China
Africa
Explore chapters and articles related to this topic
In vitro studies
Published in Ze Zhang, Mahmoud Rouabhia, Simon E. Moulton, Conductive Polymers, 2018
A. Lee Miller, Huan Wang, Michael J. Yaszemski, Lichun Lu
There is a great deal of variability of neurons. They vary in function, size (small, medium, large, and giant), and shape (spherical, conical, star shaped, fusiform, polyhedral, and pyramidal). Neurons can be classified by their morphology or function. A broad and basic classification of neurons categorizes them based on their functions into motor, sensory, or interneurons. Motor neurons carry information from the central nervous system to organs, glands, and muscles (hence they are also termed efferent neurons). Sensory neurons send information to the central nervous system from internal organs or from external stimuli (hence they are also termed afferent neurons). Interneurons relay signals between motor and sensory neurons. They communicate with each other via chemical and electrical synapses in a process known as synaptic transmission (Shepherd 2000).
Emerging memristive neurons for neuromorphic computing and sensing
Published in Science and Technology of Advanced Materials, 2023
Zhiyuan Li, Wei Tang, Beining Zhang, Rui Yang, Xiangshui Miao
In terms of neuromorphic-sensing system, how to match the varied types of sensors with the memristive neuron is a huge challenge. A scaling resistor has been incorporated to configurate the neuronal circuit (see Section 6). However, this scheme only focuses on converting perceptual signals into spikes, sacrificing the performance of the sensor system. Therefore, it is needed to evaluate and compare the sensor characteristics, specifically the sensitivity, selectivity, and reliability, in artificial sensory neurons, to further realize highly matching of signals between biology and electronics. In addition, mechanical compliance of artificial sensory system is important for the biocompatible neuromorphic electronics. The artificial sensory neuron should be tolerant of the mechanical deformation. The structures and materials of flexible device need to be optimized to achieve stable neuronal responses based on various sensory signals regardless of mechanical strain. In the future, a stretchable and biocompatible artificial neuron will expand to skin-attachable and implantable neuromorphic electronics for wearable computing, health monitoring, and sensorimotor neural signal transmission.
Recent advances in neuromorphic transistors for artificial perception applications
Published in Science and Technology of Advanced Materials, 2023
Inspired by the supermodal sensory fusion in sensory nervous system, Wan et al. [48] developed a bimodal artificial sensory neuron (BASE) based on ionic/electronic hybrid neuromorphic electronic devices to implement the visual-haptic fusion, as schematically shown in Figure 22(a). The BASE collects optical and pressure information from photodetector and pressure sensors, respectively. Then, the bimodal information can be transmitted through an ionic cable. At last, information is integrated and induces post-synaptic currents on a synaptic transistor. Figure 22(b) shows the responses of the bimodal artificial sensory neuron to external stimuli. In addition, based on bimodal information sensory cues, sensory neurons can be stimulated at multiple levels to successfully simulate motion control, manipulating skeletal myotubes and a robotic hand, as shown in Figure 22(c). More interestingly, by simulating the multi-transparency pattern recognition task, the enhanced recognition ability realized on the fusion of visual/haptic cues is confirmed, as shown in Figure 22(d). The results show that the highly integrated perceptual system constructed by simulating sensory fusion at the neuron-level has far-reaching significances for neurorobotics and artificial intelligence.
Emerging electrolyte-gated transistors for neuromorphic perception
Published in Science and Technology of Advanced Materials, 2023
Cui Sun, Xuerong Liu, Qian Jiang, Xiaoyu Ye, Xiaojian Zhu, Run-Wei Li
The human visual system receives over 80% of the information from the environment. Emulation of biological visual function is important for the development of machine vision technologies that finds broad applications in autonomous driving, video analysis, and intelligent manufacturing [67,131,132]. Kim et al. [133] reported an artificial nervous system (Figure 7(a)) consisting of a quantum dot (QD) photodiode, a retentive electric double layer (EDL) transistor, and CMOS-based neuron circuits, corresponding to the visual receptor, synapse, and neuron, respectively. The visual receptor and sensory neuron circuits convert the incident light signals into electrical signals, which were analyzed by the neuronal circuit. The EGT-based artificial synapse shows the strengthened synaptic strength corresponding to LTM, after sufficient optical stimulation, which accelerates the action potential generation speed of the downstream artificial neuron. By connecting the artificial nervous system with a robot hand, an artificial stimulus-response system capable of mimicking the conscious response behavior is constructed. After sufficient training, the time required for activation of the robot hand to the light stimulus is reduced from 2.56 s to 0.23 s. This artificial stimulus-response system provides a new perspective for the development of artificial intelligence-based systems for neurological disorder patients.