Table 1.
Number of cancer cases in the three datasets “core features,” “core features and TM,” and “all features” before (blue) and after (green) anonymization when including all eight classes vs. only including four classes Lung, Upper GI, Pancreas, and Liver.
Number of classes | Cases in “core features” dataset (nine features) | Cases in “core features and TM” dataset (11 features) | Cases in “all features” dataset (30 features) | |||
---|---|---|---|---|---|---|
Before anonymization | After anonymization | Before anonymization | After anonymization | Before anonymization | After anonymization | |
8 | 13,861 | 13,764 | 13,861 | 13,712 | 336 | 328 |
4 | 4,295 | 4,271 | 4,295 | 4,271 | 299 | 297 |