The surface hydrophobicity of a microbial cell is known to be one of the important factors in its adhesion to an interface. To date, such property has been altered by either genetic modification or external pH, temperature, and nutrient control. Here we report a new strategy to engineer a microbial cell surface and discover the unique dynamic trapping of hydrophilic cells at an air/water interface via hydrophobicity switching. We demonstrate the surface transformation and hydrophobicity switching of Escherichia coli (E. coli) by metal nanoparticles. By employing real-time dark-field imaging, we directly observe that hydrophobic gold nanoparticle-coated E. coli, unlike its naked counterpart, is irreversibly trapped at the air/water interface because of elevated hydrophobicity. We show that our surface transformation method and resulting dynamic interfacial trapping can be generally extended to Gram-positive bateria, Gram-negative bacteria, and fungi. As the dynamic interfacial trapping allows the preconcentration of microbial cells, high intensity of scattering light, in-plane focusing, and near-field enhancement, we are able to directly quantify E. coli as low as 1.0 × 103 cells/ml by using a smartphone with an image analyzer. We also establish the identification of different microbial cells by the characteristic Raman transitions directly measured from the interfacially trapped cells.

The World Health Organization and Korea National Health Insurance assert that the number of alopecia patients is increasing every year, and approximately 70 percent of adults suffer from scalp problems. Although alopecia is a genetic problem, it is difficult to diagnose at an early stage. Although deep-learning-based approaches have been effective for medical image analyses, it is challenging to generate deep learning models for alopecia detection and analysis because creating an alopecia image dataset is challenging. In this paper, we present an approach for generating a model specialized for alopecia analysis that achieves high accuracy by applying data preprocessing, data augmentation, and an ensemble of deep learning models that have been effective for medical image analyses. We use an alopecia image dataset containing 526 good, 13,156 mild, 3742 moderate, and 825 severe alopecia images. The dataset was further augmented by applying normalization, geometry-based augmentation (rotate, vertical flip, horizontal flip, crop, and affine transformation), and PCA augmentation. We compare the performance of a single deep learning model using ResNet, ResNeXt, DenseNet, XceptionNet, and ensembles of these models. The best result was achieved when DenseNet, XceptionNet, and ResNet were combined to achieve an accuracy of 95.75 and an F1 score of 87.05.

No abstract