UniHead: Unifying Multi-Perception for Detection Heads.

The detection head constitutes a pivotal component within object detectors, tasked with executing both classification and localization functions. Regrettably, the commonly used parallel head often lacks omni perceptual capabilities, such as deformation perception (DP), global perception (GP), and cross-task perception (CTP). Despite numerous methods attempting to enhance these abilities from a single aspect, achieving a comprehensive and unified solution remains a significant challenge. In response to this challenge, we develop an innovative detection head, termed UniHead, to unify three perceptual abilities simultaneously. More precisely, our approach: 1) introduces DP, enabling the model to adaptively sample object features; 2) proposes a dual-axial aggregation transformer (DAT) to adeptly model long-range dependencies, thereby achieving GP; and 3) devises a cross-task interaction transformer (CIT) that facilitates interaction between the classification and localization branches, thus aligning the two tasks. As a plug-and-play method, the proposed UniHead can be conveniently integrated with existing detectors. Extensive experiments on the COCO dataset demonstrate that our UniHead can bring significant improvements to many detectors. For instance, the UniHead can obtain + 2.7 AP gains in RetinaNet, + 2.9 AP gains in FreeAnchor, and + 2.1 AP gains in GFL. The code is available at https://github.com/zht8506/UniHead.

Comparing Sonazoid contrast-enhanced ultrasound to contrast-enhanced CT and MRI for differentially diagnosing renal lesions: a prospective multicenter study.

PURPOSE: To evaluate the diagnostic performance of contrast-enhanced (CE) ultrasound using Sonazoid (SNZ-CEUS) by comparing with contrast-enhanced computed tomography (CE-CT) and contrast-enhanced magnetic resonance imaging (CE-MRI) for differentiating benign and malignant renal masses. MATERIALS AND METHODS: 306 consecutive patients (from 7 centers) with renal masses (40 benign tumors, 266 malignant tumors) diagnosed by both SNZ-CEUS, CE-CT or CE-MRI were enrolled between September 2020 and February 2021. The examinations were performed within 7 days, but the sequence was not fixed. Histologic results were available for 301 of 306 (98.37%) lesions and 5 lesions were considered benign after at least 2 year follow-up without change in size and image characteristics. The diagnostic performances were evaluated by sensitivity, specificity, positive predictive value, negative predictive value, and compared by McNemar's test. RESULTS: In the head-to-head comparison, SNZ-CEUS and CE-MRI had comparable sensitivity (95.60 vs. 94.51%, P = 0.997), specificity (65.22 vs. 73.91%, P = 0.752), positive predictive value (91.58 vs. 93.48%) and negative predictive value (78.95 vs. 77.27%); SNZ-CEUS and CE-CT showed similar sensitivity (97.31 vs. 96.24%, P = 0.724); however, SNZ-CEUS had relatively lower than specificity than CE-CT (59.09 vs. 68.18%, P = 0.683). For nodules > 4 cm, CE-MRI demonstrated higher specificity than SNZ-CEUS (90.91 vs. 72.73%, P = 0.617) without compromise the sensitivity. CONCLUSIONS: SNZ-CEUS, CE-CT, and CE-MRI demonstrate desirable and comparable sensitivity for the differentiation of renal mass. However, the specificity of all three imaging modalities is not satisfactory. SNZ-CEUS may be a suitable alternative modality for patients with renal dysfunction and those allergic to gadolinium or iodine-based agents.

Sodium lignosulfonate/graphene composites for efficient desalination by incorporating CoS to control pore size.

The phenomenon of overlapping double layers due to micropores inhibits capacitive deionization performance, which is improved by increasing the pore size. In this study, a novel ternary composite electrode (sodium lignosulfonate/reduced graphene oxide/cobalt sulfide, LGC) was designed using a two-step hydrothermal method. CoS with high pseudocapacitance modifies sodium lignosulfonate and graphene connected by hydrogen bonding, benefiting from the constitutive steric structure. The electrochemical performance was significantly enhanced, and the desalination capacity substantially improved. The LGC electrode specific capacitance was as high as 354.47 F g-1 at a 1 A g-1 current density. The desalination capacity of the capacitive deionization device comprising LGC and activated carbon in 1 M NaCl electrolyte reached 28.04 mg g-1 at an operating condition of 1.2 V, 7 mL min-1. Additionally, the LGC electrodes degraded naturally post the experiment by simply removing the CoS, suggesting that the LGC composites are promising material for capacitive deionization electrodes.

Research on a three-dimensional SERS substrate based on a CNTs/Ag@Au/SiO2 composite for detection of fipronil and imidacloprid pesticides.

Surface-enhanced Raman scattering (SERS) has a unique fingerprint spectrum, which allows for rapid, highly sensitive, and non-destructive detection without the need for sample pretreatment. However, SERS substrates have disadvantages such as short storage time and poor reproducibility. In this study, carbon nanotubes, gold, and silver were combined to take advantage of their inherent structural and characteristic properties that enhance the Raman effect. A new type of SERS composite substrate, CNTs/Ag@Au/SiO2, was prepared using a hydrothermal method and seed growth method. The substrate was characterized using transmission electron microscopy (TEM), and the average distance between the core-shell nanoparticles was found to be 3.1 nm, which is more suitable than other gold-silver combined core-shell structures and significantly improves the SERS enhancement factor. The substrate demonstrated high sensitivity even at low concentrations of probe molecules and good uniformity at five randomly selected locations. After storage for 45 days, the substrate still exhibited good stability. In most gold-silver combined core-shell structures, the detection limit for Rhodamine 6G (R6G) is 10-9 mol L-1, while in this substrate, the detection limit for R6G is 10-11 mol L-1. Furthermore, the contribution of the substrate's enhancement was deeply investigated using finite-difference time-domain (FDTD), which revealed that the substrate's hotspots were present in two forms: the "hotspots generated between Ag@Au nanoparticles" and the "hotspots generated between Ag@Au nanoparticles and carbon nanotubes". These two forms of hotspots also demonstrated that the performance brought about by the preparation of the substrate structure was reliable. The simulation results were compared with the experimental results, and the analysis showed that the real environment would have an impact on the substrate's structure during the actual substrate preparation process. Finally, the substrate was used for detecting the pesticide fipronil, and the results showed clear peaks even at a concentration of 0.1 mg L-1. The results indicated that the Raman intensity was linearly exponential with the fipronil solution concentration, with a determination coefficient of R2 = 0.991. This study provides a new SERS substrate for pesticide residue detection and further explores the improvement of pesticide detection limits.

Toward a No-Reference Quality Metric for Camera-Captured Images.

Existing no-reference (NR) image quality assessment (IQA) metrics are still not convincing for evaluating the quality of the camera-captured images. Toward tackling this issue, we, in this article, establish a novel NR quality metric for quantifying the quality of the camera-captured images reliably. Since the image quality is hierarchically perceived from the low-level preliminary visual perception to the high-level semantic comprehension in the human brain, in our proposed metric, we characterize the image quality by exploiting both the low-level image properties and the high-level semantics of the image. Specifically, we extract a series of low-level features to characterize the fundamental image properties, including the brightness, saturation, contrast, noiseness, sharpness, and naturalness, which are highly indicative of the camera-captured image quality. Correspondingly, the high-level features are designed to characterize the semantics of the image. The low-level and high-level perceptual features play complementary roles in measuring the image quality. To infer the image quality, we employ the support vector regression (SVR) to map all the informative features to a single quality score. Thorough tests conducted on two standard camera-captured image databases demonstrate the effectiveness of the proposed quality metric in assessing the image quality and its superiority over the state-of-the-art NR quality metrics. The source code of the proposed metric for camera-captured images is released at https://github.com/YT2015?tab=repositories.

An all-cellulose sponge with a nanofiller-assisted hierarchical cellular structure for fruit maintaining freshness.

Cellulose sponges with compressibility and resilience are an ideal packaging material for fruits with fragile skin. Here, a soft and elastic all-cellulose sponge (CS) with a hierarchical cellular structure was fabricated, where the long molecular chain cellulose constructed major pores, the cellulose at nanoscale acted as an elastic nanofiller to fill the gaps of long molecular chain cellulose fibers and constructed minor pores. With these two kinds of pores, this structure can absorb strain hierarchically. The sponge can protect fruits from mechanical damage when dropped or repeated vibration. Furthermore, the CS modified with chlorogenic acid (C-CGAS) had excellent antibacterial and antifungal abilities. Therefore, C-CGAS could extend the storage time of strawberries to 18 days without any microbial invasion, which is the longest storage time reported thus far. This study provides a new idea for the preparation of polymer sponges and a new design for the development of antimicrobial packaging materials.

Natural phenolics and flavonoids modified the hierarchical cellular cellulose sponges for efficient water disinfection.

Disinfecting microbially contaminated water in a safe and sustainable way is a great challenge. The phenolics and flavonoids in plants are ideal antibacterial agents for biosafety. Herein, an all-plant-derived antibacterial sponge with a hierarchical cellular structure was prepared from natural phenolic and flavonoid crude extracts and cellulose for water purification. After being modified by quercetin, the quercetin-cellulose sponge (Q-CS) exhibits great toughness, elasticity and a large specific surface area, which benefited from a unique hierarchical cellular structure. Q-CS shows ultrahigh water flux (4.5 × 105 L·m-2·h-1·bar-1) and excellent antibacterial abilities against Escherichia coli and Staphylococcus aureus (<1.87 % viability). Sponges modified with crude pyrola (P-CS) and mulberry leaf extracts (M-CS) have similar properties compared with Q-CS. The good performances of P-CS and M-CS show the strong antibacterial applications of natural crude extract. This work provides a strategy for fabricating sustainable and safe antibacterial sponges for water disinfection.

Soft and elastic silver nanoparticle-cellulose sponge as fresh-keeping packaging to protect strawberries from physical damage and microbial invasion.

Strawberry is a nutritious food that is susceptible to mechanical injury and microbiological infection. Traditional coatings for strawberry packaging provide resistance against microbial infection but not against mechanical damage. In this study, a soft and elastic cellulose sponge modified with silver nanoparticles (AgNPs@CS-1:1) was prepared as strawberry packaging material, and it provided effective protection against mechanical damage. In addition, after 1000 cyclic compression, AgNPs@CS-1:1 presented only 16.80% unrecoverable deformation and still had elasticity, suggesting its fatigue resistance and durable protection for strawberry against damage caused by repeated vibrations during transportation. In addition, AgNPs@CS-1:1 had good antibacterial (E. coli and S. aureus) and antifungal (Rhizopus stolonifer) abilities. The storage time of strawberries packaged by AgNPs@CS-1:1 was extended to 12 days without microbial invasion. Thus, AgNPs@CS-1:1 provided dual protection at the physical and microbial levels. This study proposes a new method for the preservation of strawberries based on the utilization of cellulose.

Simulation and Mitigation of the Wrap-Around Artifact in the MRI Image.

Magnetic resonance imaging (MRI) is an essential clinical imaging modality for diagnosis and medical research, while various artifacts occur during the acquisition of MRI image, resulting in severe degradation of the perceptual quality and diagnostic efficacy. To tackle such challenges, this study deals with one of the most frequent artifact sources, namely the wrap-around artifact. In particular, given that the MRI data are limited and difficult to access, we first propose a method to simulate the wrap-around artifact on the artifact-free MRI image to increase the quantity of MRI data. Then, an image restoration technique, based on the deep neural networks, is proposed for wrap-around artifact reduction and overall perceptual quality improvement. This study presents a comprehensive analysis regarding both the occurrence of and reduction in the wrap-around artifact, with the aim of facilitating the detection and mitigation of MRI artifacts in clinical situations.

Effect of different planting areas on the chemical compositions and hypoglycemic and antioxidant activities of mulberry leaf extracts in Southern China.

Guangdong, Guangxi and Chongqing are emerging sericulture areas in China where the production of mulberry leaves is huge. In order to identity high quality mulberry leaves that are suitable for healthy products to expand planting, 24 samples from three regions (Guangdong, Guangxi, Chongqing) in the south of China were quantified for two alkaloids (1-deoxynojirimycin and fagomine) and five phenols (chlorogenic acid, rutin, isoquercitrin, etc.) using high-performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS). Additionally, the total phenolic and total flavonoid contents, antioxidant and glycosidase inhibitory activities (hypoglycemic activity) were tested using different assays (DPPH, ABTS, FRAP) to comprehensively evaluate the quality of the mulberry leaves. The contents of DNJ and fagomine ranged from 0.401±0.003 to 5.309±0.036 mg/g and from 0.279±0.031 to 2.300±0.060 mg/g, respectively. The main phenolic constituents were chlorogenic acid, rutin and isoquercitrin, with chlorogenic acid present in the highest concentrations, ranging from 3.104±0.191 to 10.050±0.143 mg/g. The antioxidant activity exhibited a tendency as follows: Guangxi > Guangdong > Chongqing, except for two samples from Chongqing, which showed the highest antioxidant activity. Based on our study, mulberry leaves from Guangdong and Guangxi could be future sources of natural hypoglycemic and antioxidant products.