The visual cues from multiple support regions of different sizes and resolutions are complementary in classifying a candidate box in object detection. Effective integration of local and contextual visual cues from these regions has become a fundamental problem in object detection. In this paper, we propose a gated bi-directional CNN (GBD-Net) to pass messages among features from different support regions during both feature learning and feature extraction. Such message passing can be implemented through convolution between neighboring support regions in two directions and can be conducted in various layers. Therefore, local and contextual visual patterns can validate the existence of each other by learning their nonlinear relationships and their close interactions are modeled in a more complex way. It is also shown that message passing is not always helpful but dependent on individual samples. Gated functions are therefore needed to control message transmission, whose on-or-offs are controlled by extra visual evidence from the input sample. The effectiveness of GBD-Net is shown through experiments on three object detection datasets, ImageNet, Pascal VOC2007 and Microsoft COCO. Besides the GBD-Net, this paper also shows the details of our approach in winning the ImageNet object detection challenge of 2016, with source code provided on https://github.com/craftGBD/craftGBD. In this winning system, the modified GBD-Net, new pretraining scheme and better region proposal designs are provided. We also show the effectiveness of different network structures and existing techniques for object detection, such as multi-scale testing, left-right flip, bounding box voting, NMS, and context.
With COVID-19 spreading around the world, many countries are exposed to the imported case risk from inbound international flights. Several governments issued restrictions on inbound flights to mitigate such risk. But with the pandemic controlled in many countries, some decide to reopen the economy by relaxing the international air travel bans. As the virus has still been prevailing in many regions, this relaxation raises the alarm to import overseas cases and results in the revival of local pandemic. This study proposes a risk index to measure one country's imported case risk from inbound international flights. The index combines both daily dynamic international air connectivity data and the updated global COVID-19 data. It can measure the risk at the country, province and even specific route level. The proposed index was applied to China, which is the first country to experience and control COVID-19 pandemic while later becoming exposed to high imported case risk after the epidemic centers switched to Europe and the US afterward. The calculated risk indexes for each Chinese province or region show both spatial and temporal patterns from January to April 2020. It is found that China's strict restriction on inbound flights since March 26 was very effective to cut the imported case risk by half than doing nothing. But the overall index level kept rising because of the deteriorating pandemic conditions around the world. Hong Kong and Taiwan are the regions facing the highest imported case risk due to their superior international air connectivity and looser restriction on inbound flights. Shandong Province had the highest risk in February and early March due to its well-developed air connectivity with South Korea and Japan when the pandemic peaked in these two countries. Since mid-March, the imported case risk from Europe and the US dramatically increased. Last, we discuss policy implications for the relevant stakeholders to use our index to dynamically adjust the international air travel restrictions. This risk index can also be applied to other contexts and countries to relax restrictions on particular low-risk routes while still restricting the high-risk ones. This would balance the essential air travels need and the requirement to minimize the imported case risk.
The tumor microenvironment (TME) is a complex mixture of cell types whose interactions affect tumor growth and clinical outcome. To discover such interactions, we developed CODEFACS (COnfident DEconvolution For All Cell Subsets), a tool deconvolving cell type–specific gene expression in each sample from bulk expression, and LIRICS (Ligand–Receptor Interactions between Cell Subsets), a statistical framework prioritizing clinically relevant ligand–receptor interactions between cell types from the deconvolved data. We first demonstrate the superiority of CODEFACS versus the state-of-the-art deconvolution method CIBERSORTx. Second, analyzing The Cancer Genome Atlas, we uncover cell type–specific ligand–receptor interactions uniquely associated with mismatch-repair deficiency across different cancer types, providing additional insights into their enhanced sensitivity to anti–programmed cell death protein 1 (PD-1) therapy compared with other tumors with high neoantigen burden. Finally, we identify a subset of cell type–specific ligand–receptor interactions in the melanoma TME that stratify survival of patients receiving anti–PD-1 therapy better than some recently published bulk transcriptomics-based methods. Significance: This work presents two new computational methods that can deconvolve a large collection of bulk tumor gene expression profiles into their respective cell type–specific gene expression profiles and identify cell type–specific ligand–receptor interactions predictive of response to immune-checkpoint blockade therapy. This article is highlighted in the In This Issue feature, p. 873
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.