Metasurface biosensors have become the core label-free and rapid-detection technology in bioanalysis. Lung cancer and brain cancer are the first leading causes of cancer death among adults and adolescents, respectively, where poor early diagnosis results from expensive detection costs and time consumption. To tackle the above problems, here, we introduce a terahertz-domain metasurface biosensor for cancer diagnosis, relying on a perfectly symmetrical periodic surface structure, which significantly exhibits polarization-insensitivity at 2.05 THz and the high-sensitivity of 504 GHz/RIU (RIU = refractive index unit). According to the frequency shifts and transmittance variations, four cell types are successfully distinguished from each other. The minimum number of cells is required for thousands of cells to display the differences of spectra, which is 1/30 of clinical methods. Furthermore, the results were consistent with pathological results (the gold standard in clinic) by Gaussian curve fitting. The proposed biosensor has really achieved the characterization of cells in normal and cancerous state. This detection strategy dramatically reduced the cost of detection by reuse and time consumption was reduced to 1/20 of the pathology testing. In addition, it is flexible to set samples and easy to realize automatic operation due to the great polarization-insensitivity of the proposed biosensor, which can further reduce labor costs in the future. It is envisioned that the proposed biosensor will present immense potential in the fields of cancer detection, distinguishing different cancers, and identifying primary lesion cancer or metastatic cancer.

Hearing impairment is a common disease affecting a substantial proportion of the global population. Currently, the most effective clinical treatment for patients with sensorineural deafness is to implant an artificial electronic cochlea. However, the improvements to hearing perception are variable and limited among healthy subjects. Moreover, cochlear implants have disadvantages, such as crosstalk derived from the currents that spread into non-target tissue between the electrodes. Here, in this work, we describe terahertz wave modulation, a new and minimally invasive technology that can enhance hearing perception in animals by reversible modulation of currents in cochlear hair cells. Using single-cell electrophysiology, guinea pig audiometry, and molecular dynamics simulations, we show that THM can reversibly increase mechano-electrical transducer currents (~ 50% higher) and voltage-gated K⁺ currents in cochlear hair cells through collective resonance of –C=O groups. In addition, measurement of auditory brainstem response (ABR) in guinea pigs treated with THM indicated a ~ 10 dB increase in hearing sensitivity. This study thus reports a new method of highly spatially selective hearing enhancement without introducing any exogeneous gene, which has potential applications for treatment of hearing disorders as well as several other areas of neuroscience.

Collagen, one of the major components in the mammalian connective tissues, plays an essential role in many vital physiological processes. Many common diseases, such as fibrosis, overuse injuries, and bone fracture, are associated with collagen arrangement defects. However, the underlying mechanism of collagen arrangement defects remains elusive. In this study, we applied infrared scattering-type scanning near-field optical microscopy to study collagen fibrils’ structural properties. Experimentally, we observed two types of collagen fibrils’ arrangement with different periodic characteristics. A crystal sliding model was employed to explain this observation qualitatively. Our results suggest that the collagen dislocation propagates in collagen fibrils, which may shed light on many collagen diseases’ pathogenesis. These findings help to understand the regulation mechanism of hierarchical biological structure.

We establish a preliminary model of neural signal generation and transmission based on our previous research, and use this model to study signal transmission on unmyelinated nerves. In our model, the characteristics of neural signals are studied both on a long-time and a short time scale. On the long-time scale, the model is consistent with the circuit model. On the short time scale, the neural system exhibits a THz and infrared electromagnetic oscillation but the energy envelope curve of the rapidly oscillating signal varies slowly. In addition, the numerical method is used to solve the equations of neural signal generation and transmission, and the effects of the temperature on signal transmission are studied. It is found that overly high and overly low temperatures are not conducive to the transmission of neural signals.