold-out experiments were employed. Results. On 20-dimensional latent common space, area under receiver operating characteristic curve (AUC), precision, accuracy, recall and F1-score are 0.9531 +/- 0.0183, 0.9260 +/- 0.0184, 0.9136 +/- 0.0174, 0.9468 +/- 0.0106 and 0.9362 +/- 0.0125 for the training cohort respectively, and 0.8984 +/- 0.0536, 0.8671 +/- 0.0489, 0.8500 +/- 0.0599, 0.9118 +/- 0.0550 and 0.8882 +/- 0.0440 for the validation cohort respectively (reported as mean +/- standard deviation). It shows a better discrimination capability than single-view methods, our previous method, and eight baseline methods. When the dimension was reduced to 2, the model not only has effective prediction performance, but also is convenient for data visualization. Conclusions. Our proposed method by integrating radiomics features of primary tumor and LN can be helpful in predicting lymph node metastasis in patients of GC. It shows multi-view learning has great potential for guiding the prognosis and treatment decision-making in GC. |
URI | http://hdl.handle.net/20.500.11897/638516 |
ISSN | 0031-9155 |
DOI | 10.1088/1361-6560/ac515b |
Indexed | SCI(E) |
Appears in Collections: | 北京航天中心医院 |