Search

Automatically segmenting specific tissues or structures from medical images is a straightforward task for deep learning models. However, identifying a few specific objects from a group of similar targets can be a challenging task. This study focuses on the segmentation of certain specific intervertebral discs from lateral spine images acquired from an MRI scanner. In this research, an approach is proposed that utilizes MultiResUNet models and employs saliency maps for target intervertebral disc segmentation. First, a sub-image cropping method is used to separate the target discs. This method uses MultiResUNet to predict the saliency maps of target discs and crop sub-images for easier segmentation. Then, MultiResUNet is used to segment the target discs in these sub-images. The distance maps of the segmented discs are then calculated and combined with their original image for data augmentation to predict the remaining target discs. The training set and test set use 2674 and 308 MRI images, respectively. Experimental results demonstrate that the proposed method significantly enhances segmentation accuracy to about 98%. The performance of this approach highlights its effectiveness in segmenting specific intervertebral discs from closely similar discs. [ABSTRACT FROM AUTHOR]

Image dehazing is a crucial image processing step for outdoor vision systems. However, images recovered through conventional image dehazing methods that use either haze-relevant priors or heuristic cues to estimate transmission maps may not lead to sufficiently accurate haze removal from single images. The most commonly observed effects are darkened and brightened artifacts on some areas of the recovered images, which cause considerable loss of fidelity, brightness, and sharpness. This paper develops a variational image dehazing method on the basis of a color-transfer image dehazing model that is superior to conventional image dehazing methods. By creating a color-transfer image dehazing model to remove haze obscuration and acquire information regarding the coefficients of the model by using the devised convolutional neural network-based deep framework as a supervised learning strategy, an image fidelity, brightness, and sharpness can be effectively restored. The experimental results verify through quantitative and qualitative evaluations of either synthesized or real haze images, and the proposed method outperforms existing single image dehazing methods. [ABSTRACT FROM AUTHOR]

Solution-processed vanadium-doped molybdenum oxide films (V)MoO x films with mole ratios of Mo:V = 1:0, 1:0.05, 1:0.2, 1:0.5, 0:1, are fabricated as hole transport layer (HTL) in organic photovoltaics with active layer blend comprising poly(3-hexylthiophene) (P3HT) and [6,6]-phenyl-C 61 butyric acid methyl ester (PCBM). The device structure is ITO/(V)MoO x /P3HT:PCBM/ZnO NP/Al, and the working area is 0.16 cm 2 . The result shows that the device using V 0.05 MoO x HTL has the best performance, including power conversion efficiency of 2.16%, V oc of 0.6 V, J sc of 6.93 mA/cm 2 , and FF of 51.9%. Using ultraviolet photoelectron spectroscopy (UPS), we can define the energy levels of valence band edge and Fermi level of (V)MoO x films. UPS analysis indicates that V 0.05 MoO x has the smallest energy band offset between its valence band edge to the HOMO of P3HT, which is advantageous for hole transporting from P3HT to ITO anode via the V 0.05 MoO x HTL. In addition, V 0.05 MoO x film shows the lowest electrical resistivity among all (V)MoO x films, which is further beneficial for hole transportation. [ABSTRACT FROM AUTHOR]

Single crystal Sb 2 Te 3 topological insulator nanoflakes with the thickness of 25 nm and 456 nm were synthesized via vapor phase deposition method. The Hall resistance and magnetoresistance of the nanoflakes have been measured at temperatures 2 K and 300 K in the fields up to 9 T. The magnetoresistance and Hall resistance of the nanoflakes demonstrate significant differences, so that despite ordinary magnetoresistance and Hall effect obtained for 25 nm sample 450 nm nanoflake demonstrates unusual magnetoresistance and nonlinear Hall resistance. The sense of curvature of both R xx ( B ) and R xy ( B ) dependences is inversed at high temperature. The experimental data have been analyzed in the frame of a multichannel transport model. The difference in the behavior is attributed to the existence of the charge carriers with high and low mobility, as well as to their relative contribution (which varies depending on the temperature) to the magneto-transport of the nanoflakes. [ABSTRACT FROM AUTHOR]

Magnetotransport properties of a single crystal Sb 2 Te 3 topological insulator nanoflake with the thickness of 25 nm, synthesized via a vapor phase deposition method, are studied. The mobility and number of carriers are obtained through the Hall resistance measurement data. The magnetoresistance shows pronounced weak antilocalization effect. Temperature dependence of the phase coherence length confirms the 2D nature of the observed weak antilocalization effect. The fluctuations in the magnetoresistance are attributed to the combined contribution of the universal conductance fluctuation and the Shubnikov de Haas oscillations of the 2D electron gas. The Shubnikov de Haas oscillations reveal two well defined frequencies ascribed to states with spin-up and spin-down polarizations. The Rashba parameter and the energy gap between the two spin polarized subbands have been obtained and were equal to δ= 0.75×10 −11 eV m and Δ E= 2.4 meV correspondingly. [ABSTRACT FROM AUTHOR]

Freestanding micrometer thick graphene oxide (GO) membranes combine high stiffness, low mass density, and high loss coefficient. This unique combination of properties is ideal for efficient and broadband electro‐acoustic transduction, relying on membrane lightness, stiffness, and internal damping. Here, the viscoelastic response of GO membranes is measured, and the application of ≈100 µm thick GO membranes is demonstrated in dynamic loudspeakers. Using dynamic mechanical analysis and the time‐temperature superposition principle of polymer rheology, it is found that the stiffness of GO membrane increases by more than 50% over 1–20 kHz while damping decreases by less than 20%. GO membranes exhibit 45% higher damping than aluminum membranes in loudspeakers assemblies. Consequently, GO membranes enable the upshift of loudspeaker breakup frequency by 16 to 12 octave above speakers assembled with aluminum, polyethylene terephthalate, titanium, and oak wood membranes. GO is thus found to be an exceptional material for electro‐acoustic transduction. [ABSTRACT FROM AUTHOR]

Freestanding micrometer thick graphene oxide (GO) membranes combine high stiffness, low mass density, and high loss coefficient. This unique combination of properties is ideal for efficient and broadband electro‐acoustic transduction, relying on membrane lightness, stiffness, and internal damping. Here, the viscoelastic response of GO membranes is measured, and the application of ≈100 µm thick GO membranes is demonstrated in dynamic loudspeakers. Using dynamic mechanical analysis and the time‐temperature superposition principle of polymer rheology, it is found that the stiffness of GO membrane increases by more than 50% over 1–20 kHz while damping decreases by less than 20%. GO membranes exhibit 45% higher damping than aluminum membranes in loudspeakers assemblies. Consequently, GO membranes enable the upshift of loudspeaker breakup frequency by 16 to 12 octave above speakers assembled with aluminum, polyethylene terephthalate, titanium, and oak wood membranes. GO is thus found to be an exceptional material for electro‐acoustic transduction. [ABSTRACT FROM AUTHOR]

The automatic segmentation of intervertebral discs from medical images is an important task for an intelligent clinical system. In this study, a deep learning model based on the MultiResUNet model for the automatic segmentation of specific intervertebral discs is presented. MultiResUNet can easily segment all intervertebral discs in MRI images; however, when only certain specific intervertebral discs need to be segmented, problems with segmentation errors, misalignment, and noise occur. In order to solve these problems, a two-stage MultiResUNet model is proposed. Connected-component labeling, automatic cropping, and distance transform are used in the proposed method. The experimental results show that the segmentation errors and misalignments of specific intervertebral discs are greatly reduced, and the segmentation accuracy is increased to about 94%. The performance of the proposed method proves its usefulness for the automatic segmentation of specific intervertebral discs over other deep learning models, such as the U-Net, CNN-based, Attention U-Net, and MultiResUNet models. [ABSTRACT FROM AUTHOR]