Investigation on electrochemical performance of striped, ß12 and χ3 Borophene as anode materials for lithium-ion batteries.

By developing next-generation lithium-ion batteries (LIBS), demand for exploring novel anode materials with exclusive electrochemical features and ultra-high capacity is increasing. In the current research, first-principles theory, and density functional theory (DFT) calculations were conducted to extensively investigate and compare the capability of three different borophene nanolayers, including striped, ß12, and χ3 borophene, as a novel candidate for anode electrode in LIBs. We first predicted the most preferential Li atom adsorption sites on the three borophene structures. The predicted average formation energies for striped, ß12, and χ3 borophene were obtained 3.123, 3.184, and 3.216 eV, respectively. The positive value of formation energy exhibits the sufficient stability of the structures. Moreover, the negative adsorption energy proved that Li atom insertion on all borophene monolayers is thermodynamically favorable. In order to simulate the lithiation process, we gradually increased the concentration of Li atoms. We found that the fully lithiated striped, ß12 and χ3 borophenes could provide ultra-high specific capacities of 1700, 1983, and 1859 mAh/g, respectively. Structural analysis proved that the surface area expansion rate of the striped borophene in a fully lithiated state was 1%, which was lower than those of ß12 and χ3 borophene with 3.33% and 2.63%, respectively. The analyses of electronic properties confirmed that borophenes were inherently metallic and superior ion conductive agents, even after fully lithiated state. Ion diffusion was studied using Nudged elastic band method and the value of diffusion energy barrier ranged from 0.03 to 0.36 eV which was lower than other promising 2D anode materials. Furthermore, open-circuit voltage results demonstrated that the electronic potential of modeled borophenes was low enough to be in the acceptable range of under 1.2V. All these reports exhibited that borophene nanolayers with excellent specific capacity and superior conductivity were desired candidates for anode materials of next generation LIBs.

E-CatBoost: An efficient machine learning framework for predicting ICU mortality using the eICU Collaborative Research Database.

Improving the Intensive Care Unit (ICU) management network and building cost-effective and well-managed healthcare systems are high priorities for healthcare units. Creating accurate and explainable mortality prediction models helps identify the most critical risk factors in the patients' survival/death status and early detect the most in-need patients. This study proposes a highly accurate and efficient machine learning model for predicting ICU mortality status upon discharge using the information available during the first 24 hours of admission. The most important features in mortality prediction are identified, and the effects of changing each feature on the prediction are studied. We used supervised machine learning models and illness severity scoring systems to benchmark the mortality prediction. We also implemented a combination of SHAP, LIME, partial dependence, and individual conditional expectation plots to explain the predictions made by the best-performing model (CatBoost). We proposed E-CatBoost, an optimized and efficient patient mortality prediction model, which can accurately predict the patients' discharge status using only ten input features. We used eICU-CRD v2.0 to train and validate the models; the dataset contains information on over 200,000 ICU admissions. The patients were divided into twelve disease groups, and models were fitted and tuned for each group. The models' predictive performance was evaluated using the area under a receiver operating curve (AUROC). The AUROC scores were 0.86 [std:0.02] to 0.92 [std:0.02] for CatBoost and 0.83 [std:0.02] to 0.91 [std:0.03] for E-CatBoost models across the defined disease groups; if measured over the entire patient population, their AUROC scores were 7 to 18 and 2 to 12 percent higher than the baseline models, respectively. Based on SHAP explanations, we found age, heart rate, respiratory rate, blood urine nitrogen, and creatinine level as the most critical cross-disease features in mortality predictions.