Continuous Monitoring with AI-Enhanced BioMEMS Sensors: A Focus on Sustainable Energy Harvesting and Predictive Analytics

Mingchen Cai, Hao Sun, Tianyue Yang, Hongxin Hu, Xu‐Bing Li, Jia YuanMicromachines 2025 · DOI

TL;DR

This review explores the exciting convergence of BioMEMS sensors, sustainable energy harvesting, and AI for continuous, intelligent monitoring. It highlights how combining advanced materials, self-powered systems, and sophisticated AI analytics is paving the way for next-generation health and environmental sensing platforms.

What Problem Does This Paper Address?

The paper addresses the critical challenge of enabling truly autonomous and intelligent continuous monitoring systems. Current BioMEMS sensors, while advanced, face limitations: they often rely on bulky, short-lived batteries, and the massive data they generate is difficult to interpret in real-time. This limits their application in resource-scarce environments or for long-term, unobtrusive health monitoring. The paper aims to bridge this gap by reviewing the state-of-the-art in self-powered BioMEMS and AI-driven data analysis, proposing a roadmap for integrated, intelligent sensing.

Methods

This review synthesizes existing research across three key areas:

  • BioMEMS Sensor Evolution: It details advancements in sensing materials (nanomaterials like carbon black, metal nanoparticles, quantum dots, CNTs, nanowires, graphene, and MXenes), micro/nano-scale architectures, and fabrication techniques that enhance sensitivity, flexibility, and biocompatibility for continuous monitoring.
  • Sustainable Energy Harvesting: The paper examines self-powered solutions for BioMEMS, focusing on:
  • Piezoelectric Nanogenerators (PENGs): Converting mechanical energy into electricity.
  • Triboelectric Nanogenerators (TENGs): Harnessing energy from contact and friction.
  • Moisture Electricity Generators (MEGs): Generating power from ambient humidity. These technologies aim to eliminate reliance on traditional batteries.

  • AI for Predictive Analytics: The review discusses the role of AI, particularly machine learning (ML) and deep learning (DL), in interpreting BioMEMS data. This includes classic ML algorithms, deep learning models (DNNs, CNNs, RNNs), large language models (LLMs), and large multimodal models (LMMs) for tasks like signal processing, anomaly detection, and predictive modeling. The paper also touches upon near-sensor and in-sensor computing for efficient data processing.

Datasets

Not explicitly discussed in the paper, as it is a review article synthesizing existing research rather than presenting new experimental results.

Key Insights and Ideas

  • Synergy is Key: The true power lies not in individual advancements but in their synergistic integration – combining advanced materials, ubiquitous energy harvesting, and intelligent AI analytics.
  • Beyond Batteries: The future of BioMEMS is self-powered. Energy harvesting technologies are maturing, offering viable alternatives to conventional batteries for continuous operation.
  • AI as the "Brain": AI is transforming BioMEMS from mere data collectors into intelligent systems capable of real-time interpretation, prediction, and decision-making.
  • Edge Computing is Crucial: For wearable and implantable devices, processing data closer to the sensor (near-sensor and in-sensor computing) is essential to reduce latency, power consumption, and data transmission burdens.
  • Multimodality is the Next Frontier: Large multimodal models hold promise for integrating diverse data streams (images, audio, text, sensor data) for a more holistic understanding of physiological and environmental states.

Key Results

As this is a review paper, it doesn't present new experimental results. Instead, it synthesizes findings from numerous studies, highlighting:

  • Advancements in BioMEMS Materials: Nanomaterials offer tunable properties for enhanced sensitivity and selectivity.
  • Progress in Energy Harvesting: PENGs, TENGs, and MEGs are demonstrating increasing efficiency and adaptability for wearable and flexible platforms.
  • AI's Impact on Data Analysis: ML and DL models are achieving high accuracy in tasks like physiological signal monitoring and biochemical concentration tracking, with LLMs and LMMs showing potential for complex data interpretation.
  • Examples of Integrated Systems: The paper showcases examples of wearable sensors, point-of-care devices, and environmental monitoring systems that integrate these technologies.

Limitations and Future Work

Limitations:

  • Energy Constraints: While energy harvesting is advancing, the inherent variability and sensitivity to ambient conditions remain a challenge for consistent power supply.
  • Computational Demands: Modern AI models, especially deep learning architectures, require significant computational resources and memory, posing integration challenges for microscale, battery-limited BioMEMS devices.
  • Data Scarcity and Heterogeneity: Obtaining labeled, high-quality biosensor data is difficult, leading to issues like model overfitting and poor generalization, especially in personalized or decentralized deployments.
  • Material Durability and Biocompatibility: Long-term stability, resistance to environmental degradation, and rigorous biocompatibility testing for novel materials are still areas needing further characterization.
  • Regulatory Hurdles: The pace of innovation often outstrips regulatory frameworks, creating bottlenecks for clinical translation.
  • Interpretability: Ensuring model transparency and explainability is crucial, especially for medical applications where regulatory acceptance and trust are paramount.

Future Work:

  • Co-optimization: Developing integrated systems that synergistically optimize energy harvesters, sensor materials, and embedded AI frameworks.
  • Multimodal Energy Systems: Creating hybrid energy harvesters that combine different mechanisms for more stable power output.
  • Domain-Adaptive Machine Learning: Developing AI models tailored to specific BioMEMS applications and data modalities.
  • Lightweight and Edge-Optimized AI: Research into model compression, quantization, and specialized hardware accelerators for efficient on-device inference.
  • Federated and Transfer Learning: Leveraging these techniques to address data scarcity and privacy concerns.
  • Digital Twins: Integrating BioMEMS data with digital twin technologies for real-time simulation and visualization of physiological and environmental dynamics.
  • Advanced Materials: Continued development of multifunctional, biocompatible, and long-lasting nanocomposites and biodegradable semiconductors.

Summary updated

Abstract

Continuous monitoring of environmental and physiological parameters is essential for early diagnostics, real-time decision making, and intelligent system adaptation. Recent advancements in bio-microelectromechanical systems (BioMEMS) sensors have significantly enhanced our ability to track key metrics in real time. However, continuous monitoring demands sustainable energy supply solutions, especially for on-site energy replenishment in areas with limited resources. Artificial intelligence (AI), particularly large language models, offers new avenues for interpreting the vast amounts of data generated by these sensors. Despite this potential, fully integrated systems that combine self-powered BioMEMS sensing with AI-based analytics remain in the early stages of development. This review first examines the evolution of BioMEMS sensors, focusing on advances in sensing materials, micro/nano-scale architectures, and fabrication techniques that enable high sensitivity, flexibility, and biocompatibility for continuous monitoring applications. We then examine recent advances in energy harvesting technologies, such as piezoelectric nanogenerators, triboelectric nanogenerators and moisture electricity generators, which enable self-powered BioMEMS sensors to operate continuously and reducereliance on traditional batteries. Finally, we discuss the role of AI in BioMEMS sensing, particularly in predictive analytics, to analyze continuous monitoring data, identify patterns, trends, and anomalies, and transform this data into actionable insights. This comprehensive analysis aims to provide a roadmap for future continuous BioMEMS sensing, revealing the potential unlocked by combining materials science, energy harvesting, and artificial intelligence.
BibTeX
@inproceedings{cai2025continuous,
  author = {Mingchen Cai and Hao Sun and Tianyue Yang and Hongxin Hu and Xu‐Bing Li and Jia Yuan},
  title = {{Continuous Monitoring with AI-Enhanced BioMEMS Sensors: A Focus on Sustainable Energy Harvesting and Predictive Analytics}},
  booktitle = {Micromachines},
  year = {2025},
  doi = {10.3390/mi16080902},
}
Figure