image: (a) Left: encoding of blink information, the schematic indicates that the brain generates specific commands and stimulates the eyes to blink accordingly when a specific situation is encountered. Centre: EMI lens. Right: decoding the blink information for machine interface, healthcare, and AR/VR. (b) I, The exploded structure diagram of the blink-recognizable EMI lens, II, digital photographs of the EMI lens in different orientations. (c) The internal circuit of the EMI lens and the reading circuit of the frequency signal. (d) The encoding and decoding process of blink information. The capacitance of the sensor in designed EMI lens changes when the eyes switch between different states (eyes open, squinting, and eyes closed), and then covert to frequency, which is recorded by the external coil and transmitted to the electrical circles for signal process, finally decoding for the drone control application.
Credit: ©Science China Press
EMI has emerged as a promising paradigm for human-computer interaction, yet its development has been hindered by several technical challenges including limited signal accuracy, poor wearability, and visual interference. To address these issues, research team led by Prof. Guozhen Shen (Beijing Institute of Technology) and Prof. Zhiyong Fan (Hong Kong University of Science and Technology) has developed an innovative flexible electronic solution - a smart contact lens integrated with an LC resonant circuit that achieves both high sensitivity and excellent biocompatibility for wireless EMI applications.
Unlike brain-computer interfaces (BCI) based on electroencephalography which needs complex algorithms and electrical circuits, EMI accomplishes the command based on consciousness information generated from the brain via simple ocular movements, the accuracy of which is much higher than that of BCI. Ocular movements primarily include blinks and eye rotation. Existing devices for monitoring ocular movements mainly rely on charge coupled devices (CCD) cameras or metal coil-embedded contact lenses, but the former requires complex external hardware, while the latter affects wearing comfort and field of vision due to rigid components. Compared to eye rotation, blinking offers significant advantages: its visibility, stability, and natural pressure characteristics (the eyelid exerts approximately 30 mmHg of pressure on the cornea during blinking) make it easier to capture with highly sensitive sensors. Additionally, parameters such as blink count, duration, and left/right eye can be encoded to generate diverse commands. Therefore, blink-based EMI systems demonstrate substantial application potential.
The research team designed an EMI system using a multilayer-structured flexible smart contact lens (EMI lens) as the core component. The EMI lens employs flexible materials as the substrate, integrating Ti3C2Tx MXene electrode layers, a honeycomb-structured microporous dielectric layer, and an induction coil to form a complete LC resonant circuit. Pressure variations alter the spacing of the microstructured dielectric layer, thereby changing the capacitance value. This change is converted into measurable frequency signals (detectable by external vector network analyzers) through the LC resonant circuit, enabling wireless pressure monitoring.
Without compromising vision or wearing comfort, the EMI lens can sensitively detect corneal deformation caused by intraocular pressure (IOP) changes and eyelid pressure induced by blinking. The system features dual functional modes: within the normal IOP range (10-21 mmHg), signals are converted into real-time monitoring data; when specific pressure (~30 mmHg) is detected, algorithms translate blink signals into control commands.
In wearability tests, subjects showed no significant physiological rejection or discomfort, confirming the feasibility of practical EMI lens applications. Additionally, the human eye performs 10-20 unconscious blinks per minute, which constitute interference signals in EMI. The EMI lens-based system achieves precise recognition by analyzing blink duration and pressure amplitude, effectively distinguishing conscious from unconscious blinks, with good accuracy in practical tests.
The research team developed a blink-based control command encoding/decoding mechanism that maps different blinking behaviors to flight commands, experimentally validating the feasibility of controlling multidimensional drone movements through blinking. In vivo rabbit tests further confirmed the system's reliability, with normal physiological conditions observed post-experiment. These results fully demonstrate the practical value of the EMI lens-based system in medical monitoring and human-machine control.
Journal
National Science Review