Search

Background An artificial-intelligence (AI) model for predicting the prognosis or mortality of coronavirus disease 2019 (COVID-19) patients will allow efficient allocation of limited medical resources. We developed an early mortality prediction ensemble model for COVID-19 using AI models with initial chest X-ray and electronic health record (EHR) data. Results We used convolutional neural network (CNN) models (Inception-ResNet-V2 and EfficientNet) for chest X-ray analysis and multilayer perceptron (MLP), Extreme Gradient Boosting (XGBoost), and random forest (RF) models for EHR data analysis. The Gradient-weighted Class Activation Mapping and Shapley Additive Explanations (SHAP) methods were used to determine the effects of these features on COVID-19. We developed an ensemble model (Area under the receiver operating characteristic curve of 0.8698) using a soft voting method with weight differences for CNN, XGBoost, MLP, and RF models. To resolve the data imbalance, we conducted F1-score optimization by adjusting the cutoff values to optimize the model performance (F1 score of 0.77). Conclusions Our study is meaningful in that we developed an early mortality prediction model using only the initial chest X-ray and EHR data of COVID-19 patients. Early prediction of the clinical courses of patients is helpful for not only treatment but also bed management. Our results confirmed the performance improvement of the ensemble model achieved by combining AI models. Through the SHAP method, laboratory tests that indicate the factors affecting COVID-19 mortality were discovered, highlighting the importance of these tests in managing COVID-19 patients.

Background Integrins play a critical role in carcinogenesis. Integrin β1 localization is regulated by the guanosine-5′-triphosphate hydrolase Rab25 and integrin β1 levels are elevated in the serum of colon cancer patients; thus, the present study examined the effects of epidermal growth factor (EGF) and Rab25 on integrin β1 localization in colon cancer cells. Methods HCT116 human colon cancer cells were treated with increasing concentrations of EGF, and cell proliferation and protein expression were monitored by MTT and western blot analyses, respectively. Cell fractionation was performed to determine integrin β1 localization in the membrane and cytosol. Integrin β1 extracellular shedding was monitored by enzyme-linked immunosorbent assays (ELISAs) with culture supernatants from stimulated cells. HCT116 cells were transfected with Rab25-specific siRNA to determine the significance of Rab25 in integrin β1 trafficking in the presence of EGF. Results Total integrin β1 expression increased in response to EGF and subsequently decreased at 24 h post-stimulation. A similar decrease was observed in purified membrane fractions, whereas no changes were observed in cytosolic levels. ELISAs using media from stimulated cell cultures demonstrated increased integrin β1 levels corresponding to the decrease observed in membrane fractions, suggesting that EGF induces integrin receptor shedding. EGF stimulation in Rab25-knockdown cells resulted in integrin β1 accumulation in the membrane, suggesting that Rab25 promotes integrin endocytosis. Conclusions Integrin β1 is shed from colon cancer cells in response to EGF stimulation in a Rab25-dependent manner. These results further the present understanding of the role of integrin β1 in colon cancer progression.

ENA Actimineral Resource A (ENA-A) is alkaline water that is composed of refined edible cuttlefish bone and two different species of seaweed, Phymatolithon calcareum and Lithothamnion corallioides. In the present study, ENA-A was investigated as an antioxidant to protect against CCl(4)-induced oxidative stress and hepatotoxicity in rats. Liver injury was induced by either subacute or chronic CCl(4) administration, and the rats had free access to tap water mixed with 0% (control group) or 10% (v/v) ENA-A for 5 or 8 weeks. The results of histological examination and measurement of antioxidant activity showed that the reactive oxygen species production, lipid peroxidation, induction of CYP2E1 were decreased and the antioxidant activity, including glutathione and catalase production, was increased in the ENA-A groups as compared with the control group. On 2-DE gel analysis of the proteomes, 13 differentially expressed proteins were obtained in the ENA-A groups as compared with the control group. Antioxidant proteins, including glutathione S-transferase, kelch-like ECH-associated protein 1, and peroxiredoxin 1, were increased with hepatocyte nuclear factor 3-beta and serum albumin precursor, and kininogen precursor decreased more in the ENA-A groups than compared to the control group. In conclusion, our results suggest that ENA-A does indeed have some protective capabilities against CCl(4)-induced liver injury through its antioxidant function.