Sijie Chen1, Wenjing Zhou2, Jinghui Tu1, Jian Li1, Bo Wang3,4, Xiaofei Mo3,4, Geng Tian3,4, Kebo Lv1, Zhijian Huang5. 1. Department of Mathematics, Ocean University of China, Qingdao, China. 2. Department of Oncology, Hiser Medical Center of Qingdao, Qingdao, China. 3. Qingdao Geneis Institute of Big Data Mining and Precision Medicine, Qingdao, China. 4. Geneis Beijing Co., Ltd., Beijing, China. 5. Department of Breast Surgical Oncology, Fujian Cancer Hospital & Fujian Medical University Cancer Hospital, Fuzhou, China.
Abstract
PURPOSE: Establish a suitable machine learning model to identify its primary lesions for primary metastatic tumors in an integrated learning approach, making it more accurate to improve primary lesions' diagnostic efficiency. METHODS: After deleting the features whose expression level is lower than the threshold, we use two methods to perform feature selection and use XGBoost for classification. After the optimal model is selected through 10-fold cross-validation, it is verified on an independent test set. RESULTS: Selecting features with around 800 genes for training, the R 2-score of a 10-fold CV of training data can reach 96.38%, and the R 2-score of test data can reach 83.3%. CONCLUSION: These findings suggest that by combining tumor data with machine learning methods, each cancer has its corresponding classification accuracy, which can be used to predict primary metastatic tumors' location. The machine-learning-based method can be used as an orthogonal diagnostic method to judge the machine learning model processing and clinical actual pathological conditions.
PURPOSE: Establish a suitable machine learning model to identify its primary lesions for primary metastatic tumors in an integrated learning approach, making it more accurate to improve primary lesions' diagnostic efficiency. METHODS: After deleting the features whose expression level is lower than the threshold, we use two methods to perform feature selection and use XGBoost for classification. After the optimal model is selected through 10-fold cross-validation, it is verified on an independent test set. RESULTS: Selecting features with around 800 genes for training, the R 2-score of a 10-fold CV of training data can reach 96.38%, and the R 2-score of test data can reach 83.3%. CONCLUSION: These findings suggest that by combining tumor data with machine learning methods, each cancer has its corresponding classification accuracy, which can be used to predict primary metastatic tumors' location. The machine-learning-based method can be used as an orthogonal diagnostic method to judge the machine learning model processing and clinical actual pathological conditions.
Authors: Sarah E Kerr; Catherine A Schnabel; Peggy S Sullivan; Yi Zhang; Veena Singh; Brittany Carey; Mark G Erlander; W Edward Highsmith; Sarah M Dry; Elena F Brachtel Journal: Clin Cancer Res Date: 2012-05-30 Impact factor: 12.531
Authors: Angela Hamblin; Sarah Wordsworth; Jilles M Fermont; Suzanne Page; Kulvinder Kaur; Carme Camps; Pamela Kaisaki; Avinash Gupta; Denis Talbot; Mark Middleton; Shirley Henderson; Anthony Cutts; Dimitrios V Vavoulis; Nick Housby; Ian Tomlinson; Jenny C Taylor; Anna Schuh Journal: PLoS Med Date: 2017-02-14 Impact factor: 11.069
Authors: Péter Mendik; Levente Dobronyi; Ferenc Hári; Csaba Kerepesi; Leonardo Maia-Moço; Donát Buszlai; Peter Csermely; Daniel V Veres Journal: Nucleic Acids Res Date: 2019-01-08 Impact factor: 16.971