CNN-LSTM deep learning based forecasting model for COVID-19 infection cases in Nigeria, South Africa and Botswana.

Background: COVID-19 pandemic has indeed plunged the global community especially African countries into an alarming difficult situation culminating into a great deal amounts of catastrophes such as economic recession, political instability and loss of jobs. The pandemic spreads exponentially and causes loss of lives. Following the outbreak of the omicron new variant of concern, forecasting and identification of the COVID-19 infection cases is very vital for government at various levels. Hence, having knowledge of the spread at a particular point in time, swift actions can be taken by government at various levels with a view to accordingly formulate new policies and modalities towards minimizing the trajectory of the consequences of COVID-19 pandemic to both public health and economic sectors. Methods: Here, a potent combination of Convolutional Neural Network (CNN) learning algorithm along with Long Short Term Memory (LSTM) learning algorithm has been proposed in this work in order to produce a hybrid of a deep learning algorithm Convolutional Neural Network - Long Short Term Memory (CNN-LSTM) for forecasting COVID-19 infection cases particularly in Nigeria, South Africa and Botswana. Forecasting models for COVID-19 infection cases in Nigeria, South Africa and Botswana, were developed for 10 days using deep learning-based approaches namely CNN, LSTM and CNN-LSTM deep learning algorithm respectively. Results: The models were evaluated on the basis of four standard performance evaluation metrics which include accuracy, MSE, MAE and RMSE respectively. However, the CNN-LSTM deep learning-based forecasting model achieved the best accuracy of 98.30%, 97.60%, and 97.74% for Nigeria, South Africa and Botswana respectively; and in the same manner, achieved lesser MSE, MAE and RMSE values compared to models developed with CNN and LSTM respectively. Conclusions: Taken together, the CNN-LSTM deep learning-based forecasting model for COVID-19 infection cases in Nigeria, South Africa and Botswana dramatically surpasses the two other DL based forecasting models (CNN and LSTM) for COVID-19 infection cases in Nigeria, South Africa and Botswana in terms of not only the best accuracy of with 98.30%, 97.60%, and 97.74% but also in terms of lesser MSE, MAE and RMSE.

Predictors of civil and forensic inpatient psychiatric readmissions at a Public Mental Health Hospital.

Inpatient psychiatric readmissions are increasingly prevalent and associated with worse prognostic outcomes and high economic costs, regardless of the medicolegal ramifications that necessitate them. Unlike most general medical readmissions, psychiatric readmissions are commonly warranted for both medical and legal purposes. However, studies focusing on analyzing the predictors of inpatient psychiatric readmission and their relationship to civil versus forensic readmission are limited. The purpose of this study was to examine the predictors of psychiatric readmission among civil and forensic patients admitted to a psychiatric hospital. In this retrospective chart review, we extrapolated data from medical records of 741 patients admitted from 2012 to 2017 with follow up until 2019. Analyses involved chi-square tests for comparing the distribution of demographic and clinical variables between forensic and civil readmission, and Cox regression to determine predictors of time to first readmission. Our results show that race, diagnosis, restraint/seclusion, type of admission, and disposition are significantly associated with an increased risk of psychiatric readmission. This study has important implications for healthcare providers and policy makers in revising mental health policies and improving systems-based practices for the mental health system. Future efforts in improving community psychiatric services and enhancing inpatient therapeutic environment may reduce psychiatric readmissions.

Status of malondialdehyde, catalase and superoxide dismutase levels/activities in schoolchildren with iron deficiency and iron-deficiency anemia of Kashere and its environs in Gombe State, Nigeria.

BACKGROUND: Iron-deficiency anemia (IDA) or iron deficiency (ID) is by far the most common form of disorder affecting the cognitive development, physical growth and school performance of children in developing countries including Nigeria. OBJECTIVES: In the present study, we aimed to examine whether IDA or ID, or both are associated with oxidative stress or otherwise by assessing the perturbations in oxidative stress markers including malondialdehyde (MDA), catalase (CAT) and superoxide dismutase (SOD). METHODS: Here, a total of eighty-one IDA, ID, and healthy control subjects of twenty-seven replicates each, were recruited and investigated. Human serum MDA, CAT and SOD levels were quantitatively analyzed using Enzyme-Linked Immunosorbant Assay. RESULTS: Mean serum MDA levels of IDA (5.10 ± 2.35 mmol/L) and ID (4.05 ± 1.35 mmol/L) groups were found to perturb significantly (p < 0.05), being higher than those of control (3.30 ± 0.95 mmol/L) subjects. Similarly, mean serum MDA levels of IDA (5.10 ± 2.35 mmol/L) group was found to be significantly (p < 0.05) higher when compared with ID (4.05 ± 1.35 mmol/L) subjects. Conversely, mean serum CAT and SOD activities of IDA (8.35 ± 2.21 ng/mL and 340.70 ± 153.65 ng/mL) group were found to differ significantly (p < 0.05), and those of ID (9.40 ± 1.47 ng/mL and 435.00 ± 144.75 ng/mL) subjects were found to perturb slightly (p > 0.05), being lower than those of control (10.40 ± 4.31 ng/mL and 482.12 ± 258.37 ng/mL) subjects. CONCLUSIONS: Taken together, the results of the present study showed that lipid peroxidation was dramatically increased in both IDA and ID subjects in hydroperoxide-superoxide-dependent manner; in contrast, enzymatic antioxidant capacity was drastically decreased in both IDA and ID groups as evidenced by biochemical markers.

Transmission electron microscopy of carbon-coated and iron-doped titania nanoparticles.

We present a study on the properties of iron (Fe)-doped and carbon (C)-coated titania (TiO2) nanoparticles (NPs) which has been compiled by using x-ray diffraction (XRD), transmission electron microscopy (TEM), and x-ray photoelectron spectroscopy (XPS). These TiO2 NPs were prepared by using the flame synthesis method. This method allows the simultaneous C coating and Fe doping of TiO2 NPs. XRD investigations revealed that the phase of the prepared NPs was anatase TiO2. Conventional TEM analysis showed that the average size of the TiO2 NPs was about 65 nm and that the NPs were uniformly coated with the element C. Furthermore, from the x-ray energy dispersive spectrometry analysis, it was found that about 8 at.% Fe was present in the synthesized samples. High-resolution TEM (HRTEM) revealed the graphitized carbon structure of the layer surrounding the prepared TiO2 NPs. HRTEM analysis further revealed that the NPs possessed the crystalline structure of anatase titania. Energy-filtered TEM (EFTEM) analysis showed the C coating and Fe doping of the NPs. The ratio of L3 and L2 peaks for the Ti-L23 and Fe-L23 edges present in the core loss electron energy loss spectroscopy (EELS) revealed a +4 oxidation state for the Ti and a +3 oxidation state for the Fe. These EELS results were further confirmed with XPS analysis. The electronic properties of the samples were investigated by applying Kramers-Kronig analysis to the low-loss EELS spectra acquired from the prepared NPs. The presented results showed that the band gap energy of the TiO2 NPs decreased from an original value of 3.2 eV to about 2.2 eV, which is quite close to the ideal band gap energy of 1.65 eV for photocatalysis semiconductors. The observed decrease in band gap energy of the TiO2 NPs was attributed to the presence of Fe atoms at the lattice sites of the anatase TiO2 lattice. In short, C-coated and Fe-doped TiO2 NPs were synthesized with a rather cost-effective and comparatively easily scalable method. The presented analysis enables us to predict the excellent efficiency of these NPs for solar-cell and photo-catalysis applications.