Culaclii Stanislav, Wang Po-Min, Taccola Giuliano, Yang William, Bailey Brett, Chen Yan-Peng, Lo Yi-Kai, Liu Wentai
Department of Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States.
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, United States.
Front Neurosci. 2021 Jul 29;15:697731. doi: 10.3389/fnins.2021.697731. eCollection 2021.
Novel neural stimulation protocols mimicking biological signals and patterns have demonstrated significant advantages as compared to traditional protocols based on uniform periodic square pulses. At the same time, the treatments for neural disorders which employ such protocols require the stimulator to be integrated into miniaturized wearable devices or implantable neural prostheses. Unfortunately, most miniaturized stimulator designs show none or very limited ability to deliver biomimetic protocols due to the architecture of their control logic, which generates the waveform. Most such designs are integrated into a single System-on-Chip (SoC) for the size reduction and the option to implement them as neural implants. But their on-chip stimulation controllers are fixed and limited in memory and computing power, preventing them from accommodating the amplitude and timing variances, and the waveform data parameters necessary to output biomimetic stimulation. To that end, a new stimulator architecture is proposed, which distributes the control logic over three component tiers - software, microcontroller firmware and digital circuits of the SoC, which is compatible with existing and future biomimetic protocols and with integration into implantable neural prosthetics. A portable prototype with the proposed architecture is designed and demonstrated in a bench-top test with various known biomimetic output waveforms. The prototype is also tested to deliver a complex, continuous biomimetic stimulation to a rat model of a spinal-cord injury. By delivering this unique biomimetic stimulation, the device is shown to successfully reestablish the connectivity of the spinal cord post-injury and thus restore motor outputs in the rat model.
与基于均匀周期性方波脉冲的传统协议相比,模仿生物信号和模式的新型神经刺激协议已显示出显著优势。与此同时,采用此类协议治疗神经疾病需要将刺激器集成到小型可穿戴设备或植入式神经假体中。不幸的是,由于其产生波形的控制逻辑架构,大多数小型刺激器设计在提供仿生协议方面表现出无能力或能力非常有限。大多数此类设计都集成到单个片上系统(SoC)中,以减小尺寸并选择将其实现为神经植入物。但是它们的片上刺激控制器是固定的,内存和计算能力有限,这使得它们无法适应幅度和时间变化以及输出仿生刺激所需的波形数据参数。为此,提出了一种新的刺激器架构,该架构将控制逻辑分布在三个组件层上——软件、微控制器固件和SoC的数字电路,它与现有的和未来的仿生协议兼容,并可集成到植入式神经假体中。设计了具有所提出架构的便携式原型,并在台式测试中用各种已知的仿生输出波形进行了演示。该原型还经过测试,可向脊髓损伤大鼠模型提供复杂的、连续的仿生刺激。通过提供这种独特的仿生刺激,该设备被证明能够成功重建损伤后脊髓的连接性,从而恢复大鼠模型中的运动输出。
Front Neurosci. 2021-7-29
Int IEEE EMBS Conf Neural Eng. 2019-3
IEEE J Solid-State Circuits. 2015-1-1
IEEE Trans Biomed Circuits Syst. 2013-12
Biomed Eng Online. 2017-8-14
IEEE Trans Biomed Circuits Syst. 2017-6
Annu Int Conf IEEE Eng Med Biol Soc. 2010
Biosensors (Basel). 2024-6-28
IEEE Trans Biomed Circuits Syst. 2017-8
IEEE Trans Biomed Circuits Syst. 2022-8
Analog Integr Circuits Signal Process. 2018-4
Int IEEE EMBS Conf Neural Eng. 2019-3
J Neural Eng. 2018-7-19