Negative Pseudo Labeling using Class Proportion for Semantic Segmentation in Pathology

TL;DR

This paper introduces a weakly-supervised cell tracking method that trains CNNs using only cell detection annotations, achieving performance comparable to fully supervised methods by analyzing co-detection outputs.

Contribution

It presents a novel backward-and-forward propagation technique to infer cell associations from weakly-supervised detection CNNs in pathology images.

Findings

Achieves near state-of-the-art performance with weak supervision

Successfully infers cell associations without explicit annotation

Demonstrates effectiveness in pathology cell tracking

Abstract

We propose a weakly-supervised cell tracking method that can train a convolutional neural network (CNN) by using only the annotation of "cell detection" (i.e., the coordinates of cell positions) without association information, in which cell positions can be easily obtained by nuclear staining. First, we train a co-detection CNN that detects cells in successive frames by using weak-labels. Our key assumption is that the co-detection CNN implicitly learns association in addition to detection. To obtain the association information, we propose a backward-and-forward propagation method that analyzes the correspondence of cell positions in the detection maps output of the co-detection CNN. Experiments demonstrated that the proposed method can match positions by analyzing the co-detection CNN. Even though the method uses only weak supervision, the performance of our method was almost the same as the state-of-the-art supervised method.