Self-confocal NIR-II fluorescence microscopy for multifunctional in vivo imaging

Fluorescence imaging in the second near-infrared window(NIR-II,900–1880 nm)with less scattering background in biological tissues has been combined with the confocal microscopic system for achieving deep in vivo imaging with high spatial resolution.However,the traditional NIR-IIfluorescence confocal microscope with separate excitation focus and detection pinhole makes it possess low confocal e±ciency,as well as di±cultly to adjust.Two types of upgraded NIR-IIfluorescence confocal microscopes,sharing the same pinhole by excitation and emission focus,leading to higher confocal e±ciency,are built in this work.One type is-ber-pinhole-based confocal microscope applicable to CW laser excitation.It is constructed forfluorescence intensity imaging with large depth,high stabilization and low cost,which could replace multiphotonfluorescence microscopy in some applications(e.g.,cerebrovascular and hepatocellular imaging).The other type is air-pinhole-based confocal microscope applicable to femtosecond(fs)laser excitation.It can be employed not only for NIR-IIfluorescence intensity imaging,but also for multi-channelfluorescence lifetime imaging to recognize different structures with similarfluorescence spectrum.Moreover,it can be facilely combined with multiphotonfluorescence microscopy.A single fs pulsed laser is utilized to achieve up-conversion(visible multiphotonfluorescence)and down-conversion(NIR-II one-photonfluorescence)excitation simultaneously,extending imaging spectral channels,and thus facilitates multi-structure and multi-functional observation.

稀疏二维突触和CMOS神经元混合神经形态硬件用于字符识别

神经形态计算能够有效处理数密集型任务.本文提出一种基于二维MoS2突触和CMOS神经元混合集成的神经形态硬件系统,可兼具二维突触器件的灵活可塑性与CMOS神经元的成熟度、可靠性.受益于混合集成,进一步提出特征提取和频率编码方式,降低了硬件所需的突触与神经元数量;基于板级全硬件实现高准确率(98.8%)和低功耗(11.4μW)的字符识别功能.本工作为构建高面积效率和能量效率的神经形态硬件系统提供了一个可行的方案.

Initial Experience of NIR-II Fluorescence Imaging-Guided Surgery in Foot and Ankle Surgery

Optical imaging in the second near-infrared(NIR-II;900-1880 nm)window is currently a popular research topic in the field of biomedical imaging.This study aimed to explore the application value of NIR-II fluorescence imaging in foot and ankle surgeries.A lab-established NIR-II fluorescence surgical navigation system was developed and used to navigate foot and ankle surgeries which enabled obtaining more high-spatial-frequency information and a higher signal-to-background ratio(SBR)in NIR-II fluorescence images compared to NIR-I fluorescence images;our result demonstrates that NIR-II imaging could provide higher-contrast and larger-depth images to surgeons.Three types of clinical application scenarios(diabetic foot,calcaneal fracture,and lower extremity trauma)were included in this study.Using the NIR-II fluorescence imaging technique,we observed the ischemic region in the diabetic foot before morphological alterations,accurately determined the boundary of the ischemic region in the surgical incision,and fully assessed the blood supply condition of the flap.NIR-II fluorescence imaging can help surgeons precisely judge surgical margins,detect ischemic lesions early,and dynamically trace the perfusion process.We believe that portable and reliable NIR-II fluorescence imaging equipment and additional functional fluorescent probes can play crucial roles in precision surgery.

金属氧化物对磷酸盐吸附的预测及分子机制

金属氧化物吸附剂由于其优异的性能和稳定性而成为有前景的磷酸盐去除材料。揭示纳米金属氧化物吸附磷酸盐的关键影响因素和吸附机制,对吸附磷酸盐金属基材料的设计至关重要.因此,本研究测定了32种金属氧化物的磷酸盐吸附量,通过ReliefF算法识别影响金属氧化物吸附磷酸盐的关键描述符,基于此建立了对磷酸盐吸附的预测模型,并采用密度泛函理论(density functional theory,DFT)模拟了磷酸盐在6种典型金属氧化物表面的吸附。结果显示,金属氧化物的电负性和阳离子电荷是决定磷酸盐吸附的关键因素采用6种机器学习算法构建了可筛选高吸附量金属氧化物的预测模型,决策树模型的训练集和验证集的预测准确度超过90%.DFT计算结果表明,金属原子d轨道与磷酸基团氧原子p轨道的相互作用是决定金属原子和氧原子成键的关键,阐明了金属原子d轨道能级决定磷酸基团吸附能的新机制.本研究所建立的预测模型可为筛选吸附磷酸盐的金属基材料提供依据,DFT计算揭示的分子机制可为吸附剂的设计提供理论基础.

An artificial neural network chip based on two-dimensional semiconductor

Recently,research on two-dimensional(2D)semiconductors has begun to translate from the fundamen-tal investigation into rudimentary functional circuits.In this work,we unveil the first functional MoS2 artificial neural network(ANN)chip,including multiply-and-accumulate(MAC),memory and activation function circuits.Such MoS2 ANN chip is realized through fabricating 818 field-effect transistors(FETs)on a wafer-scale and high-homogeneity MoS2 film,with a gate-last process to realize top gate structured FETs.A 62-level simulation program with integrated circuit emphasis(SPICE)model is utilized to design and optimize our analog ANN circuits.To demonstrate a practical application,a tactile digit sensing recognition was demonstrated based on our ANN circuits.After training,the digit recognition rate exceeds 97%.Our work not only demonstrates the protentional of 2D semiconductors in wafer-scale inte-grated circuits,but also paves the way for its future application in AI computation.

Perfecting and extending the near-infrared imaging window

In vivo fluorescence imaging in the second near-infrared window(NIR-II)has been considered as a promising technique for visualizing mammals.However,the definition of the NIR-II region and the mechanism accounting for the excellent performance still need to be perfected.Herein,we simulate the photon propagation in the NIR region(to 2340 nm),confirm the positive contribution of moderate light absorption by water in intravital imaging and perfect the NIR-II window as 900–1880 nm,where 1400–1500 and 1700–1880 nm are defined as NIR-IIx and NIR-IIc regions,respectively.Moreover,2080–2340 nm is newly proposed as the third near-infrared(NIR-III)window,which is believed to provide the best imaging quality.The wide-field fluorescence microscopy in the brain is performed around the NIRIIx region,with excellent optical sectioning strength and the largest imaging depth of intravital NIR-II fluorescence microscopy to date.We also propose 1400 nm long-pass detection in off-peak NIR-II imaging whose performance exceeds that of NIR-IIb imaging,using bright fluorophores with short emission wavelength.