Table 5

Prediction performance of machine-learning algorithms: 24-month survival

Area under the ROC curve (AUC)
Cancer typeEAR onlyECO onlyEAR+ECO
Genitourinary0.73 (0.69 to 0.78)0.84 (0.81 to 0.88)*0.86 (0.82 to 0.89)*,†
Colorectal0.76 (0.72 to 0.80)0.76 (0.72 to 0.80)0.76 (0.72 to 0.80)
Lung0.74 (0.69 to 0.78)0.78 (0.73 to 0.82)*0.82 (0.79 to 0.86)*,†
Breast0.67 (0.61 to 0.73)0.86 (0.82 to 0.90)*0.88 (0.84 to 0.92)*
Haematological0.73 (0.68 to 0.77)0.70 (0.66 to 0.75)0.80 (0.76 to 0.84)*,†
Upper gastrointestinal0.81 (0.77 to 0.85)0.77 (0.72 to 0.81)0.87 (0.83 to 0.9)*,†
Skin0.71 (0.65 to 0.76)0.85 (0.8 to 0.89)*0.94 (0.92 to 0.97)*,†
Head and neck0.74 (0.7 to 0.78)0.66 (0.51 to 0.61)0.71 (0.67 to 0.76)†
Gynaecological0.96 (0.94 to 0.99)0.99 (0.98 to 1)*0.97 (0.95 to 0.99)
CNS0.83 (0.78 to 0.89)0.87 (0.82 to 0.93)0.96 (0.93 to 0.99)*,†
Unknown primary0.74 (0.7 to 0.79)0.78 (0.74 to 0.82)*0.8 (0.76 to 0.84)*
  • *Significantly greater than EAR only.

  • †Significantly greater than ECO only.

  • CNS, central nervous system; EAR, electronic administrative records; ECO, Evaluation of Cancer Outcomes; ROC, receiver operating characteristic.