Skip to main content

Table 1 Indication of the origin of the seven datasets that were used.

From: Pooling breast cancer datasets has a synergetic effect on classification performance and improves signature stability

     All ER ER pos ER neg
Publication: Label Survival Total Poor Good Poor Good Poor Good
Desmedt et al. [4] Des (D) DMFS 120 29 91 16 65 13 26
Minn et al. [22] Min (Mn) DMFS 62 21 41 9 30 12 11
Miller et al. [18] Mil (Ml) SOS 193 37 156 26 117 11 39
Pawitan et al. [23] Paw (P) SOS 142 22 120 14 99 8 21
Loi et al. [24] Loi (L) DMFS 120 28 92 21 71 7 21
Chin et al. [25] Chi (C) DMFS 86 23 63 14 50 9 13
Vijver et al. [3] Vij (V) DMFS 248 70 178 44 149 26 29
  1. The Survival column indicates the type of survival data that was used; DMFS Distant Metastasis Free Survival; SOS breast cancer Specific Overall Survival. Lastly, for each dataset the total number of samples is indicated along with the number of poor/good samples per ER subgroup.