A diagonal masking self-attention-based multi-scale network for motor imagery classification.

Electroencephalography (EEG)-based motor imagery (MI) is a promising paradigm for brain-computer interface (BCI), but the non-stationarity and low signal-to-noise ratio of EEG signals make it a challenging task. To achieve high-precision motor imagery classification, we propose a Diagonal Masking Self-Attention-based Multi-Scale Network (DMSA-MSNet) to fully develop, extract, and emphasize features from different scales. First, for local features, a multi-scale temporal-spatial block is proposed to extract features from different receptive fields. Second, an adaptive branch fusion block is specifically designed to bridge the semantic gap between these coded features from different scales. Finally, in order to analyze global information over long ranges, adiagonal masking self-attentionblock is introduced, whichhighlightsthe most valuable features in the data. The proposed DMSA-MSNet outperforms state-of-the-art models on the BCI Competition IV 2a and the BCI Competition IV 2b datasets. Our study achieves rich information extraction from EEG signals and provides an effective solution for motor imagery classification. The code is available at https://github.com/HyperSystemAndImageProc/A-diagonal-masking-self-attention-based-Multi-Scale-Network-for-motor-imagery-classification. .

Irradiance uniformity reshaping method for solar simulators based on computer-generated spatial filter films.

Irradiance uniformity is critical to the accuracy of photovoltaic device test results. Therefore, to post-correct the irradiance uniformity inherent in artificial lighting systems, a spatial irradiance filter scheme for film patterns is proposed based on the physical phenomenon of a positively related relationship between inkjet concentration and the transparency of the flexible film. The scheme first establishes the characteristic equation between the irradiance absorption and pattern grayscale values and then generates the spatial filtering pattern by utilizing the light intensity distribution to be calibrated, matrix operations, and bilinear interpolation. To evaluate its performance, an STM32 microprocessor-based irradiance distribution measurement system was developed and used to test and verify single lamp, planar array, and curved surface array light sources. The results reveal that the corrected irradiance uniformity improves by 15.5%, 24.01 %, and 13.11%, all of which achieve the Class A irradiance uniformity of the IEC 60904-9 standard.

Hyperspectral upgrade solution for biomicroscope combined with Transformer network to classify infectious bacteria.

Infectious diseases caused by bacterial pathogens pose a significant public health threat, emphasizing the need for swift and accurate bacterial species detection methods. Hyperspectral microscopic imaging (HMI) offers nondestructive, rapid, and data-rich advantages, making it a promising tool for microbial detection. In this research, we present a highly compatible and cost-effective approach to extend a standard biomicroscope system into a hyperspectral biomicroscope using a prism-grating-prism configuration. Using this prototype, we generate 600 hyperspectral data cubes for Listeria, Bacillus typhi, Bacillus pestis, and Bacillus anthracis. Additionally, we propose a Transformer-based classification network that achieves a 99.44% accuracy in classifying these infectious pathogens, outperforming traditional methods. Our results suggest that the successful combination of HMI and the optimized Transformer-based classification network highlights the potential for rapid and precise detection of infectious disease pathogens .

Multi-scale feature pyramid fusion network for medical image segmentation.

PURPOSE: Medical image segmentation is the most widely used technique in diagnostic and clinical research. However, accurate segmentation of target organs from blurred border regions and low-contrast adjacent organs in Computed tomography (CT) imaging is crucial for clinical diagnosis and treatment. METHODS: In this article, we propose a Multi-Scale Feature Pyramid Fusion Network (MS-Net) based on the codec structure formed by the combination of Multi-Scale Attention Module (MSAM) and Stacked Feature Pyramid Module (SFPM). Among them, MSAM is used to skip connections, which aims to extract different levels of context details by dynamically adjusting the receptive fields under different network depths; the SFPM including multi-scale strategies and multi-layer Feature Perception Module (FPM) is nested in the network at the deepest point, which aims to better focus the network's attention on the target organ by adaptively increasing the weight of the features of interest. RESULTS: Experiments demonstrate that the proposed MS-Net significantly improved the Dice score from 91.74% to 94.54% on CHAOS, from 97.59% to 98.59% on Lung, and from 82.55% to 86.06% on ISIC 2018, compared with U-Net. Additionally, comparisons with other six state-of-the-art codec structures also show the presented network has great advantages on evaluation indicators such as Miou, Dice, ACC and AUC. CONCLUSION: The experimental results show that both the MSAM and SFPM techniques proposed in this paper can assist the network to improve the segmentation effect, so that the proposed MS-Net method achieves better results in the CHAOS, Lung and ISIC 2018 segmentation tasks.

Pass-Transistor Logic Circuits Based on Wafer-Scale 2D Semiconductors.

2D semiconductors, such as molybdenum disulfide (MoS2 ), have attracted tremendous attention in constructing advanced monolithic integrated circuits (ICs) for future flexible and energy-efficient electronics. However, the development of large-scale ICs based on 2D materials is still in its early stage, mainly due to the non-uniformity of the individual devices and little investigation of device and circuit-level optimization. Herein, a 4-inch high-quality monolayer MoS2 film is successfully synthesized, which is then used to fabricate top-gated (TG) MoS2 field-effect transistors with wafer-scale uniformity. Some basic circuits such as static random access memory and ring oscillators are examined. A pass-transistor logic configuration based on pseudo-NMOS is then employed to design more complex MoS2 logic circuits, which are successfully fabricated with proper logic functions tested. These preliminary integration efforts show the promising potential of wafer-scale 2D semiconductors for application in complex ICs.

Explorations on Growth of Blue-Green-Yellow-Red InGaN Quantum Dots by Plasma-Assisted Molecular Beam Epitaxy.

Self-assembled growth of blue-green-yellow-red InGaN quantum dots (QDs) on GaN templates using plasma-assisted molecular beam epitaxy were investigated. We concluded that growth conditions, including small N2 flow and high growth temperature are beneficial to the formation of InGaN QDs and improve the crystal quality. The lower In/Ga flux ratio and lower growth temperature are favorable for the formation of QDs of long emission wavelength. Moreover, the nitrogen modulation epitaxy method can extend the wavelength of QDs from green to red. As a result, visible light emissions from 460 nm to 622 nm have been achieved. Furthermore, a 505 nm green light-emitting diode (LED) based on InGaN/GaN MQDs was prepared. The LED has a low external quantum efficiency of 0.14% and shows an efficiency droop with increasing injection current. However, electroluminescence spectra exhibited a strong wavelength stability, with a negligible shift of less than 1.0 nm as injection current density increased from 8 A/cm2 to 160 A/cm2, owing to the screening of polarization-related electric field in QDs.

Effects and mechanisms of In surfactant on high Al-content AlGaN grown by plasma-assisted molecular beam epitaxy.

High Al-content AlGaN epilayers were grown on AlN template by using indium (In) surfactant with plasma-assisted molecular beam epitaxy (PA-MBE), and deep ultraviolet emission at 235 nm was obtained at room temperature. The effects and mechanisms of In-surfactant on the crystalline quality and optical properties of AlGaN were investigated. It was found that In-surfactant could facilitate two-dimensional AlGaN growth by reducing activation barrier for Al/Ga atoms to cross steps and effectively increasing the migration rate on the growth surface, and thus improve surface morphology and decrease defect density. The photoluminescence measurements revealed that the optical properties were remarkably improved by adopting In as surfactant, and phase separation was also effectively eliminated. Furthermore, the concentration of impurities including oxygen and silicon was decreased, which is attributed to higher defects formation energy for these impurities with In-surfactant assisted epitaxy growth.

A Self-Powered Transparent Photodetector Based on Detached Vertical (In,Ga)N Nanowires with 360° Omnidirectional Detection for Underwater Wireless Optical Communication.

Underwater wireless optical communication (UWOC) is a wireless communication technology using visible light to transmit data in an underwater environment, which has wide applications. Based on lift-off (In,Ga)N nanowires, this work has proposed and successfully demonstrated a self-powered photoelectrochemical (PEC) photodetector (PD) with excellent transmissivity. The transparent functionality of the PD is critical for 360° omnidirectional underwater detection, which was realized by detaching the (In,Ga)N nanowires from the opaque epitaxial substrates to the indium tin oxide (ITO)/glass. It was also found that the insulating SiO2 layer can enhance the photocurrent by about 12 times. The core-shell structure of the nanowires is beneficial for generating carriers and contributing to the photocurrent. Furthermore, a communication system with ASCII code is set to demonstrate the PD detection in underwater communication. This work paves an effective way to develop 360° omnidirectional PDs for the wide applications in UWOC system and underwater photodetection.

Improved photocatalytic activity and stability of InGaN quantum dots/C3N4heterojunction photoelectrode for CO2reduction and hydrogen production.

Photocatalytic conversion of CO2to produce fuel is considered a promising approach to reduce CO2emissions and tackle energy crisis. GaN-based materials have been studied for CO2reduction because of their excellent optical properties and band structure. However, low photocatalytic activity and severe photocorrosion of GaN-based photoelectrode greatly limit their applications. In this work, photocatalytic activity was improved by adopting InGaN quantum dots (QDs) combined with C3N4nano-sheets as photoanode, and thus the efficiency of CO2reduction and the selectivity of hydrogen production were increased significantly. In addition, the photoelectron-chemical corrosion of photoelectrodes has been apparently controlled. InGaN QDs/C3N4has the highest CO and H2productions rates of 14.69µmol mol-1h-1and 140µmol mol-1h-1which were 2.2 times and 14.5 times than that of InGaN film photoelectrode, respectively. The enhancement of photocatalytic activity is attributed to C3N4modification and a large electric dipole forming on the surface of InGaN QDs, which facilitate the separation and transfer of photo-generated carriers and thus promote CO2reduction reaction. This work provides a promising strategy for the development of GaN-based photoanodes with superior stability and efficiency.

Dual-wavelength visible photodetector based on vertical (In,Ga)N nanowires grown by molecular beam epitaxy.

Due to the wide applications of blue and red photodetectors, dual-wavelength (blue/red) photodetectors are promising for future markets. In this work, a dual-wavelength photodetector based on vertical (In,Ga)N nanowires and graphene has been fabricated successfully. By using the transparent graphene, both blue and red responses can be clearly detected. The rise time of response can reach 3.5 ms. Furthermore, the underlying mechanism of double responses has also been analyzed. The main reason contributing to the dual-wavelength response could be the different diameters of nanowires, leading to different In components within (In,Ga)N sections.

Visible to short wavelength infrared In2Se3-nanoflake photodetector gated by a ferroelectric polymer.

Photodetectors based on two-dimensional (2D) transition-metal dichalcogenides have been studied extensively in recent years. However, the detective spectral ranges, dark current and response time are still unsatisfactory, even under high gate and source-drain bias. In this work, the photodetectors of In2Se3 have been fabricated on a ferroelectric field effect transistor structure. Based on this structure, high performance photodetectors have been achieved with a broad photoresponse spectrum (visible to 1550 nm) and quick response (200 µs). Most importantly, with the intrinsic huge electric field derived from the polarization of ferroelectric polymer (P(VDF-TrFE)) gating, a low dark current of the photodetector can be achieved without additional gate bias. These studies present a crucial step for further practical applications for 2D semiconductors.

Lattice-Polarity-Driven Epitaxy of Hexagonal Semiconductor Nanowires.

Lattice-polarity-driven epitaxy of hexagonal semiconductor nanowires (NWs) is demonstrated on InN NWs. In-polarity InN NWs form typical hexagonal structure with pyramidal growth front, whereas N-polarity InN NWs slowly turn to the shape of hexagonal pyramid and then convert to an inverted pyramid growth, forming diagonal pyramids with flat surfaces and finally coalescence with each other. This contrary growth behavior driven by lattice-polarity is most likely due to the relatively lower growth rate of the (0001̅) plane, which results from the fact that the diffusion barriers of In and N adatoms on the (0001) plane (0.18 and 1.0 eV, respectively) are about 2-fold larger in magnitude than those on the (0001̅) plane (0.07 and 0.52 eV), as calculated by first-principles density functional theory (DFT). The formation of diagonal pyramids for the N-polarity hexagonal NWs affords a novel way to locate quantum dot in the kink position, suggesting a new recipe for the fabrication of dot-based devices.

Electrical spin injection from room-temperature ferromagnetic (Ga, Mn)N in nitride-based spin-polarized light-emitting diodes.

We present the electrical spin injection from room-temperature ferromagnetic (Ga, Mn)N in nitride-based spin-polarized light-emitting diodes. The electroluminescence spectra from the spin LED indicate the existence of the spin polarization via optical polarization of emitted light up to room temperature. This demonstrates that the spin injection from the (Ga, Mn)N layer into (In, Ga)N quantum wells was achieved persisting up to room temperature by comparing it with the magnetic field dependence of the Hall resistance, which is proportional to the out-of-plane magnetization. These results support that (Ga, Mn)N is an appropriate material for a spin injection source in room-temperature operating semiconductor spintronic devices.