Uncertainty quantification in skin cancer classification using three-way decision-based Bayesian deep learning.

Accurate automated medical image recognition, including classification and segmentation, is one of the most challenging tasks in medical image analysis. Recently, deep learning methods have achieved remarkable success in medical image classification and segmentation, clearly becoming the state-of-the-art methods. However, most of these methods are unable to provide uncertainty quantification (UQ) for their output, often being overconfident, which can lead to disastrous consequences. Bayesian Deep Learning (BDL) methods can be used to quantify uncertainty of traditional deep learning methods, and thus address this issue. We apply three uncertainty quantification methods to deal with uncertainty during skin cancer image classification. They are as follows: Monte Carlo (MC) dropout, Ensemble MC (EMC) dropout and Deep Ensemble (DE). To further resolve the remaining uncertainty after applying the MC, EMC and DE methods, we describe a novel hybrid dynamic BDL model, taking into account uncertainty, based on the Three-Way Decision (TWD) theory. The proposed dynamic model enables us to use different UQ methods and different deep neural networks in distinct classification phases. So, the elements of each phase can be adjusted according to the dataset under consideration. In this study, two best UQ methods (i.e., DE and EMC) are applied in two classification phases (the first and second phases) to analyze two well-known skin cancer datasets, preventing one from making overconfident decisions when it comes to diagnosing the disease. The accuracy and the F1-score of our final solution are, respectively, 88.95% and 89.00% for the first dataset, and 90.96% and 91.00% for the second dataset. Our results suggest that the proposed TWDBDL model can be used effectively at different stages of medical image analysis.

Microwave-assisted extraction of arabinan-rich pectic polysaccharides from melon peels: Optimization, purification, bioactivity, and techno-functionality.

The effects of water to solids ratio (WSR, 10-30 mL/g), power (180-540 W), and irradiation time (IT, 5-15 min) in microwave-assisted extraction (MAE) were optimized to extract polysaccharides from melon peels (PMP). The maximum extraction yield (32.81 %) was obtained under 20.94 mL/g WSR, 414.4 W power, and 12.75 min IT. The main monosaccharide composition of purified PMP with an average molecular weight of 5.71 × 104 kDa were d-galacturonic acid, arabinose, glucose, and galactose. An ascending dose-dependent antiradical and antioxidant behavior for PMP (0-5.0 mg/mL) was found. The initial foaming capacity (38.6-110.3 %) and foaming stability (5.2-65.2 %) were significantly increased as a function of PMP concentration (1.0-5.0 %), while they reduced by increasing the mixing time (p < 0.05). The highest emulsifying activity index (44.1 m2/g) and emulsifying stability (69.3 %) at 5.0 % PMPs were determined. PMP gels with FTIR-identified functional groups can be formulated in new gluten-free functional products.

Garlic (Allium sativum L.): a potential unique therapeutic food rich in organosulfur and flavonoid compounds to fight with COVID-19.

Coronavirus disease 2019 (COVID-19) is the current major health crisis in the world. A successful strategy to combat the COVID-19 pandemic is the improvement of nutritional pattern. Garlic is one of the most efficient natural antibiotics against the wide spectrum of viruses and bacteria. Organosulfur (e.g., allicin and alliin) and flavonoid (e.g., quercetin) compounds are responsible for immunomodulatory effects of this healthy spice. The viral replication process is accelerated with the main structural protease of SARS-CoV-2. The formation of hydrogen bonds between this serine-type protease and garlic bioactives in the active site regions inhibits the COVID-19 outbreak. The daily dietary intake of garlic and its derived-products as an adjuvant therapy may improve side effects and toxicity of the main therapeutic drugs with reducing the used dose.

Microwave-assisted extraction of polysaccharides from the marshmallow roots: Optimization, purification, structure, and bioactivity.

The process optimization and biological characterization of marshmallow root polysaccharides (MRPs) obtained from the microwave-assisted extraction (MAE) were studied. The highest MAE-yield (14.47%) was optimized at 457.32 W and 75 °C for 26 min. The extracted crude polysaccharides were purified using ion-exchange and gel-filtration chromatographies and eluted a single symmetrical narrow peak, showing a homogenous fraction (MRP-P1) with a molecular weight of 4.87 × 104 Da. The surface morphology of polysaccharides and functional groups of MRP-P1 were determined by employing scanning electron microscopy and Fourier-transform infrared spectroscopy, respectively. The major monosaccharide composition of MRPs were the three monomers of rhamnose, galactose, and glucose. The antioxidant, antimicrobial, and antitumor activities were increased in a concentration-dependent manner (1.0-10.0 mg/mL). MRP-P1 exhibited a strong in vitro antiproliferative activity against lung (A549), liver (HepG2), and breast (MCF-7) cancer cells. The anticancer activity of polysaccharides extracted under optimal MAE conditions was highly associated with their antioxidant and antibacterial functions.

Correction to: Future nutrient-dense diets rich in vitamin D: a new insight toward the reduction of adverse impacts of viral infections similar to COVID-19.

[This corrects the article DOI: 10.1186/s41110-020-00122-4.].