A convolutional neural network with pixel-wise sparse graph reasoning for COVID-19 lesion segmentation in CT images.

The COVID-19 pandemic has extremely threatened human health, and automated algorithms are needed to segment infected regions in the lung using computed tomography (CT). Although several deep convolutional neural networks (DCNNs) have proposed for this purpose, their performance on this task is suppressed due to the limited local receptive field and deficient global reasoning ability. To address these issues, we propose a segmentation network with a novel pixel-wise sparse graph reasoning (PSGR) module for the segmentation of COVID-19 infected regions in CT images. The PSGR module, which is inserted between the encoder and decoder of the network, can improve the modeling of global contextual information. In the PSGR module, a graph is first constructed by projecting each pixel on a node based on the features produced by the encoder. Then, we convert the graph into a sparsely-connected one by keeping K strongest connections to each uncertainly segmented pixel. Finally, the global reasoning is performed on the sparsely-connected graph. Our segmentation network was evaluated on three publicly available datasets and compared with a variety of widely-used segmentation models. Our results demonstrate that (1) the proposed PSGR module can capture the long-range dependencies effectively and (2) the segmentation model equipped with this PSGR module can accurately segment COVID-19 infected regions in CT images and outperform all other competing models.

3D bi-directional transformer U-Net for medical image segmentation.

As one of the popular deep learning methods, deep convolutional neural networks (DCNNs) have been widely adopted in segmentation tasks and have received positive feedback. However, in segmentation tasks, DCNN-based frameworks are known for their incompetence in dealing with global relations within imaging features. Although several techniques have been proposed to enhance the global reasoning of DCNN, these models are either not able to gain satisfying performances compared with traditional fully-convolutional structures or not capable of utilizing the basic advantages of CNN-based networks (namely the ability of local reasoning). In this study, compared with current attempts to combine FCNs and global reasoning methods, we fully extracted the ability of self-attention by designing a novel attention mechanism for 3D computation and proposed a new segmentation framework (named 3DTU) for three-dimensional medical image segmentation tasks. This new framework processes images in an end-to-end manner and executes 3D computation on both the encoder side (which contains a 3D transformer) and the decoder side (which is based on a 3D DCNN). We tested our framework on two independent datasets that consist of 3D MRI and CT images. Experimental results clearly demonstrate that our method outperforms several state-of-the-art segmentation methods in various metrics.

Evaluation Method of Equalization of Basic Medical Services from the Spatial Perspective: The Case of Xinjiang, China

Protecting residents' health and improving equality are important goals of the United Nations Sustainable Development Goals. The recent outbreak of COVID-19 has placed a heavy burden on the medical systems of many countries and been disastrous for the low-income population of the world, which has further increased economic, health, and lifelong inequality in society. One way to improve the population's health is to equalize basic medical services. A scientific evaluation of the status quo or the equalization of basic medical services (EBMS) is the basic prerequisite and an important basis for realizing the equitable allocation of medical resources. Traditional evaluation methods ignore the spatial characteristics of medical services, mostly using the indicator of equal weight evaluation, which restricts the objectivity of the evaluation results. Given this, this research proposes a set of EBMS evaluation methods from a spatial perspective and takes the Xinjiang Uygur Autonomous Region of China (Xinjiang) as an example for studying the status quo of EBMS. This study puts forward a set of EBMS evaluation methods from a geospatial perspective and makes full use of spatial analysis and information theory techniques to construct a two-level evaluation indicator that takes into account the spatial characteristics of EBMS. The entropy weight method and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) method have been used to reveal the current status quo of EBMS in Xinjiang to objectively reflect the differences in EBMS. When using the entropy and TOPSIS methods, the evaluation is always based on the data so that the results can more objectively reveal the medical resources available to the residents. Therefore, the government can realize a reasonable allocation of medical resources.