image: (a) Optical photographs of NeuroCam. (b) Validation of NeuroCam for recording brain signals in an epileptic rabbit model. (c) High-spatial-resolution recording results obtained with NeuroCam.
Credit: Xing Sheng @ Tsinghua University, China
Accurately and efficiently capturing large-scale neural signals is a core objective of next-generation brain-machine interface (BMI) technology, as well as a critical breakthrough direction in neuroscience and neuroengineering. Electrocorticography (ECoG) devices, by conforming to the cerebral cortex, enable the monitoring of coordinated neural population activity across extensive brain regions, and have been successfully applied in key scenarios such as speech synthesis, motor decoding, and epileptogenic zone localization.
However, traditional passive ECoG electrodes are typically millimeter-scale in size, far larger than cortical columns (only tens to hundreds of micrometers). This makes it difficult to accurately capture neural signals from subdivided brain regions. Meanwhile, high-resolution recording relies on high-density electrode arrays, but passive electrodes require independent wires for each channel, limiting the array scale and the efficiency of signal readout. In recent years, emerging active electrode arrays have overcome the fan-out bottleneck using time-division multiplexing technology. Yet their complex, non-standard manufacturing processes hinder the low-cost and large-scale production, restricting further practical applications.
Nowadays, finding an integrated approach to overcome the multiple challenges of flexible ECoG devices—including limited channel counts, low density, complex wiring, and difficulty in large-scale manufacturing—and meeting the demand for high-spatial-resolution recording of large-scale brain signals, have become a critical challenge.
This study develops a flexible, multiplexed, high-density ECoG device named NeuroCam. Featuring up to 4096 channels and compatibility with large-scale production, NeuroCam achieves an integrated breakthrough in overcoming the key challenges of existing devices, such as limited channels, low density, complex wiring, and poor scalability. It opens up a new pathway for high-spatial-resolution recording of large-scale brain signals, and provides strong support for the next phase of leapfrog development in BMI technology.
Built on metal oxide thin-film transistors, the array integrates 4096 recording channels on a single flexible substrate, achieving a channel density of 44 sites/mm2. Through its multiplexed design, signal readout requires only 128 input/output (I/O) wires, effectively resolving the wiring challenges of high-channel devices. Additionally, the device applies industrial-grade manufacturing processes, which not only ensure the feasibility of large-scale production but also guarantee the consistent performance across all channels.
In in vivo experiments on epileptic rabbit models, NeuroCam (4096 channels) accurately captured the spatiotemporal changes of epileptic potentials in real time and synchronously, while clearly localizing epileptogenic zones, demonstrating exceptional spatial resolution and large-area brain region recording capabilities. In vitro experiments further verified the device’s key performance attributes, including biocompatibility, electrical stability, and bending resistance.
NeuroCam exhibits significant advantages in key metrics including channel count and density. It provides a new tool for decoding complex neural activities and advances the development of high-performance BMI technology. Meanwhile, the breakthrough opens up new opportunities for neuroscience research and neuroengineering applications related to the diagnosis and treatment of neurological disorders such as epilepsy.