Learning-to-augment strategy using noisy and denoised data: Improving generalizability of deep CNN for the detection of COVID-19 in X-ray images.

Chest X-ray images are used in deep convolutional neural networks for the detection of COVID-19, the greatest human challenge of the 21st century. Robustness to noise and improvement of generalization are the major challenges in designing these networks. In this paper, we introduce a strategy for data augmentation using the determination of the type and value of noise density to improve the robustness and generalization of deep CNNs for COVID-19 detection. Firstly, we present a learning-to-augment approach that generates new noisy variants of the original image data with optimized noise density. We apply a Bayesian optimization technique to control and choose the optimal noise type and its parameters. Secondly, we propose a novel data augmentation strategy, based on denoised X-ray images, that uses the distance between denoised and original pixels to generate new data. We develop an autoencoder model to create new data using denoised images corrupted by the Gaussian and impulse noise. A database of chest X-ray images, containing COVID-19 positive, healthy, and non-COVID pneumonia cases, is used to fine-tune the pre-trained networks (AlexNet, ShuffleNet, ResNet18, and GoogleNet). The proposed method performs better results compared to the state-of-the-art learning to augment strategies in terms of sensitivity (0.808), specificity (0.915), and F-Measure (0.737). The source code of the proposed method is available at https://github.com/mohamadmomeny/Learning-to-augment-strategy.

Development of a colloidal gold-based immunochromatographic test strip for screening of microalbuminuria.

A rapid immunochromatography (ICG) assay based on antibody colloidal gold nanoparticles specific to human serum albumin (HSA) was developed, and its applications for primary screening of HSA in the urine were evaluated. A monoclonal antibody (MAb) specific to HSA was produced from cloned hybridoma cells (EMRC1) and used to develop an ICG strip. The nanocolloidal gold, with an average particle diameter of 20 nm, was synthesized and labeled MAb as the detection reagent. An antibody colloidal gold probe was applied on the conjugate pad, and HSA antigen was immobilized to a nitrocellulose membrane as the capture reagent to prepare the ICG strip test. This test required only 10 min to accomplish a semiquantitative detection of albumin. The sensitivity to urinary albumin was found to be approximately 20 µg/mL, and the analytical range was 20-25 µg/mL. The reliability of the testing procedures was examined by carrying out the ICG strip test with 40 urine samples and comparing the results of these tests with those obtained via immunoturbidimetry. The ICG strip was adequately sensitive and accurate for a rapid screening of HSA in the urine.

Colloidal nanogold-based immunochromatographic strip test for the detection of digoxin toxicity.

Digoxin is widely used as a cardiac glycoside drug in the treatment of various heart conditions. Because it is a toxic drug, it should be regularly monitored in the serum of patients under treatment. In this study, colloidal nanogold is synthesized and the preparation of nanogold-labeled monoclonal antibody probe to digoxin is described under optimal conditions. In addition, an immunochromatographic (IC) method for digoxin analysis employing nanogold-labeled probe is developed. With this technique, it requires only 5 min to complete the quantitative detection of digoxin. The detection time is decreased 20-30 times in comparison to radioimmunoassay (RIA). The sensitivity to digoxin was about 2 ng/ml by naked eye, which is within the therapeutic and toxic ranges of digoxin. The results of serum samples obtained by IC strip were in agreement with those obtained by RIA. The IC strip was sufficiently sensitive and accurate to be used for the rapid detection of digoxin in serum samples.