Table 1: Patient characteristics of TCGA cohort.

From: Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features

TCGA data set
CharacteristicsSummary
Lung adenocarcinoma patientsN=515
Age66.0±9.9 years
Gender46.3% Male; 53.7% female
Number of tumour histopathology image seriesN=831
Number of histopathology image series of adjacent benign tissueN=243
Number of histopathology image tilesN=5,739,972
  
Grade
 Grade 162 (12.0%)
 Grade 1–211 (2.14%)
 Grade 2180 (35.0%)
 Grade 2–339 (7.57%)
 Grade 3170 (33.0%)
 Grade 45 (0.97%)
 Grade unavailable48 (9.3%)
  
Stage
 Stage I254 (49.3%)
 Stage II119 (23.1%)
 Stage III81 (15.7%)
 Stage IV25 (4.9%)
 Stage unavailable36 (7.0%)
Lung squamous cell carcinoma patientsN=502
Age66.7±12.4 years
Gender74.1% Male; 25.9% female
Number of tumour histopathology image seriesN=761
Number of histopathology image series of adjacent benign tissueN=351
Number of histopathology image tilesN=5,033,634
  
Grade
 Grade 19 (1.79%)
 Grades 1–24 (0.80%)
 Grade 2198 (39.4%)
 Grades 2–334 (6.77%)
 Grade 3225 (44.8%)
 Grades 3–42 (0.40%)
 Grade 49 (1.79%)
 Grade unavailable21 (4.2%)
  
Stage
 Stage I242 (48.2%)
 Stage II156 (31.1%)
 Stage III87 (17.3%)
 Stage IV7 (1.4%)
 Stage unavailable10 (2.0%)
  1. Abbreviation: TCGA, The Cancer Genome Atlas.