Xiaofeng Sheng

In addition to its flavor and nutritional value, the origin of kelp has become a crucial factor influencing consumer choices. Nevertheless, research on kelp’s origin traceability by volatile organic compound (VOC) analysis is lacking, and the application of deep learning in this field remains scarce due to its black-box nature. To address this gap, we attempted to identify the origin of kelp by analyzing its VOCs in conjunction with explainable deep learning. In this work, we identified 115 distinct VOCs in kelp samples using gas chromatography coupled with ion mobility spectroscopy (GC-IMS), of which 68 categories were discernible. Consequently, we developed a comprehensible one-dimensional convolutional neural network (1D-CNN) model that incorporated 107 VOCs exhibiting significant regional disparities (p < 0.05). The model successfully discerns the origin of kelp, achieving perfect metrics across accuracy (100%), precision (100%), recall (100%), F1 score (100%), and AUC (1.0). SHapley Additive exPlanations (SHAP) analysis highlighted the impact of features such as 1-Octen-3-ol-M, (+)-limonene, allyl sulfide-D, 1-hydroxy-2-propanone-D, and (E)-2-hexen-1-al-M on the model output. This research provides deeper insights into how critical product features correlate with specific geographic information, which in turn boosts consumer trust and promotes practical utilization in actual settings.

In this work, a rapid method for the simultaneous determination of N and S in seafood was established based on a solid sampling absorption-desorption system coupled with a thermal conductivity detector. This setup mainly includes a solid sampling system, a gas line unit for controlling high-purity oxygen and helium, a combustion and reduction furnace, a purification column system for moisture, halogen, SO₂, and CO₂, and a thermal conductivity detector. After two stages of purging with 20 s of He sweeping (250 mL/min), N₂ residue in the sample-containing chamber can be reduced to 2, H₂O, and CO₂ were absorbed by the adsorption column in turn, the purification process executed the vaporization of the N-containing analyte, and then N₂ was detected by the thermal conductivity cell for the quantification of N. Subsequently, the adsorbed SO₂ was released after heating the SO₂ adsorption column and then transported to the thermal conductivity cell for the detection and quantification of S. After the instrumental optimization, the linear range was 2.0–100 mg and the correlation coefficient (R) was more than 0.999. The limit of detection (LOD) for N was 0.66 μg and the R was less than 4.0%, while the recovery rate ranged from 95.33 to 102.8%. At the same time, the LOD for S was 2.29 μg and the R was less than 6.0%, while the recovery rate ranged from 92.26 to 105.5%. The method was validated using certified reference materials (CRMs) and the measured N and S concentrations agreed with the certified values. The method indicated good accuracy and precision for the simultaneous detection of N and S in seafood samples. The total time of analysis was less than 6 min without the sample preparation process, fulfilling the fast detection of N and S in seafood. The establishment of this method filled the blank space in the area of the simultaneous and rapid determination of N and S in aquatic product solids. Thus, it provided technical support effectively to the requirements of risk assessment and detection in cases where supervision inspection was time-dependent.