Flexible deformation sensors inevitably suffer from sensitivity degradation and severe measurement errors during long-term cyclic stretching due to structural fatigue. Traditional material-level optimizations are costly and lack dynamic adaptability. Herein, we propose an artificial intelligence (AI)-driven predict-and-compensate framework for the online calibration of flexible sensors. To overcome training sample scarcity, a generative adversarial network (GAN) performs temporal data augmentation. Subsequently, a hybrid deep learning framework integrating long short-term memory (LSTM) networks and a Sequence Attention mechanism is employed. This architecture accurately captures both local signal fluctuations and multiscale long-term decay trends, enabling precise multi-step prediction and output compensation. Experimental evaluations validate that this strategy significantly suppresses sensor response drift. Under cyclic loading, an initially substantial relative measurement error of 48.63% plummets to 7.16% post-calibration, with typical errors consistently reduced to the ~1% level. Furthermore, when deployed in a smart glove gesture recognition system, this method successfully restores the recognition accuracy from a fatigue-induced low of 75.73% (after 200 stretch cycles) back to 97.70%. This generative and attention-based deep learning paradigm offers robust, real-time error calibration, providing a highly viable solution for extending the long-term reliability and stability of flexible sensor systems.

In this paper, we build four-part cone models to explore the coupling effect of seven cone fiber couplers. Moreover, this is the first study of the coupling effect of four layers of biological couplers in animals and other biological lives. We simulate the four layers cone couplers by using the beam propagation method, and we assume the input beam is located at the outer fiber of the central cone. Our simulation results showed that there are two wavelength regions (short and long wavelength regions) with the strongest coupling, where the most power of input optical powers of the central cones will transfer to the six surrounding cones after transmitting through the four layers of cone couplers. However, within a wavelength region of ±75 nm near to the peak wavelengths, located in the yellow–green wavelength range, the splitting ratios at the output of the outer segment of the central cone are always greater than the sum of the splitting ratios of the six surrounding cones. These cone couplers may play an important role in color preprocessing (e.g., doing opponent color processing partially). The cone fiber coupler effect and light absorption of cones are considered separately in our models. By taking account of both the cone fiber coupling effect and absorption of outer segment of L cone, we find the multiplication of the relative optical power of cone couplers, the spectral sensitivity data of the L cone, and a normalized coefficient that matches with the photopic luminous efficiency of the human eye well. This is the attempt to use both the cone fiber coupling effect and the absorption of L cones to explain the photopic luminous efficiency. The splitting ratios of the central cones are greater than 80% at peak wavelengths located in the yellow-green wavelength range, and this can help to explain why the human eye is more sensitive to green light.

Alcoholic liver disease (ALD) is the major immunometabolic disorder characterized by dysregulated lipid metabolism and persistent activation of hepatic innate immune responses. Vincamine (Vin), isolated from Vinca minor, has been reported to exhibit diverse pharmacological activities. Nevertheless, the immunomodulatory effects and underlying mechanisms of Vin on ALD remain poorly understood. This study aimed to explore the protective effects of Vin against alcohol-induced hepatic injury, with a particular focus on immunometabolic regulation of inflammasome activation. An acute ethanol-fed mouse model was established using male C57BL/6 mice. In vitro experiments were performed using ethanol-stimulated AML12 and HepG2 hepatocytes, as well as LPS/ATP-activated murine peritoneal macrophages. Vin significantly attenuated hepatic steatosis in ethanol-fed mice through modulation of lipid metabolism-associated genes, notably SREBP1 and PPARα. Vin also modulated glycolysis-related pathways through regulation of CDK1, GLUT1, and HIF-1α. In vivo, Vin markedly reduced alcohol-induced immune cell infiltration and suppressed activation of the TLR4-NLRP3 inflammasome pathway. In hepatocytes, Vin alleviated ethanol-induced lipid accumulation, accompanied by downregulation of CDK1 and GLUT1 and reduced NLRP3 activation and IL-1β secretion. In macrophages, Vin inhibited LPS/ATP-induced IL-1β and Caspase-1 expression. Mechanistically, CDK1 deficiency impaired glycolytic activity, as reflected by reduced expression of GLUT1, HIF-1α, and LDHA, and subsequently alleviated lipid accumulation and inflammatory responses under ethanol exposure. Vin exerts protective effects against alcohol-induced hepatic steatosis and inflammation by modulating a CDK1-associated metabolic-inflammatory axis involving glycolysis and NLRP3 inflammasome activation.