郑州大学王其炎&卫荣汉&西安交大马伟‖【ACS Sensors】未来神经接口:AI赋能仿生芯片,实现多巴胺动态的实时解码与闭环调控
- 2026-04-04 09:18:32
点击蓝字
关注我们
介绍 /Introduction
研究图文/Introduction
Figure 1. Biomimetic microfluidic electrochemical chip with a closed-loop AI decision engine system enables dopamine dynamics tracking and intervention. The platform combines wrinkled MoS2 electrodes, whose intestinal mucosa inspired morphology enhances sensing, with a closed-loop AI decision engine that reads dopamine signals from the chip, recognizes affective states based on those signals, and executes interventions via machine-driven decision-making. The system is successfully validated in murine models, tracking dopamine fluctuations in response to water and sugar stimuli and demonstrating its capacity to autonomously adjust behavioral strategies.
Figure 2. Bioinspired architecture of the intestine-like wrinkle MoS2 electrode. (a) Schematic illustration of the wrinkled-MoS2 structure as working electrode over the microfluidic electrochemical chip. (b,d) SEM images of the wrinkled-MoS2 and sheet-MoS2 electrode, respectively. (c,e) Fluid simulations performed in fluent depicting the velocity profiles of the wrinkled-MoS2 and sheet-MoS2 electrode in pure water, indicating that wrinkled-MoS2 produces significantly more micro-vortices. (f) Molecular dynamics simulation showing the diffusion of dopamine (DA) molecules over the electrodes, and its results of the adsorption amount (g) and adsorption energy (h). (i) Amperometric responses of the wrinkled-MoS2 elec‐ trode at various concentrations at 0.15 V. (j) Determination of the limit of detection (LOD) for the electrodes. (k) Evaluation of the anti-interference performance of the wrinkled-MoS2 electrode for dopamine detection, demonstrating low interference from ascorbic acid (AA), uric acid (UA), lacticacid (Lac), glucose (Gluc), and urea.
Figure 3. Microfluidic electrochemical chip device system based on the wrinkled-MoS2 electrode.
结论 /Conclusion
综上所述，本文开发了一种将仿生微流控设计与自适应AI控制相结合的闭环神经化学传感系统。褶皱的肠道状二硫化钼电极仅在4.1微升生物体液中即可实现37 nM的超灵敏多巴胺检测，通过增强的传质作用克服了分析物补充的限制。该系统嵌入无线可穿戴平台，能够实时持续监测神经化学动态。AI引擎结合了基于XGBoost的分类器与基于LLaMA的语义推理，实现了毫秒级神经化学波动的解读以及自主、行为适应性的干预。在自由活动大鼠上的体内研究证实了由内源性多巴胺动力学驱动的精准闭环调控。这一智能系统 exemplifies 了神经化学传感的新范式，其中实时感知、解读与干预无缝集成，用于精准神经调控和适应性神经精神疾病治疗。
文献直通车 /Literature information
Biomimetic Electrochemical Chip Integrated with Closed-Loop AI for Dynamic Dopamine Decoding and NeuromodulationXinran Li, Xiao Wu, Qiyan Wang,* Yunhe Zhu, Liyuan Zhang, Yifan Song, Benkun Lv, Jiyu Chen, Wu Fan, Guobi Chai, Qidong Zhang,* Ronghan Wei,* and Wei Ma*原文链接：https://doi.org/10.1021/acssensors.5c03495
点击左下角“阅读原文”，轻松获取原文信息
欢迎扫码关注
科研路上，我们一路同行！
免责声明
「原创」仅代表原创编译，水平有限，仅供学术交流，本平台不主张原文的版权，如有侵权，请联系删除。文献解读或作者简历如有疏漏之处，我们深表歉意，请作者团队及时联系chemsensing@163.com，我们会在第一时间进行修改或撤稿重发，感谢您的谅解！
END
投稿、荐稿、合作请联系邮箱
chemsensing@163.com
