Skip to main content
. 2022 Oct 25;247(22):2025–2037. doi: 10.1177/15353702221126560

Table 4.

Performance evaluation of the CLAM and MIL models on validation and test sets.

Model Valid set Test set
Accuracy (ACC) (%) Area under the ROC curve (AUC)
(%)
ACC
(%)
AUC
(%)
CLAM 85.57 95.19 87.04 95.06
MIL 81.16 90.13 82.21 90.54

CLAM: constrained attention multiple instance learning; MIL: multiple instance learning.