成果简介
在皮肤接触界面持续监测压力和温度，对于预防卧床患者的组织损伤和循环系统相关并发症至关重要。然而，现有的医疗压力传感器大多体积庞大、有线连接且依赖电池供电，这限制了其长期使用的适用性。本文，韩国电气研究院Myungwoo Choi等研究人员在《Mater Horiz》期刊发表名为“A battery-free, wireless graphene pressure sensor for machine learning-assisted posture classification and VR/AR visualization in smart healthcare environments”的论文，研究提出一种无电池无线多模态传感平台，其中单层石墨烯作为高性能压力传感活性层，实现高灵敏度（1.75×10⁻³ kPa⁻¹，应变系数=8.6）和卓越稳定性（超过1000次工作循环）。
该平台可在无外部电源的情况下，实时可逆地检测皮肤-设备界面的压力与温度。通过深度学习算法（特别是深度神经网络DNNs），采集信号可被分类为不同坐姿模式，从而实现患者状态的智能持续监测。此外，集成增强现实/虚拟现实（AR/VR）界面可实时可视化压力分布，支持沉浸式远程医疗监护。本研究综合推出基于石墨烯的智能传感平台，无缝融合无线操作、人工智能驱动分析及AR/VR可视化技术，为先进患者监测提供个性化交互式智能医疗解决方案。
图文导读
图1、 Design and operation of the battery-free wireless graphene pressure sensing system. (a) Exploded schematic of the wireless graphene pressure sensor composed of a soft PDMS micropad, encapsulated graphene active layer, supporting frame, and NFC-integrated fPCB for wireless data transmission and power harvesting. (b) Structural layout and optical image of the encapsulated graphene layer. (c) Finite element analysis (FEA) results of localized strain in the central graphene region under applied pressure. (d) Block diagram of the NFC-based wireless architecture for simultaneous pressure and temperature sensing. (e and f) Photographs of the assembled and PDMS-encapsulated device, demonstrating flexibility, skin conformity, and biocompatibility. (g) FEA of serpentine interconnects under stretching confirms mechanical stability. (h) Conceptual illustration of machine-learning-assisted posture classification and AR/VR-based visualization enabled by the graphene pressure sensing platform for real-time, patient-centered healthcare monitoring.
图2.Structural characterization and pressure-sensing performance of CVD-grown graphene used as the active layer. (a) Schematic illustration of the pressure-sensing active layer based on a CVD-grown graphene film encapsulated with a polyimide (PI) layer. (b) Optical microscopy (OM) image of the graphene film on a PI substrate with Au electrodes. (c) Schematic comparison between conventional CVD-grown graphene (C-Gr), which contains adlayers and grain boundaries, and single-layer, high-quality graphene (SH-Gr). (d) Scanning electron microscopy (SEM) images showing the surface morphologies of C-Gr and SH-Gr. (e) Raman spectra of C-Gr and SH-Gr. (f and g) High-resolution transmission electron microscopy (HR-TEM) images of C-Gr and SH-Gr, respectively. (h) Fractional change in the resistance of graphene pressure sensors under applied pressures of 1, 4, 7, and 10 kPa (C-Gr, white; SH-Gr, blue). (i) Resistance response of the graphene pressure sensor under loading and unloading at 15 kPa. (j) Dynamic response of the graphene pressure sensor under cyclic pressures of 0.5, 5, 10, and 15 kPa. (k) Long-term stability of the graphene pressure sensor over 1000 cyclic loadings at 5 kPa.
图3、Wireless characterization of the battery-free graphene pressure sensor. (a) Schematic illustration and block diagram of the battery-free, wireless sensing platform for simultaneous pressure and temperature monitoring. The system comprises three main components: (i) a battery-free wireless sensor integrating an NFC SoC for pressure (ADC1) and temperature (ADC2) measurement; (ii) a primary antenna and reader that provide wireless power transfer and data communication; and (iii) a laptop for real-time monitoring and visualization of post-processed data through a graphical user interface (GUI). (b) Response of the battery-free wireless sensor under loading and unloading at 15 kPa. (c) Change in ADC values of the wireless sensor under dynamic loadings of 10 and 6 kPa. (d) Reproducible ADC variations of the wireless sensor over four consecutive loading–unloading cycles at 10 kPa. (e) Change in ADC values of the wireless sensor as the temperature increases and decreases.
图4、Posture monitoring and classification using data from the battery-free wireless graphene pressure sensor. (a) Photograph showing the placement of five sensors on the lower back, hips, and thighs for posture monitoring. (b) Five sitting postures—normal, lean forward, lean backward, lean left, and lean right—used for the monitoring experiment. (c) Continuous measurements of pressure (P) and temperature (T) obtained from the sensors during sequential posture transitions. (d) Schematic of the deep neural network (DNN) model employed for posture classification. (e) Classification accuracy for five cross-validation splits (SP1–SP5). (f) Confusion matrix summarizing the prediction performance across the five sitting postures.
图6、Augmented reality (AR) and virtual reality (VR) monitoring system integrated with the battery-free wireless graphene sensors. (a) Block diagram of the AR/VR monitoring system. (b) Switching between AR and VR modes using the passthrough function on the Meta Quest 3 headset. (c) Operational sequence of the AR system in four steps. (d) Representative AR frame showing real-time bar overlays of pressure (red) and temperature (blue). (e) Representative VR frame for multi-user remote monitoring and evaluation.
小结
综上所述，本文开发了一种无电池无线压力传感系统，该系统采用单层高品质石墨烯作为高性能活性层，并整合了机器学习辅助分类与AR/VR可视化技术，适用于智能医疗应用。该石墨烯传感器具备高灵敏度、卓越的机械柔韧性及稳定的信号传导特性，可在无需外部电源的情况下，对皮肤界面上的细微压力与温度变化进行可靠的实时监测。通过将无线传感平台与人工智能驱动的数据分析相结合，成功实现了多种坐姿的分类识别，证明该系统能将复杂生理信号转化为具有实际意义的行为信息。此外，AR与VR可视化技术的融合实现了压力分布的直观实时远程评估，凸显该平台在沉浸式临床康复监测领域的应用潜力。这项工作展现了先进材料、智能算法与沉浸式可视化技术的协同融合，为构建连接人体生理与数字智能的个性化、预测性、交互式智能医疗系统开创了新范式。
文献:https://doi.org/10.1039/D5MH02270C
本文来自材料分析与应用，本文观点不代表石墨烯网立场，转载请联系原作者。
