Farmland pest recognition based on Cascade RCNN Combined with Swin-Transformer.

Agricultural pests and diseases pose major losses to agricultural productivity, leading to significant economic losses and food safety risks. However, accurately identifying and controlling these pests is still very challenging due to the scarcity of labeling data for agricultural pests and the wide variety of pest species with different morphologies. To this end, we propose a two-stage target detection method that combines Cascade RCNN and Swin Transformer models. To address the scarcity of labeled data, we employ random cut-and-paste and traditional online enhancement techniques to expand the pest dataset and use Swin Transformer for basic feature extraction. Subsequently, we designed the SCF-FPN module to enhance the basic features to extract richer pest features. Specifically, the SCF component provides a self-attentive mechanism with a flexible sliding window to enable adaptive feature extraction based on different pest features. Meanwhile, the feature pyramid network (FPN) enriches multiple levels of features and enhances the discriminative ability of the whole network. Finally, to further improve our detection results, we incorporated non-maximum suppression (Soft NMS) and Cascade R-CNN's cascade structure into the optimization process to ensure more accurate and reliable prediction results. In a detection task involving 28 pest species, our algorithm achieves 92.5%, 91.8%, and 93.7% precision in terms of accuracy, recall, and mean average precision (mAP), respectively, which is an improvement of 12.1%, 5.4%, and 7.6% compared to the original baseline model. The results demonstrate that our method can accurately identify and localize farmland pests, which can help improve farmland's ecological environment.

A Three-Dimensional Paper-Based Isoelectric Focusing Device for Direct Analysis of Proteins in Physiological Samples.

On-site protein analysis is crucial for disease diagnosis in community and family medicine in which microfluidic paper-based analytical devices (µPADs) have attracted growing attention. However, the practical applications of µPADs in protein analysis for physiological samples with high complexity is still limited. Herein, we developed a three-dimensional (3D) paper-based isoelectric focusing (IEF) platform, which is composed of power supply, reservoirs, and separation channel and made by the origami and stacking method, to simultaneously separate and enrich proteins in both low-salt and high-salt samples. Under the optimized experimental conditions, standard proteins (bovine hemoglobin (BHb) and phycocyanin (Phy)) were separated within 18 min under a 36 V power supply and obtained a 10-fold enrichment using the 3D paper-based IEF platform. Then, the capability of the 3D paper-based IEF platform for direct pretreatment of high-salt samples using a 12 V battery as power supply was measured through separating three standard proteins in saline (0.9% NaCl) with separation resolution (SR) > 1.29. Through further coupling with colorimetric and lateral flow strip measurements, the 3D paper-based IEF platform was applied to directly pretreat and quantitatively analyze microalbuminuria and C-reactive proteins in clinical urine and serum samples with analytical results with relative deviations of <8.4% and < 13.1%, respectively, to the clinical test results. This work proposes a new strategy to minimize the difficulty of directly processing high-salt samples with the traditional IEF system and provides a versatile, miniaturized, and low voltage demand analytical platform for on-site analysis of proteins in physiological samples.