Cell classification with worse-case boosting for intelligent cervical cancer screening

Youyi Song, Jing Zou, Kup-Sze Choi, Baiying Lei, Jing Qin

Abstract: Highlights•A new training algorithm for effectively learning from underrepresented data.•Boosting the learning performance on underrepresented data in cervical cell classification.•More training iterations and a large loss weight are dynamically assigned to worse-case data.•The classification performance on underrepresented data is significantly improved.