Devices based on rapid, paper-based, isothermal nucleic acid amplification techniques have recently emerged with the potential to fill a growing need for highly sensitive point-of-care diagnostics throughout the world. As this field develops, such devices will require optimized materials that promote amplification and sample preparation. Herein, we systematically investigated isothermal nucleic acid amplification in materials currently used in rapid diagnostics (cellulose paper, glass fiber, and nitrocellulose) and two additional porous membranes with upstream sample preparation capabilities (polyethersulfone and polycarbonate). We compared amplification efficiency from four separate DNA and RNA targets (Bordetella pertussis, Chlamydia trachomatis, Neisseria gonorrhoeae, and Influenza A H1N1) within these materials using two different isothermal amplification schemes, helicase dependent amplification (tHDA) and loop-mediated isothermal amplification (LAMP), and traditional PCR. We found that the current paper-based diagnostic membranes inhibited nucleic acid amplification when compared to membrane-free controls; however, polyethersulfone allowed for efficient amplification in both LAMP and tHDA reactions. Further, observing the performance of traditional PCR amplification within these membranes was not predicative of their effects on in situ LAMP and tHDA. Polyethersulfone is a new material for paper-based nucleic acid amplification, yet provides an optimal support for rapid molecular diagnostics for point-of-care applications.

Abstract

Aiming at the problems of small metal surface defect samples in industrial production and the difficulty of data annotation in supervised segmentation algorithms, a background reconstruction method based on Cycle generative adversarial networks is proposed to realize metal surface defect segmentation in combination with the traditional threshold segmentation algorithm. Firstly, the corresponding defect-free template is reconstructed from the defect image using Cycle generative adversarial networks, and the defect image and the reconstructed template are subjected to the differential subjected to eliminate the influence of the background texture of the defect sample. Finally, the segmentation process is performed using the adaptive thresholding segmentation method. In order to adapt to the small sample training as well as to improve the performance of background reconstruction of the generative network, the U-Net network structure is used as the generator, and the attention mechanism is also introduced. Meanwhile, the L1 loss and Multi-scale SSIM loss are combined to design the cycle consistency loss for training. The experimental results show that the method in this paper can accomplish good defect segmentation results using a small number of defective samples.

No abstract