PoolNet+: Exploring the Potential of Pooling for Salient Object Detection.

We explore the potential of pooling techniques on the task of salient object detection by expanding its role in convolutional neural networks. In general, two pooling-based modules are proposed. A global guidance module (GGM) is first built based on the bottom-up pathway of the U-shape architecture, which aims to guide the location information of the potential salient objects into layers at different feature levels. A feature aggregation module (FAM) is further designed to seamlessly fuse the coarse-level semantic information with the fine-level features in the top-down pathway. We can progressively refine the high-level semantic features with these two modules and obtain detail enriched saliency maps. Experimental results show that our proposed approach can locate the salient objects more accurately with sharpened details and substantially improve the performance compared with the existing state-of-the-art methods. Besides, our approach is fast and can run at a speed of 53 FPS when processing a 300 ×400 image. To make our approach better applied to mobile applications, we take MobileNetV2 as our backbone and re-tailor the structure of our pooling-based modules. Our mobile version model achieves a running speed of 66 FPS yet still performs better than most existing state-of-the-art methods. To verify the generalization ability of the proposed method, we apply it to the edge detection, RGB-D salient object detection, and camouflaged object detection tasks, and our method achieves better results than the corresponding state-of-the-art methods of these three tasks. Code can be found at http://mmcheng.net/poolnet/.

Ratiometric fluorescence sensing of formaldehyde in food samples based on bifunctional MOF.

A new fluorescence strategy was described for ratiometric sensing of formaldehyde (FA) with bifunctional MOF, which acted as a fluorescence reporter as well as biomimetic peroxidase. With the assistance of H2O2, NH2-MIL-101 (Fe) catalyzes the oxidation of non-luminescent substrate o-phenylenediamine (OPD) to produce fluorescent product (oxOPD) with the maximum emission at 570 nm. Besides, intrinsic fluorescence of MOF (λem = 445 nm) was quenched by oxOPD through inner filter effect (IFE). However, FA and OPD reacted to generate Schiff bases, which competitively consumed OPD inhibiting the generation of oxOPD. Under the excitation wavelength of 375 nm, a ratiometric strategy was designed to detect FA with the fluorescence intensity ratio at 445 nm and 570 nm (F445/F570) as readout signal. This strategy exhibited a wide linear range (0.1-50 µM) and low detection limit of 0.03 µM. This method was confirmed for FA detection in food samples. In addition to establishing a new method to detect FA, this work will open new applications of MOF in food safety.

Enhanced detection of ascorbic acid with cascaded fluorescence recovery of a dual-nanoquencher system.

An innovative strategy with target-triggered cascade fluorescence recovery of a dual-nanoquencher system was developed to detect ascorbic acid (AA). Herein, manganese dioxide (MnO2) nanosheets and gold nanoparticles (AuNPs) were used as nanoquenchers simultaneously. Owing to their synergistic effects, the fluorescence of 2,3-diaminophenazine (DAP) was decreased efficiently, thus minimizing the background fluorescence. The introduction of AA triggered the decomposition of MnO2 into Mn2+, which induced the aggregation of AuNPs. Both the decomposed MnO2 and aggregated AuNPs possess weak quenching abilities towards DAP. Such a cascade amplification strategy enhanced the detection sensitivity for AA with a LOD as low as 6.7 nM, which was two orders of magnitude lower than that of MnO2-based fluorescence assay. Furthermore, this amplification strategy was successfully applied to detect AA in food samples.

Nifuroxazide in combination with CpG ODN exerts greater efficacy against hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is a common malignant tumour in China that remains a major challenge to the medical community, and effective treatment is urgently needed. Due to complex tumorigenesis, monotherapy shows poor therapeutic effects, and combined treatment becomes a necessary option. YW002, a CpG ODN-containing sequence, has been proven to enhance antitumor effects in tumour-bearing mouse models. Moreover, as a broad-spectrum antimicrobial drug, nifuroxazide exhibited an anti-HCC effect through activation of p-Stat3. Here, we tested the effect of nifuroxazide on HCC in vitro and then explored the therapeutic effect of combined nifuroxazide and CpG ODN on HCC in vivo. Nifuroxazide inhibited proliferation, induced apoptosis and suppressed migration and invasion in HepG2 cells in vitro. The combination therapy using nifuroxazide and CpG ODN significantly suppressed the growth of tumours in tumour-bearing mice with few side effects and achieved better therapeutic effects on HCC than monotherapy. Moreover, combined nifuroxazide and CpG ODN therapy significantly induced apoptosis, enhanced the infiltration of CD4+ and CD8+ T lymphocytes and macrophages in tumour tissue, and increased the ratio of CD4+ and CD8+ T lymphocytes in the spleens of tumour-bearing mice. The introduction of this combination therapy combining nifuroxazide and CpG ODN provided a new strategy for HCC treatment.

Rethinking the U-Shape Structure for Salient Object Detection.

The U-shape structure has shown its advantage in salient object detection for efficiently combining multi-scale features. However, most existing U-shape-based methods focused on improving the bottom-up and top-down pathways while ignoring the connections between them. This paper shows that we can achieve the cross-scale information interaction by centralizing these connections, hence obtaining semantically stronger and positionally more precise features. To inspire the newly proposed strategy's potential, we further design a relative global calibration module that can simultaneously process multi-scale inputs without spatial interpolation. Our approach can aggregate features more effectively while introducing only a few additional parameters. Our approach can cooperate with various existing U-shape-based salient object detection methods by substituting the connections between the bottom-up and top-down pathways. Experimental results demonstrate that our proposed approach performs favorably against the previous state-of-the-arts on five widely used benchmarks with less computational complexity. The source code will be publicly available.