Machine-Learning Mental-Fatigue-Measuring μm-Thick\nElastic Epidermal Electronics (MMMEEE)

Electrophysiological\n(EP) signals are key biomarkers for monitoring\nmental fatigue (MF) and general health, but state-of-the-art wearable\nEP-based MF monitoring systems are bulky and require user-specific,\nlabeled data. Ultrathin epidermal electrodes with high performance\nare ideal for constructing imperceptive EP sensing systems; however,\nthe lack of a simple and scalable fabrication delays their application\nin MF recognition. Here, we report a facile, scalable printing-welding-transferring\nstrategy (PWT) for printing μm-thickness micropatterned silver\nnanowires (AgNWs)/sticky polydimethylsiloxane, welding the AgNWs via\nplasmonic effect, and transferring the electrode to skin as tattoos.\nThe PWT provides electrodes with conformability, comfort, and stability\nfor EP sensing. Leveraging the facile and scalable PWT, we develop\n plug-and-play wireless multimodal epidermal electronics integrated\nwith an unsupervised transfer learning (UTL) scheme for MF recognition\nacross various users. The UTL adaptively minimizes the intersubject\ndifference and achieves high accuracy, without demand of expensive\ncomputation and labels from target users.