12967_2020_2297_MOESM1_ESM.docx (1.48 MB)
Download file

Additional file 1 of Glandular orientation and shape determined by computational pathology could identify aggressive tumor for early colon carcinoma: a triple-center study

Download (1.48 MB)
journal contribution
posted on 17.03.2020, 04:40 by Meng-Yao Ji, Lei Yuan, Shi-Min Lu, Meng-Ting Gao, Zhi Zeng, Na Zhan, Yi-Juan Ding, Zheng-Ru Liu, Ping-Xiao Huang, Cheng Lu, Wei-Guo Dong
Additional file 1: File S1. TMA Construction. File S2. Description of Gland Co-occurrence Morphological Feature Extraction. File S3. Immunohistochemistry. Figure S1. The workflow of patient selection. Figure S2. Kaplan–Meier curves of perineural invasion, vascular invasion on D2. Table S1. Summary of gland morphometric features. Table S2. A comprehensive list of all 797 quantitative features. Table S3. Patient characteristics of TCGA cohort. Table S4. The top 5 representative Feature and descriptions. Table S5. The performance of the classifiers on D2/D3. Table S6. Correlations between ECHBC and other major clinicopathologic features and disease recurrence on D2. Table S7. Comparative analysis of the image classifier and immunohistochemistry&CEA on D2. Table S8. The performances of the image classifier on TCGA cohort. Table S9. Multivariate survival analysis conducted on D4. Table S10. Ki67 and CEA Multivariate survival analysis conducted on D1/D2.


Natural Science Foundation of Hubei Province National Natural Science Foundation of China Innovation Seed Funding of Wuhan University Central Universities