Mixed-Supervised Dual-Network for Medical Image Segmentation

Abstract

Deep learning based medical image segmentation models usually require large datasets with high-quality dense segmentations to train, which are very time-consuming and expensive to prepare. One way to tackle this challenge is using the mixed-supervised learning framework, in which only a part of data is densely annotated with segmentation label and the rest is weakly labeled with bounding boxes. The model is trained jointly in a multi-task learning setting. In this paper, we propose Mixed-Supervised Dual-Network (MSDN), a novel architecture which consists of two separate networks for the detection and segmentation tasks respectively, and a series of connection modules between the layers of the two networks. These connection modules are used to transfer useful information from the auxiliary detection task to help the segmentation task. We propose to use a recent technique called Squeeze and Excitation in the connection module to boost the transfer. We conduct experiments on two medical image segmentation datasets. The proposed MSDN model outperforms multiple baselines.

Deep learning applications in pulmonary medical imaging: recent updates and insights on COVID-19

“…Wang et al [117] proposed a mixed supervised dualnetwork (MSDN) that consists of a network for detection and another one for segmentation, and used "Squeeze and Excitation" for transferring information from auxiliary detection to help segmentation.…”

Coarse-to-Fine Lung Nodule Segmentation in CT Images With Image Enhancement and Dual-Branch Network

“…Wang et al propose a mixed-supervised dual network for medical image segmentation. The model uses two independent networks to perform detection and segmentation tasks, which works well for the lung nodule segmentation task [17]. Junhua Gu proposes an improved detection model of lung nodules based on deformation convolution, which is combined with a simple and effective lung nodule size variation strategy so that the model can effectively fuse the feature maps of different sizes [18].…”

Hybrid Supervision Learning for Pathology Whole Slide Image Classification

“…Mixed supervision learning on various levels of annotations has shown its effectiveness in various machine learning applications [1]- [3] . However, in the context of computational pathology, this is still a challenging problem, as the highresolution of whole slide images makes it unattainable to conduct end-to-end training of deep learning models using existing weak or mixed supervision learning methods [1], [2], [4]- [12]. As the size of pathological whole slide images is around 100,000 × 100,000 pixels, the discriminative regions usually only occupy a small portion of the image.…”