With the rapidly aging society and increased concern for personal cardiovascular health, novel, flexible electrodes suitable for electrocardiogram (ECG) signal monitoring are in demand. Based on the excellent electrical and mechanical properties of graphene and the rapid development of graphene device fabrication technologies, graphene-based ECG electrodes have recently attracted much attention, and many flexible graphene electrodes with excellent performance have been developed. To understand the current research progress of graphene-based ECG electrodes and help researchers clarify current development conditions and directions, we systematically review the recent advances in graphene-based flexible ECG electrodes. Graphene electrodes are classified as bionic, fabric-based, biodegradable, laser-induced/scribed, modified-graphene, sponge-like, invasive, etc., based on their design concept, structural characteristics, preparation methods, and material properties. Moreover, some categories are further divided into dry or wet electrodes. Then, their performance, including electrode–skin impedance, signal-to-noise ratio, skin compatibility, and stability, is analyzed. Finally, we discuss possible development directions of graphene ECG electrodes and share our views.

Electronic skin (e-skin) has attracted tremendous interest due to its diverse potential applications, including in physiological signal detection, health monitoring, and artificial throats. However, the major drawbacks of traditional e-skin are the weak adhesion of substrates, incompatibility between sensitivity and stretchability, and its single function. These shortcomings limit the application of e-skin and increase the complexity of its multifunctional integration. Herein, the synergistic network of crosslinked SWCNTs within and between multilayered graphene layers was directly drip coated onto the PU thin film with self-adhesion to fabricate versatile e-skin. The excellent mechanical properties of prepared e-skin arise from the sufficient conductive paths guaranteed by SWCNTs in small and large deformation under various strains. The prepared e-skin exhibits a low detection limit, as small as 0.5% strain, and compatibility between sensitivity and stretchability with a gauge factor (GF) of 964 at a strain of 0–30%, and 2743 at a strain of 30–60%. In physiological signals detection application, the e-skin demonstrates the detection of subtle motions, such as artery pulse and blinking, as well as large body motions, such as knee joint bending, elbow movement, and neck movement. In artificial throat application, the e-skin integrates sound recognition and sound emitting and shows clear and distinct responses between different throat muscle movements and different words for sound signal acquisition and recognition, in conjunction with superior sound emission performance with a sound spectrum response of 71 dB (f = 12.5 kHz). Overall, the presented comprehensive study of novel materials, structures, properties, and mechanisms offers promising potential in physiological signals detection and artificial throat applications.