The Innovation
·
2024
·
IF 33.2
Hang Yin,Jincong Pang,Shan Zhao,Haodi Wu,Zihao Song,Xing Li,Zhiping Zheng,Ling Xu,Jiang Tang,Guangda Niu
X-ray detection is crucial across various sectors, but traditional techniques face challenges such as inefficient data transmission, redundant sensing, high power consumption, and complexity. The innovative idea of a retinomorphic X-ray detector shows great potential. However, its implementation has been hindered by the absence of active layers capable of both detecting X-rays and serving as memory storage. In response to this critical gap, our study integrates hybrid perovskite with hydrion-conductive organic cations to develop a groundbreaking retinomorphic X-ray detector. This novel device stands at the nexus of technological innovation, utilizing X-ray detection, memory, and preprocessing capabilities within a single hardware platform. The core mechanism underlying this innovation lies in the transport of electrons and holes within the metal halide octahedral frameworks, enabling precise X-ray detection. Concurrently, the hydrion movement through organic cations endows the device with short-term resistive memory, facilitating rapid data processing and retrieval. Notably, our retinomorphic X-ray detector boasts an array of formidable features, including reconfigurable short-term memory, a linear response curve, and an extended retention time. In practical terms, this translates into the efficient capture of motion projections with minimal redundant data, achieving a compression ratio of 18.06% and an impressive recognition accuracy of up to 98.6%. In essence, our prototype represents a paradigm shift in X-ray detection technology. With its transformative capabilities, this retinomorphic hardware is poised to revolutionize the existing X-ray detection landscape.
研究领域:
材料科学
·
医学影像
·
电子工程
✓ 已验证
产品关联
DOI
PubMed
高影响因子
Small
·
2023
·Vol. 20
·期 21
·2309626
·
IF 13.3
Boyu Xu,Dan Guo,Weikang Dong,Huiying Gao,Peng Zhu,Zhiwei Wang,Kenji Watanabe,Takashi Taniguchi,Zhaochu Luo,Fawei Zheng,Shoujun Zheng,Jiadong Zhou
Floating gate memory (FGM), composed of van der Waals (vdW) junctions with an atomically thin floating layer for charge storage, is widely employed to develop logic-in memories and in-sensor computing devices. Most research efforts of FGM are spent on achieving long-term charge storage and fast reading/writing for flash and random-access memory. However, dynamic modulation of memory time via a tunneling barrier and vdW interfaces, which is critical for synaptic computing and machine vision, is still lacking. Here, a van der Waals short-term memory with tunable memory windows and retention times from milliseconds to thousands of seconds is reported, which is approximately exponentially proportional to the thickness h-BN (hexagonal boron nitride) barrier. The specific h-BN barrier with fruitful gap states provides charge release channels for trapped charges, making the vdW device switchable between positive photoconductance and negative photoconductance with a broadband light from IR to UV range. The dynamic short-term memory with tunable photo response highlights the design strategy of novel vdW memory vis interface engineering for further intelligent information storage and optoelectronic detection.
研究领域:
材料科学
·
电子工程
·
光电子学
✓ 已验证
DOI
高影响因子
Journal of Physical Chemistry Letters
·
2025
·
IF 4.6
Ziqi Jia,Wenmin Zhong,Kui Zhou,Wei Zeng,Yan Li,Zihao Feng,Haozhe Xue,Pengfei Zhao,Xue Chen,Hongxiang Wang,Xingke Cai,Shuangmei Xue,Yongbiao Zhai,Ziyu Lv,Yan Yan,Meng Zhang,Xueqing Yang,Guanglong Ding,Su-Ting Han,Chi-Ching Kuo,Ye Zhou
Neuromorphic computing systems that integrate memory and computation offer a solution to the limitations of traditional von Neumann architectures. Optoelectronic synaptic transistors, responding to both optical and electrical signals, enable multifunctional operation with low power consumption. However, challenges such as short data retention and low processing efficiency remain. This study presents an optoelectronic synaptic transistor utilizing two-dimensional (2D) MoS2, 2D zeolitic imidazolate framework (ZIF) Zn2(bim)4, and gold (Au) nanoparticles (NPs) as semiconductor, tunneling layer, and floating gate materials, respectively. By adjusting the tunneling layer thickness, the charge-blocking capacity of Zn2(bim)4 is modulated, improving long-term data retention. The optoelectronic properties of MoS2 and the charge-trapping ability of Au NPs enable the transistor to mimic synaptic behaviors such as postsynaptic current (PSC), long-term potentiation (LTP), and transition from short-term to long-term memory (STM-LTM). This device can also be integrated into an artificial neural network (ANN) for smart healthcare applications, achieving 88.1% accuracy in electrocardiogram classification through optoelectronic dual-mode stimulation.
研究领域:
材料科学
·
纳米技术
·
光电子学
·
神经科学
·
人工智能
·
生物医学工程
✓ 已验证
产品关联
DOI
PubMed
高影响因子
Small
·
2025
·Vol. 21
·期 35
·2506100
·
IF 12.1
Zhengjun Liu,Yuxiao Fang,Zhaohui Cai,Yijun Liu,Xinyu Zhao,Qing Liu,Lixing Kang,Jianyu Yuan,Chun Zhao
Artificial neural networks based on artificial synaptic thin-film transistors (ASTFTs) offer promising opportunities for efficient and scalable information processing. However, implementing both spatial and temporal processing within a unified system requires complex hybrid network architectures, placing high demands on the tunability of individual synaptic devices. Here, we report a reconfigurable photosensitive synaptic transistor composed of FAPbI₃ colloidal quantum dots (CQDs) and an InO x channel, integrated with a modulation scheme tailored for hybrid convolutional-sequence neural computing. The device enables mode-specific plasticity through programmable optical and electrical stimuli. For spatial signal processing, it achieves a long-term potentiation (LTP) linearity of 0.062, long-term depression (LTD) linearity of 0.143, and an asymmetry ratio (AR) of 0.185. For temporal learning, the dynamic fading memory (FM) time constant is tunable from 42 to 75 ms, enabling dynamic short-term memory (STM) behavior. Utilizing these synaptic features, we construct a hybrid CNN-GRU neuromorphic system that achieves 94.2% accuracy in real-time gesture recognition within only 7 training epochs, using a custom dataset over 200 times larger than conventional benchmarks.This work demonstrates a unified neuromorphic hardware platform capable of spatiotemporal cognition, offering new strategies for the development of intelligent optoelectronic systems.
研究领域:
神经形态工程
·
光电子学
·
人工智能
·
神经网络
·
材料科学
·
认知计算
✓ 已验证
产品关联
DOI
PubMed
高影响因子
危险品化学品经营许可证（带存储）