Cell abundance aware deep learning for cell detection on highly imbalanced pathological data

TL;DR

This paper introduces a deep learning pipeline that incorporates cell abundance information to improve detection of rare cell types in imbalanced pathological data, demonstrating enhanced performance over baseline models.

Contribution

A novel deep learning method that uses cell abundance-based weighting to address class imbalance in pathological cell detection tasks.

Findings

Achieved a 2% higher F1-score than baseline models.

Outperformed baseline models in detecting rare cell types.

Scaling loss by cell abundance improves detection accuracy.

Abstract

Automated analysis of tissue sections allows a better understanding of disease biology and may reveal biomarkers that could guide prognosis or treatment selection. In digital pathology, less abundant cell types can be of biological significance, but their scarcity can result in biased and sub-optimal cell detection model. To minimize the effect of cell imbalance on cell detection, we proposed a deep learning pipeline that considers the abundance of cell types during model training. Cell weight images were generated, which assign larger weights to less abundant cells and used the weights to regularize Dice overlap loss function. The model was trained and evaluated on myeloma bone marrow trephine samples. Our model obtained a cell detection F1-score of 0.78, a 2% increase compared to baseline models, and it outperformed baseline models at detecting rare cell types. We found that scaling deep learning loss function by the abundance of cells improves cell detection performance. Our results demonstrate the importance of incorporating domain knowledge on deep learning methods for pathological data with class imbalance.