What type of study is this?

This is a Cohort Study study (also classified as: Quantitative Study, Experimental Study).

October 13, 2025Open Access

MIDOG 2025 Track 2: A Deep Learning Model for Classification of Atypical and Normal Mitotic Figures under Class and Hardness Imbalances

View Full Paper

Ask AI

Bookmark

Authors

A new method demonstrates high accuracy in classifying mitotic figures, addressing data imbalances and enhancing clinical diagnosis.

This approach achieves a mean balanced accuracy of 0.8744 in a 5-fold cross-validation of the dataset.
The model utilizes focal loss to effectively handle significant class imbalances common in histopathological data.
A comprehensive data augmentation pipeline is implemented to improve the model’s robustness against diverse tissue types.
Strong generalization is observed in preliminary leaderboard evaluations, enhancing clinical applicability.

Kotte et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark