Cost-Effective Active Labeling for Data-Efficient Cervical Cell Classification

TL;DR

This paper introduces a cost-effective active labeling approach that reduces human effort while constructing representative datasets for cervical cell classification, improving data efficiency and classification accuracy.

Contribution

It proposes a novel active labeling algorithm that efficiently estimates uncertainty to select the most informative images, significantly lowering human labeling costs.

Findings

The method achieves high classification accuracy with fewer labeled samples.

Empirical results demonstrate reduced human labeling effort without sacrificing performance.

The approach effectively enhances dataset representativeness for cervical cell classification.

Abstract

Information on the number and category of cervical cells is crucial for the diagnosis of cervical cancer. However, existing classification methods capable of automatically measuring this information require the training dataset to be representative, which consumes an expensive or even unaffordable human cost. We herein propose active labeling that enables us to construct a representative training dataset using a much smaller human cost for data-efficient cervical cell classification. This cost-effective method efficiently leverages the classifier's uncertainty on the unlabeled cervical cell images to accurately select images that are most beneficial to label. With a fast estimation of the uncertainty, this new algorithm exhibits its validity and effectiveness in enhancing the representative ability of the constructed training dataset. The extensive empirical results confirm its efficacy again in navigating the usage of human cost, opening the avenue for data-efficient cervical cell classification.