Table 1.
Data Set | n | Age (SD) | Tumor Volume (cm3) | OS (IQR) | Censored (%) | OS < 6 m (%) |
---|---|---|---|---|---|---|
BraTS | 119 | 62 ± 12 | 30.09 ± 11.77 | 374 (364) | 0 | 21.8% (26) |
TCIA | 34 | 60.3 ± 10 | 42.55 ± 15.71 | 414 (482) | 5.9% (2) | 17.6% (6) |
Institution 1 | 22 | 65.3 ± 10 | 40.77 ± 18.12 | 451 (307) | 22.7% (5) | 13.6% (3) |
Institution 2 | 28 | 57.9 ± 13 | 42.14 ± 17.16 | 466 (217) | 32.1% (9) | 0 |
After random split (70/30) | ||||||
Training | 143 | 62.1 ± 11 | 40.62 ± 10.27 | 409 (311) | 7.7% (11) | 17.5% (25) |
Testing | 60 | 60.1 ± 13 | 39.27 ± 11.35 | 404 (392) | 8.3% (5) | 16.7% (10) |
Statistical comparison between cohorts |
t = 0.10, p = 0.273 | t = 0.12, p = 0.902 | U = 0.33, p = 0.743 | χ2 = 0.02, p = 0.877 | χ2 = 1.17, p = 0.279 |
BraTS = Brain Tumor Segmentation Challenge 2020, TCIA = The Cancer Image Archive, SD = standard deviation, IQR = interquartile range.