Metadata Improves Segmentation Through Multitasking Elicitation

TL;DR

This paper demonstrates that incorporating metadata into convolutional networks via channel modulation enhances biomedical image segmentation performance and offers a simple, cost-effective way to improve deep learning models by enabling multitask switching.

Contribution

The study introduces a channel modulation mechanism to incorporate metadata into convolutional networks, improving segmentation results and facilitating multitask switching in biomedical imaging.

Findings

Metadata improves segmentation accuracy.

Channel modulation is an inexpensive addition.

Metadata facilitates multitask switching.

Abstract

Metainformation is a common companion to biomedical images. However, this potentially powerful additional source of signal from image acquisition has had limited use in deep learning methods, for semantic segmentation in particular. Here, we incorporate metadata by employing a channel modulation mechanism in convolutional networks and study its effect on semantic segmentation tasks. We demonstrate that metadata as additional input to a convolutional network can improve segmentation results while being inexpensive in implementation as a nimble add-on to popular models. We hypothesize that this benefit of metadata can be attributed to facilitating multitask switching. This aspect of metadata-driven systems is explored and discussed in detail.