Bioelectronics integrates electronic devices and biological systems with the ability to monitor and control biological processes. From homeostasis to sensorimotor reflexes, closed-loop control with feedback is a staple of most biological systems and fundamental to life itself. Apart from a few examples in bioelectronic medicine, the closed-loop control of biological processes using bioelectronics is not as widespread as in nature. We note that adoption of closed-loop control using bioelectronics has been slow because traditional control methods are difficult to apply to the complex dynamics of biological systems and their sensitivity to environmental changes. Here, we postulate that machine learning can greatly enhance the reach of bioelectronic closed-loop control and we present the advantages of machine learning compared to traditional control approaches. Potential applications of machine learning-based closed-loop control with bioelectronics include further impact in bioelectronic medicine and fine tuning of reactions and products in synthetic biology.

中文翻译：

---

通过机器学习将生物控制扩展到生物电子学

生物电子学集成了具有监视和控制生物过程能力的电子设备和生物系统。从动态平衡到感觉运动反射，带反馈的闭环控制是大多数生物系统的基础，也是生命本身的基础。除了生物电子医学中的一些例子外，使用生物电子学进行生物过程的闭环控制并不像在自然界中那样普遍。我们注意到，由于传统的控制方法难以应用于生物系统的复杂动力学及其对环境变化的敏感性，因此使用生物电子技术进行闭环控制的速度一直很慢。这里，我们假设机器学习可以极大地扩展生物电子闭环控制的范围，并且与传统控制方法相比，我们展示了机器学习的优势。基于机器学习的生物电子闭环控制的潜在应用包括对生物电子医学的进一步影响以及对合成生物学中反应和产物的微调。