CityUHK Researchers Develop Switchable-Flexibility Neural Interface for 3D Brain Access
The College of Biomedicine (CBM) is pleased to announce a significant breakthrough in neurotechnology. Through an interdisciplinary collaboration, Professor Peng Shi of the Department of Biomedical Engineering and Professor Yong Yang of the Department of Mechanical Engineering at City University of Hong Kong (CityUHK) have developed sFlex-Fold—a novel bioelectronic system that overcomes a fundamental limitation of existing neural interfaces: the inability to access the brain's complex three-dimensional (3D) structure without causing tissue damage.
AI-Powered Alloy DesignThe breakthrough is enabled by an AI-designed liquid metal alloy with a melting point precisely tuned to near body temperature. The team employed a machine learning architecture trained on 173 gallium-based alloy compositions, generating a library of over 35,000 formulations. After screening, the Ga97Ag3 alloy (GaAg) was selected, featuring a melting peak temperature of 36.2°C—critically close to body temperature.
From Rigid Implantation to Soft ConformationsFlex-Fold can be pre-moulded onto a 3D model of the target cortical region and retains its folded shape during cooling. The rigid device is then implanted as a single component, enabling non-destructive access to deep cortical sulci that are inaccessible to conventional flexible devices. Upon contact with the brain, the alloy melts, triggering a three-order-of-magnitude reduction in effective modulus—from 2.03 GPa to 1.66 MPa—allowing intimate conformation to both gyri and sulci.
Validated Performance and BiocompatibilityIn vivo tests demonstrated that sFlex-Fold captured neural signals with a correlation coefficient of 0.74—more than double the 0.32 of conventional flexible ECoG arrays. Stimulation experiments confirmed the device could reliably evoke neural responses from hard-to-reach sulcal regions. A four-week implantation study showed 100% survival among sFlex-Fold subjects, while 75% of animals implanted with conventional devices died. Histological analysis revealed minimal glial scar formation and neuroinflammation.
The research has been published in Science Advances (June 2026). By overcoming a key limitation in neural interfacing, this innovation brings new possibilities for the diagnosis and treatment of neurological disorders, promising tangible benefits for patients and the broader community. This achievement underscores CityUHK's commitment to impactful research that bridges materials science, biomedical engineering, and neuroscience.