UNet++: A Nested U-Net Architecture for Medical Image Segmentation

TLDR

The authors introduce UNet++, a nested U‑Net architecture designed to improve medical image segmentation by reducing the semantic gap between encoder and decoder feature maps. UNet++ is a deeply supervised encoder‑decoder network that connects encoder and decoder sub‑networks through nested, dense skip pathways. Across nodule, nuclei, liver, and polyp segmentation tasks, UNet++ outperforms U‑Net and wide U‑Net, achieving average IoU gains of 3.9 and 3.4 points, respectively.

Abstract

In this paper, we present UNet++, a new, more powerful architecture for medical image segmentation. Our architecture is essentially a deeply-supervised encoder-decoder network where the encoder and decoder sub-networks are connected through a series of nested, dense skip pathways. The re-designed skip pathways aim at reducing the semantic gap between the feature maps of the encoder and decoder sub-networks. We argue that the optimizer would deal with an easier learning task when the feature maps from the decoder and encoder networks are semantically similar. We have evaluated UNet++ in comparison with U-Net and wide U-Net architectures across multiple medical image segmentation tasks: nodule segmentation in the low-dose CT scans of chest, nuclei segmentation in the microscopy images, liver segmentation in abdominal CT scans, and polyp segmentation in colonoscopy videos. Our experiments demonstrate that UNet++ with deep supervision achieves an average IoU gain of 3.9 and 3.4 points over U-Net and wide U-Net, respectively.