基于双解码路径DD-UNet的脑肿瘤图像分割算法

针对医学图像中病灶区域尺度不一、边界模糊和周围组织强度不均匀所导致的分割精度降低问题,提出了一种基于双解码器的脑肿瘤图像分割模型。为了增强特征的表征力,提出了高阶微分残差模块并使用不同空洞率的扩张卷积用于提取特征编码,提高了网络模型的分割性能;引入上下文语义信息感知模块(multi scale dilation, MSD),从不同的目标尺度中提取更多的精细信息,提高了对结构细节信息的捕获能力,同时减少了编解码器之间的特征差异;在空间解码路径中使用选择性聚合空间注意力模块(spatial aggregation attention module, SAAM),增加了对有效空间特征的权重比例,减少了无效的特征干扰。在脑肿瘤数据集上进行了实验验证,实验结果表明,所提算法的Dice系数、平均交并比、敏感性、特异性、准确率等指标分别为：93.35%、90.71%、91.15%、99.94%、96.75%。     

基于无监督学习的低照度图像增强算法

针对目前低照度图像增强算法存在恢复细节丢失、网络复杂度高和配对数据集获取难度大等问题,提出了一种基于无监督学习的图像增强算法。在YIQ色彩空间中,通过构建的轻量化网络和幂指函数计算亮度通道Y的增强曲线,从而获得曝光较差区域增强和高光区域遏制的图像。该网络使用的无参考损失函数可以隐式地评估图像增强质量并驱动网络学习。实验对比结果表明,该算法在可训练参数和模型权重仅占9.5 k/88 kB的情形下,在视觉效果与图像质量指标上都取得了具有竞争力的结果。     

Promote Intratumoral Drug Release and Penetration to Counteract Docetaxel-Induced Metastasis by Photosensitizer-Modified Red Blood Cell Membrane-Coated Nanoparticle

The red blood cell membrane (RBCm) provides tight protection, lowers the immunogenicity, and prolongs the circulation time of drugs in vivo when acting as the coating of drug delivery systems. However, the cellular uptake and release of drugs are hindered by RBCm. Docetaxel (DTX) is the first-line medicine for treating triple-negative breast cancer (TNBC), but it induces tumor metastasis. To solve these dilemmas, in this study, the photosensitizer 1,1-dioctadecyl-3,3,3,3-tetramethylindotricarbocyanine iodide (DiR)-modified RBCm (DM) is prepared, which is coated onto a hybrid micelle consisting of the prodrugs of DTX and the anti-metastasis agent calcitriol (CTL), obtaining a nanoparticle, named HDC-DM. In a 4T1 tumor-bearing mouse model, after injecting HDC-DM, the intratumoral DTX and CTL concentrations are increased by 1.7 and 2.5 times compared with the free drug groups. After irradiating tumors with near-infrared laser, DiR elicits the photothermal effect, triggering the rupture of RBCm and drug release, promoting drug penetration in tumors, and inducing immunogenic cell death. The tumor growth inhibition rate is 77%, and the formation of lung metastases is reduced by 82%, with good biocompatibility. It is suggested that the combination of phototherapy, chemotherapy, and anti-metastatic therapy using HDC-DM is expected to be a powerful strategy for treating TNBC.     

One-Component Artificial Gustatory System Based on Hydrogen-Bond Organic Framework for Discrimination of Versatile Analytes

Hydrogen-bond organic frameworks (HOFs) with excellent structural and luminescent properties have emerged as a promising material for the construction of fluorescence sensors. However, designing a facile, universal and high throughput sensor with multiplex detection capacity still remains challenging. Herein, a one-component sensor array is constructed that mimics natural gustatory system for accurate and high-throughput discrimination and identification of versatile analytes. HOF as a single sensing element greatly simplifies the probe preparation in sensor array and detection procedure. Metal ions, proteins and bacteria as the model targets are rapid and accurately discriminated, presenting the universality of the system. Particularly, the system is successfully used for the classification of antibiotic mechanisms. The study expands the application scope of HOFs and provides a facile and universal system for sensing applications.     

Minimizing Energy Barrier in Intermediate Connection Layer for Monolithic Tandem WPeLEDs with Wide Color Gamut

Perovskite light-emitting diodes (PeLEDs) show promising prospects in the wide color gamut display owing to their ultra-narrow full width at half maximum (FWHM). However, up to now, all perovskite white LEDs integrated by standard red, green, and blue perovskite emitters, namely, monolithic white PeLEDs (WPeLEDs), have been rarely reported, owing to facing some issues, e.g., solvent incompatibility in solution technique, ion exchange, and energy transfer between different emission centers. Herein, centered on these issues, an optimal intermediate connection layer (ICL) of Po-T2T/LiF/Ag/HAT-CN/MoO₃ is adopted to successfully develop monolithic tandem multicolor PeLEDs and WPeLEDs for the first time. The multicolor PeLEDs can achieve the best external quantum efficiency of 1.8% and the highest luminance of 4844 cd m⁻². Besides, the red/green/blue (R/G/B) monolithic tandem WPeLED shows a standard white International Commission on Illumination coordinate of (0.33, 0.33) and achieves an extremely wide color gamut reaching  National Television Standards Committee of 130%. This study is the first to realize the standard R/G/B co-electroluminescence in a monolithic perovskite device and offers a feasible strategy for developing wide-color gamut perovskite displays.     

Interfacial Proton Supply/Filtration Regulates the Dynamics of Electrocatalytic Nitrogen Reduction Reaction: A Perspective

As a promising energy carrier, ammonia synthesis by electrocatalytic nitrogen reduction reaction (eNRR) is a promising green and low-carbon ammonia synthesis strategy that can replace the traditional Haber–Bosch process. However, the development of eNRR processes is mainly severely constrained by competitive hydrogen evolution reaction (HER), and the corresponding strategies to inhibit this adverse side reaction to obtain high eNRR selectivity are still limited. In addition, for this complex reaction involving gas–liquid–solid three-phase interface and proton/electron transfer, it is great significance to analyze and summarize the existing inhibition HER strategies from the viewpoint of dynamics. In view of this, this work reviews proton supply/filtration regulation strategy in catalytic system, allowing a systematic survey of the literature focusing on interface membrane regulation (inorganic membrane and organic membrane), electrolyte regulation (metal-mediated strategy and electrolyte ion regulation strategy) and system device design (electrode structure design and electrolytic cell device design). Constructive catalytic system design guidance is also suggested to inhibit hydrogen evolution and improve NH₃ selectivity, aiming for scalable and economically feasible applications.     

Vehicle and pedestrian detection method based on improved YOLOv4-tiny

Aiming at the problem of low detection accuracy of vehicle and pedestrian detection models, this paper proposes an improved you only look once v4 (YOLOv4)-tiny vehicle and pedestrian target detection algorithm. Convolutional block attention module (CBAM) is introduced into cross stage partial Darknet-53 (CSPDarknet53)-tiny module to enhance feature extraction capabilities. In addition, the cross stage partial dense block layer (CSP-DBL) module is used to replace the original simple convolutional module superposition, which compensates for the high-resolution characteristic information and further improves the detection accuracy of the network. Finally, the test results on the BDD100K traffic dataset show that the mean average precision (mAP) value of the final network of the proposed method is 88.74%, and the detection speed reaches 63 frames per second (FPS), which improves the detection accuracy of the network and meets the real-time detection speed.     

Balancing MXene Surface Termination and Interlayer Spacing Enables Superior Microwave Absorption

Surface chemistry and interlayer engineering determines the electrical properties of 2D MXene. However, it remains challenging to regulate the surface and interfacial chemistry of MXene simultaneously. Herein, simultaneous modulation of Ti₃C₂T_x MXene surface termination and layer spacing by alkali treatment are achieved. The electrical and electromagnetic properties of Ti₃C₂T_x are investigated in detail with respect to KOH and ammonia concentration dependence. A high concentration of KOH caused the Ti₃C₂T_x layer spacing to expand to 13.7 Å and the surface O/F ratio to increase to 33.84. Because of its weaker ionization effect, ammonia provides finer tuning compared to the drastic intercalation of KOH with a thorough sweeping of the F-containing groups. Ti₃C₂T_x is enriched with conductive -OH termination after ammonia treatment, which achieves an effective balance with the increased interlayer resistance. Therefore, NH₃H₂O-Ti₃C₂T_x achieves broad-band impedance matching and exhibits an efficient microwave loss of −49.1 dB at a low thickness of 1.7 mm, with an effective frequency bandwidth of 3.9 GHz. The results herein optimize the electrical properties of Ti₃C₂T_x using surface and interfacial chemistry to achieve broad microwave absorption, providing a framework for enhancing the electromagnetic wave loss of intrinsic MXene.     

Self-Standing Covalent Organic Framework Membranes for H2/CO2 Separation

Covalent organic frameworks (COFs) are proposed as promising candidates for engineering advanced molecular sieving membranes due to their precise pore sizes, modifiable pore environment, and superior stability. However, COFs are insoluble in common solvents and do not melt at high temperatures, which presents a great challenge for the fabrication of COF-based membranes (COFMs). Herein, for the first time, a new synthetic strategy is reported to prepare continuous and intact self-standing COFMs, including 2D N-COF membrane and 3D COF-300 membrane. Both COFMs show excellent selectivity of H₂/CO₂ mixed gas (13.8 for N-COF membrane and 11 for COF-300 membrane), and especially ultrahigh H₂ permeance (4319 GPU for N-COF membrane and 5160 GPU for COF-300 membrane), which is superior to those of COFMs reported so far. It should be noted that the overall separation performance of self-standing COFMs exceeds the Robeson upper bound. Furthermore, a theoretical study based on Grand Canonical Monte Carlo (GCMC) simulation is performed to explain the excellent separation of H₂/CO₂ through COFMs. Thus, this facile preparation method will provide a broad prospect for the development of self-standing COFMs with highly efficient H₂ purification.