Search

There are few studies utilizing only IR cameras for long-distance gender recognition, and they have shown low recognition performance due to their lack of color and texture information in IR images with a complex background. Therefore, a rough body segmentation-based gender recognition network (RBSG-Net) is proposed, with enhanced gender recognition performance achieved by emphasizing the silhouette of a person through a body segmentation network. Anthropometric loss for the segmentation network and an adaptive body attention module are also proposed, which effectively integrate the segmentation and classification networks. To enhance the analytic capabilities of the proposed framework, fractal dimension estimation was introduced into the system to gain insights into the complexity and irregularity of the body region, thereby predicting the accuracy of body segmentation. For experiments, near-infrared images from the Sun Yat-sen University multiple modality re-identification version 1 (SYSU-MM01) dataset and thermal images from the Dongguk body-based gender version 2 (DBGender-DB2) database were used. The equal error rates of gender recognition by the proposed model were 4.320% and 8.303% for these two databases, respectively, surpassing state-of-the-art methods.

The accurate recognition of a brain tumor (BT) is crucial for accurate diagnosis, intervention planning, and the evaluation of post-intervention outcomes. Conventional methods of manually identifying and delineating BTs are inefficient, prone to error, and time-consuming. Subjective methods for BT recognition are biased because of the diffuse and irregular nature of BTs, along with varying enhancement patterns and the coexistence of different tumor components. Hence, the development of an automated diagnostic system for BTs is vital for mitigating subjective bias and achieving speedy and effective BT segmentation. Recently developed deep learning (DL)-based methods have replaced subjective methods; however, these DL-based methods still have a low performance, showing room for improvement, and are limited to heterogeneous dataset analysis. Herein, we propose a DL-based parallel features aggregation network (PFA-Net) for the robust segmentation of three different regions in a BT scan, and we perform a heterogeneous dataset analysis to validate its generality. The parallel features aggregation (PFA) module exploits the local radiomic contextual spatial features of BTs at low, intermediate, and high levels for different types of tumors and aggregates them in a parallel fashion. To enhance the diagnostic capabilities of the proposed segmentation framework, we introduced the fractal dimension estimation into our system, seamlessly combined as an end-to-end task to gain insights into the complexity and irregularity of structures, thereby characterizing the intricate morphology of BTs. The proposed PFA-Net achieves the Dice scores (DSs) of 87.54%, 93.42%, and 91.02%, for the enhancing tumor region, whole tumor region, and tumor core region, respectively, with the multimodal brain tumor segmentation (BraTS)-2020 open database, surpassing the performance of existing state-of-the-art methods. Additionally, PFA-Net is validated with another open database of brain tumor progression and achieves a DS of 64.58% for heterogeneous dataset analysis, surpassing the performance of existing state-of-the-art methods.

Solid supported lipid bilayers possess a unique combination of physical properties, which make them well suited to serve as cell membrane mimics, and among these the chief is the two dimensional fluidity of the individual lipid molecules. The method of bilayer preservation requires no substrate modification can be performed using commercially available reagents.

This paper describes stepwise on-demand generation and fusion of femtolitre aqueous droplets based on interfacial tension. Sub-millisecond reaction times from droplet fusion were demonstrated, as well as a reversible chemical toggle switch based on alternating fusion of droplets containing acidic or basic solution, monitored with the pH-dependent emission of fluorescein. [ABSTRACT FROM AUTHOR]

We developed a microfluidic platform for splitting well-mixed, femtoliter-volume droplets from larger water-in-oil plugs, where the sizes of the daughter droplets were not limited by channel width. These droplets were separated from mother plugs at a microfabricated T-junction, which enabled the study of how increased confinement affected enzyme kinetics in droplets 4-10 μm in diameter. Initial rates for enzyme catalysis in the mother plugs and the largest daughter drops were close to the average bulk rate, while the rates in smaller droplets decreased linearly with increasing surface to volume ratio. Rates in the smallest droplets decreased by a factor of 4 compared to the bulk rate. Traditional methods for detecting non-specific adsorption at the water-oil interface were unable to detect evidence of enzyme adsorption, including pendant drop tensiometry, laser scanning confocal microscopy of drops containing labeled proteins in microemulsions, and epifluorescence microscopy of plugs and drops generated on-chip. We propose the slowing of enzyme reaction kinetics in the smaller droplets was the result of increased adsorption and inactivation of enzymes at the water-oil interface arising from transient interfacial shear stresses imparted on the daughter droplets as they split from the mother plugs and passed through the constricted opening of the T-junction. Such stresses are known to modulate the interfacial area and density of surfactant molecules that can passivate the interface. Bright field images of the splitting processes at the junction indicate that these stresses scaled with increasing surface to volume ratios of the droplets but were relatively insensitive to the average flow rate of plugs upstream of the junction. [ABSTRACT FROM AUTHOR]

A device with femtoliter-scale chambers and controlled reaction initiation was developed for single-molecule enzymology. Initially separated substrate and enzyme streams were rapidly mixed in a microfluidic device and encapsulated in an array of individual microreactors, allowing for enzyme kinetics to be monitored with millisecond dead times and single-molecule sensitivity. Because the arrays of chambers were fabricated by micromolding in PDMS, the chambers were monodisperse in size, and the chamber volume could be systematically controlled. Microreactors could be purged and replenished with fresh reactants for consecutive rounds of observation. Repeated experiments with statistically identical initial conditions could be performed rapidly, with zero cross-talk between chambers in the array. [ABSTRACT FROM AUTHOR]

We have developed a general method for photopatterning well-defined patches of enzymes inside a microfluidic device at any location. First, a passivating protein layer was adsorbed to the walls and floor of a poly(dimethylsiloxane)/glass microchannel. The channel was then filled with an aqueous biotin-linked dye solution. Using an Ar+ /Kr+ laser, the fluorophore moieties were bleached to create highly reactive species. These activated molecules subsequently attached themselves to the adsorbed proteins on the microchannel walls and floor via a singlet oxygen-dependent mechanism. Enzymes linked to streptavidin or avidin could then be immobilized via (strept)- avidin/biotin binding. Using this process, we were able to pattern multiple patches of sireptavidin-linked alkaline phosphatase inside a straight microfluidic channel without the use of valves under exclusively aqueous conditions. The density of alkaline phosphatase in the patches was calculated to be ∼5% of the maximum possible density by comparison with known standards. Turnover was observed via fluorogenic substrate conversion and fluorescence microscopy. A more complex two-step enzyme reaction was also designed. In this case, avidin-linked glucose oxidase and streptavidin-linked horseradish per-oxidase were sequentially patterned in separate patches inside straight microfluidic channels. Product formed at the glucose oxidase patch became the substrate for horseradish peroxidase, patterned downstream, where fluorogenic substrate turnover was recorded. [ABSTRACT FROM AUTHOR]

The molecular level details of the displacement of surface adsorbed fibrinogen from silica substrates were studied by atomic force microscopy, immunochemical assays, fluorescence microscopy, and vibrational sum frequency spectroscopy. The results showed that human plasma fibrinogen (HPF) can be readily displaced from the interface by other plasma proteins near neutral pH because the positively charged αC domains on HPF sit between the rest of the macromolecule and the underlying surface. The (αC domains make weak electrostatic contact with the substrate, which is manifest by a high degree of alignment of Lys and Arg residues. Upon cycling through acidic pH, however, the (αC domains are irreversibly removed from this position and the rest of the macromolecule is free to engage in stronger hydrogen bonding, van der Waals, and hydrophobic interactions with the surface. This results in a 170-fold decrease in the rate at which HPF can be displaced from the interface by other proteins in human plasma. [ABSTRACT FROM AUTHOR]

Describes a class of microfluidic immunoassays based upon solid supported lipid bilayers. Characteristics of the fluid bilayers used in the experiment; Method of deriving an entire binding curve in a single experiment; Advantages of the methodology for performing heterogeneous assays developed in the experiment over traditional procedures.

The keratinocyte culture model has previously been used as an in vitro method for testing skin irritating potential of common skin irritants. However, solubility limits its use for finished products. Shampoo is very soluble in water which should make it an ideal product category for the cell culture model. To determine the skin irritant potential of several commercial shampoos, we employed cultured human keratinocytes as an in vitro model. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide test (MTT) and lactic dehydrogenase release (LDH) test were used to document cell toxicity. 7 volunteers were patch tested and their reactions evaluated using laser Doppler flowmetry and compared with the in vitro data. MTT and LDH have a good negative correlation with each other. Patch test reaction, especially at high concentrations, correlates relatively well with the in vitro test, especially with shampoos of strong and weak irritancy. However, the rank order of the shampoos of moderate toxicity was not the same as in the in vitro data. This suggests that the cell culture technique cannot directly replace in vivo methods, and that data obtained by the cell culture method should be interpreted carefully. [ABSTRACT FROM AUTHOR]

Human cytomegalovirus (HCMV) infection is a major cause of morbidity and mortality in transplant patients and a leading cause of congenital birth defects (Saint Louis, 2016). Vaccination and therapeutic studies often require scalable cell culture production of wild type virus, represented by clinical isolates. Obtaining sufficient stocks of wild-type clinical HCMV is often labor intensive and inefficient due to low yield and genetic loss, presenting a barrier to studies of clinical isolates. Here we report a bioreactor method based on continuous infection, where retinal pigment epithelial (ARPE-19) cells adhered to microcarrier beads are infected in a bioreactor and used to produce high-titers of clinical isolate HCMV that maintain genetic integrity of key viral tropism factors and the viral genome. In this bioreactor, an end-stage infection can be maintained by regular addition of uninfected ARPE-19 cells, providing convenient preparation of 107-108 pfu/ml of concentrated TB40/E IE2-EYFP stocks without daily cell passaging or trypsinization. Overall, this represents a 100-fold increase in gain of virus production of 100-times compared to conventional static-culture plates, while requiring 90% less handling time. Moreover, this continuous infection environment has the potential to monitor infection dynamics with applications for real-time tracking of viral evolution.