Multi-level Activation for Segmentation of Hierarchically-nested Classes

TL;DR

This paper introduces a multi-level activation layer for CNNs that encodes hierarchical class nesting, significantly improving segmentation accuracy in biological images with nested structures.

Contribution

The authors propose a novel multi-level activation layer and compatible loss functions to incorporate hierarchical class information into CNNs for segmentation tasks.

Findings

Speeds up learning process

Outperforms standard soft-max classification

Significantly improves Dice score (p<0.007)

Abstract

For many biological image segmentation tasks, including topological knowledge, such as the nesting of classes, can greatly improve results. However, most `out-of-the-box' CNN models are still blind to such prior information. In this paper, we propose a novel approach to encode this information, through a multi-level activation layer and three compatible losses. We benchmark all of them on nuclei segmentation in bright-field microscopy cell images from the 2018 Data Science Bowl challenge, offering an exemplary segmentation task with cells and nested subcellular structures. Our scheme greatly speeds up learning, and outperforms standard multi-class classification with soft-max activation and a previously proposed method stemming from it, improving the Dice score significantly (p-values<0.007). Our approach is conceptually simple, easy to implement and can be integrated in any CNN architecture. It can be generalized to a higher number of classes, with or without further relations of containment.