Artificial Intelligence-Enhanced Waveguide "Photonic Nose"- Augmented Sensing Platform for VOC Gases in Mid-Infrared.

On-chip nanophotonic waveguide sensor is a promising solution for miniaturization and label-free detection of gas mixtures utilizing the absorption fingerprints in the mid-infrared (MIR) region. However, the quantitative detection and analysis of organic gas mixtures is still challenging and less reported due to the overlapping of the absorption spectrum. Here,an Artificial-Intelligence (AI) assisted waveguide "Photonic nose" is presented as an augmented sensing platform for gas mixture analysis in MIR. With the subwavelength grating cladding supported waveguide design and the help of machine learning algorithms, the MIR absorption spectrum of the binary organic gas mixture is distinguished from arbitrary mixing ratio and decomposed to the single-component spectra for concentration prediction. As a result, the classification of 93.57% for 19 mixing ratios is realized. In addition, the gas mixture spectrum decomposition and concentration prediction show an average root-mean-square error of 2.44 vol%. The work proves the potential for broader sensing and analytical capabilities of the MIR waveguide platform for multiple organic gas components toward MIR on-chip spectroscopy.

Development of Photonic In-Sensor Computing Based on a Mid-Infrared Silicon Waveguide Platform.

Neuromorphic in-sensor computing has provided an energy-efficient solution to smart sensor design and on-chip data processing. In recent years, various free-space-configured optoelectronic chips have been demonstrated for on-chip neuromorphic vision processing. However, on-chip waveguide-based in-sensor computing with different data modalities is still lacking. Here, by integrating a responsivity-tunable graphene photodetector onto the silicon waveguide, an on-chip waveguide-based in-sensor processing unit is realized in the mid-infrared wavelength range. The weighting operation is achieved by dynamically tuning the bias of the photodetector, which could reach 4 bit weighting precision. Three different neural network tasks are performed to demonstrate the capabilities of our device. First, image preprocessing is performed for handwritten digits and fashion product classification as a general task. Next, resistive-type glove sensor signals are reversed and applied to the photodetector as an input for gesture recognition. Finally, spectroscopic data processing for binary gas mixture classification is demonstrated by utilizing the broadband performance of the device from 3.65 to 3.8 µm. By extending the wavelength from near-infrared to mid-infrared, our work shows the capability of a waveguide-integrated tunable graphene photodetector as a viable weighting solution for photonic in-sensor computing. Furthermore, such a solution could be used for large-scale neuromorphic in-sensor computing in photonic integrated circuits at the edge.

Quantitative risk assessments of Salmonella spp. in domestic pork in China.

Pork is one of the most commonly consumed meats, and its safety has always been a concern. Recently, safety incidents caused by chemical or biological contamination such as drug residues, heavy metals, and pathogenic microorganisms in pork have been reported, and the safety of pork is a cause for concern. Salmonella spp. is one of the important foodborne pathogens that threaten human health. Pork is a high-risk vector food for Salmonella spp. infection. The assessment of the safety risk of Salmonella spp. in pork is conducive to the prevention of related foodborne diseases. In this paper, risk assessment models for Salmonella spp. in meat were developed. The quantitative risk assessment model for Salmonella spp. based on the pork supply chain showed that the annual number of cases of salmonellosis due to pork consumption in China is approximately 27 per 10,000 males and 24 per 10,000 females. Sensitivity analysis showed that the main factors affecting the risk of Salmonella spp. in pork were the display temperature, display time, and Salmonella spp. contamination concentration in pork at the sale.

On-line measurement of COD and nitrate in water against stochastic background interference based on ultraviolet-visible spectroscopy and physics-informed multi-task learning.

Traditional ultraviolet-visible spectroscopic quantitative analytical methods face challenges in simultaneous and long-term accurate measurement of chemical oxygen demand (COD) and nitrate due to spectral overlap and the interference from stochastic background caused by turbidity and chromaticity in water. Addressing these limitations, a compact dual optical path spectrum detection sensor is introduced, and a novel ultraviolet-visible spectroscopic quantitative analysis model based on physics-informed multi-task learning (PI-MTL) is designed. Incorporating a physics-informed block, the PI-MTL model integrates pre-existing physical knowledge for enhanced feature extraction specific to each task. A multi-task loss wrapper strategy is also employed, facilitating comprehensive loss evaluation and adaptation to stochastic backgrounds. This novel approach significantly outperforms conventional models in COD and nitrate measurement under stochastic background interference, achieving impressive prediction R2 values of 0.941 for COD and 0.9575 for nitrate, while reducing root mean squared error (RMSE) by 60.89 % for COD and 77.3 % for nitrate in comparison to the conventional chemometric model partial least squares regression (PLSR), and by 30.59 % and 65.96 %, respectively, in comparison to a benchmark convolutional neural network (CNN) model. The promising results emphasize its potential as a spectroscopic instrument designed for online multi-parameter water quality monitoring against stochastic background interference, enabling long-term accurate measurement of COD and nitrate levels.