Batch Mode Active Learning on the Riemannian Manifold for Automated Scoring of Nuclear Pleomorphism in Breast Cancer

Highlights：

• Active Learning on nuclear pleomorphism scoring over the Riemannian manifold is explored.

• Adaptive Batch Mode Active Learning that identifies the apt batch size along with the batch of instances to be queried, following a submodular optimization framework is proposed.

• Samples for annotation are selected based on the kernelized Riemannian distance measures such as log-Euclidean metrics and the two Bregman divergences - Stein and Jeffrey divergences.

• A superior performance achieved, when compared with the state-of-the-art algorithms, as the information from the unlabeled samples are also exploited.