Cell Detection in Domain Shift Problem Using Pseudo-Cell-Position Heatmap

TL;DR

This paper introduces an unsupervised domain adaptation technique for cell detection that leverages pseudo-cell-position heatmaps and Bayesian confidence selection to improve performance across different data conditions.

Contribution

It presents a novel semi-supervised method using pseudo-cell heatmaps and Bayesian networks to adapt cell detection models across domains without labeled target data.

Findings

Outperforms existing domain adaptation methods in multiple domain combinations

Effectively refines pseudo-cell heatmaps to improve detection accuracy

Enables incremental domain extension with minimal supervision

Abstract

The domain shift problem is an important issue in automatic cell detection. A detection network trained with training data under a specific condition (source domain) may not work well in data under other conditions (target domain). We propose an unsupervised domain adaptation method for cell detection using the pseudo-cell-position heatmap, where a cell centroid becomes a peak with a Gaussian distribution in the map. In the prediction result for the target domain, even if a peak location is correct, the signal distribution around the peak often has anon-Gaussian shape. The pseudo-cell-position heatmap is re-generated using the peak positions in the predicted heatmap to have a clear Gaussian shape. Our method selects confident pseudo-cell-position heatmaps using a Bayesian network and adds them to the training data in the next iteration. The method can incrementally extend the domain from the source domain to the target domain in a semi-supervised manner. In the experiments using 8 combinations of domains, the proposed method outperformed the existing domain adaptation methods.