Predicted Value Research Articles

How decadal predictions entered the climate services arena: an example from the agriculture sector

Predicting the variations in climate for the coming 1–10 years is of great interest for decision makers, as this time horizon coincides with the strategic planning of stakeholders from climate-vulnerable sectors such as agriculture. This study attempts to illustrate the potential value of decadal predictions in the development of climate services by establishing interactions and collaboration with stakeholders concerned with food production and security. Building on our experience from interacting with users and the increased understanding of their needs gathered over the years through our participation in various European activities and initiatives, we developed a decadal forecast product that provides tailored and user-friendly information about multi-year dry conditions for the coming five years over global wheat harvesting regions. This study revealed that the coproduction approach, where the interaction between the user and climate service provider is established at an early stage of forecast product development, is a fundamental step to successfully provide useful and ultimately actionable information to the interested stakeholders. The study also provides insights that shed light on the reasons for the delayed entry of decadal predictions in the climate services discourse and practice, obtained from surveying climate scientists and discussing with decadal prediction experts. Finally, it shows the key challenges that this new source of climate information still faces.

Machine learning model to predict mental health crises from electronic health records

The timely identification of patients who are at risk of a mental health crisis can lead to improved outcomes and to the mitigation of burdens and costs. However, the high prevalence of mental health problems means that the manual review of complex patient records to make proactive care decisions is not feasible in practice. Therefore, we developed a machine learning model that uses electronic health records to continuously monitor patients for risk of a mental health crisis over a period of 28 days. The model achieves an area under the receiver operating characteristic curve of 0.797 and an area under the precision-recall curve of 0.159, predicting crises with a sensitivity of 58% at a specificity of 85%. A follow-up 6-month prospective study evaluated our algorithm’s use in clinical practice and observed predictions to be clinically valuable in terms of either managing caseloads or mitigating the risk of crisis in 64% of cases. To our knowledge, this study is the first to continuously predict the risk of a wide range of mental health crises and to explore the added value of such predictions in clinical practice.

Trainable segmentation for transmission electron microscope images of inorganic nanoparticles.

We present a trainable segmentation method implemented within the python package ParticleSpy. The method takes user labelled pixels, which are used to train a classifier and segment images of inorganic nanoparticles from transmission electron microscope images. This implementation is based on the trainable Waikato Environment for Knowledge Analysis (WEKA) segmentation, but is written in python, allowing a large degree of flexibility and meaning it can be easily expanded using other python packages. We find that trainable segmentation offers better accuracy than global or local thresholding methods and requires as few as 100 user-labelled pixels to produce an accurate segmentation. Trainable segmentation presents a balance of accuracy and training time between global/local thresholding and neural networks, when used on transmission electron microscope images of nanoparticles. We also quantitatively investigate the effectiveness of the components of trainable segmentation, its filter kernels and classifiers, in order to demonstrate the use cases for the different filter kernels in ParticleSpy and the most accurate classifiers for different data types. A set of filter kernels is identified that are effective in distinguishing particles from background but that retain dissimilar features. In terms of classifiers, we find that different classifiers perform optimally for different image contrast; specifically, a random forest classifier performs best for high-contrast ADF images, but that QDA and Gaussian Naïve Bayes classifiers perform better for low-contrast TEM images.

Network analysis of psychometric life history indicators

Two of the controversies besetting the field of life history theory in psychology (LHT-P) are (1) whether life history strategy (LHS) is most fruitfully conceptualized as a latent reflective factor, or as a formative, descriptive construct and (2) whether the instruments most commonly used to measure psychometric LHS adequately cover its components. Psychological network analysis, as an alternative to latent factor modeling, affords an opportunity to map the relationships among narrower constructs hypothesized to comprise psychometric LHS. We recruited 1064 U.S. undergraduates, who completed instruments comprising 23 network nodes, including the K-SF-42 scales, Kruger's (2017) mating effort and parental effort scales, and measures of childhood environmental harshness, current stability of resource access, sociosexuality, future orientation, and the dark triad. We tested pre-registered hypotheses pertaining to node centrality and node clustering (communities), and found mixed support for our predictions. Observed values of node predictability suggested that these 23 indicators comprise a self-determined network. The mating effort and parental effort scales were among the most central nodes, along with developmental SES, current resource stability, Machiavellianism and psychopathy. Contrary to prediction, most of the K-SF-42 subscales clustered into a single community of nodes. Our results add to recent literature casting doubt on the usefulness of widely used psychometric LH instruments. We expect that future research will reveal considerable heterogeneity in the sources of associations between the variables that have been characterized LHS indicators.

Thermodynamic optimization of the constitutive binaries of the LiCl/RbCl/CaCl2-NaCl-UCl3-PuCl3 systems

This study is aiming at building a set of self-consistent thermodynamic database of the constitutive binaries of the LiCl/RbCl/CaCl2-NaCl-UCl3-PuCl3 systems, and evaluating whether LiCl, RbCl or CaCl2 is an appropriate additive of NaCl-UCl3-PuCl3 matrix fuel of Molten Salt Reactor in terms of CALculation of PHAse Diagrams (CALPHAD) method. Substitutional solution model was adopted to describe the liquid phase and solid solution phase, while compound energy formalism was used to depict the intermediate phases. The model parameters were optimized based on the experimental data and theoretical predicted values. The thermodynamic parameters for describing the constitutive binaries of LiCl/RbCl/CaCl2-NaCl-UCl3-PuCl3 were ultimately obtained. Phase diagram information such as liquidus surface projections, isothermal sections and vertical sections of the LiCl/RbCl/CaCl2-NaCl-UCl3-PuCl3 system were predicted from the models with the optimized parameters. The effect of LiCl, RbCl and CaCl2 on the isothermal sections and melting temperature of NaCl-UCl3-PuCl3 matrix fuel was also investigated. The results show that RbCl is suitable as the candidate additive component of NaCl-UCl3-PuCl3 matrix fuel from the thermodynamic viewpoint, while LiCl and CaCl2 not. This work not only enriches the thermodynamic databases of NaCl-UCl3-PuCl3 matrix fuel, but also provides a direct guidance for the design and preparation of nuclear fuel in the Molten Salt Reactor.

Identifying Acute Cardiac Hazard in Early Drug Discovery Using a Calcium Transient High-Throughput Assay in Human-Induced Pluripotent Stem Cell-Derived Cardiomyocytes.

Introduction: Early identification of cardiac risk is essential for reducing late-stage attrition in drug development. We adapted the previously published cardiac hazard risk-scoring system using a calcium transient assay in human stem cell-derived CMs for the identification of cardiac risks recorded from the new hiPSC-CM line and investigated its predictivity and translational value based on the screening of a large number of reference and proprietary compounds. Methods: Evaluation of 55 reference drugs provided the translation of various pharmacological effects into a single hazard label (no, low, high, or very high hazard) using a Ca2+-sensitive fluorescent dye assay recorded by -by FDSS/µCell Functional Drug Screening System (Hamamatsu on hiPSC-CM line (FCDI iCell Cardiomyocytes2). Results: Application of the adapted hazard scoring system in the Ca2+ transient assay, using a second hiPS-CM line, provided comparable scoring results and predictivity of hazard, to the previously published scoring approach, with different pharmacological drug classes, as well as screening new chemical entities (NCE’s) using a single hazard label from four different scoring levels (no, low, high, or very high hazard). The scoring system results also showed minimal variability across three different lots of hiPSC-CMs, indicating good reproducibility of the cell line. The predictivity values (sensitivity and specificity) for drug-induced acute cardiac risk for QT-interval prolongation and Torsade de pointes (TdPs) were >95% and statistical modeling confirmed the prediction of proarrhythmic risk. The outcomes of the NCEs also showed consistency with findings in other well-established in vitro and in vivo cardiac risk assays. Conclusion: Evaluation of a large list of reference compounds and internal NCEs has confirmed the applicability of the adaptations made to the previously published novel scoring system for the hiPSC-CMs. The validation also established the predictivity for drug-induced cardiac risks with good translation to other established preclinical in vitro and in vivo assays, confirming the application of this novel scoring system in different stem cell-CM lines for early cardiac hazard identification.

Heating load prediction of residential district using hybrid model based on CNN

Optimal control of heat network in residential district requires accurate short-term heat load prediction value of each building. However, the current research on residential heat load prediction mainly relies on historical data related to heat load and does not consider the influence of building ontology parameters on different building types, resulting in low prediction accuracy. The objective of this study is to obtain accurate short-term heat load prediction values of different buildings in residential district. Based on this, the building ontology parameters were introduced, and the Lasso method was adopted to comprehensively analyze and select the influencing factors of heat load. A hybrid short-term heat load prediction model using adaptive T-distributed Satin Bowerbird (tSBO) algorithm to optimize convolutional neural network (CNN) was proposed. The actual operation data of established heating system in 10 residential districts were collected to test the performance of the model. The results show that building ontology parameters have a significant impact on the prediction accuracy of future heat load. After introducing the building ontology parameters, the mean absolute percentage error (MAPE) and root mean square error (RMSE) of the predicted values of the hybrid model were reduced by about 29.61% and 22.00%, respectively. Compared with other prediction models, the proposed hybrid model achieves an average reduction of 18.08% and 16.26% in MAPE and RMSE.

Positive predictive value of Birads IV Lesions on mammogram in detection of breast carcinoma.

To determine the positive predictive value of Breast Imaging Reporting and Data System category IV lesions, and to evaluate interobserver agreement among radiologists in interpreting the subcategories. The cross-sectional study was conducted at the Dow Institute of Radiology, Dow University of Health Sciences, Karachi, from September 2018 to February 2019, and comprised adult females aged 30-60 years diagnosed with Breast Imaging Reporting and Data System IV. Already diagnosed breast carcinoma patients were excluded. The positive predictive value of Breast Imaging Reporting and Data System category IV was calculated using histopathology findings as the gold standard. Data was analysed using SPSS 22. There were 191 patients with a mean age of 49.01±11.79 years. Positive family history of breast cancer was found in 33(17.3%) subjects. Category IV-A was noted in 53(28.0%) subjects, IV-B in 45(23.6%) and IV-C in 93(48.7%). The overall positive predictive value of category IV was 151(79.1%), IV-A 27(50.9%), IV-B 35(77.7%) and IV-C 89(95.6%). Malignant disease was significantly associated with architectural distortion and suspicious microcalcifications (p<0.001). A good agreement was observed in the categorisation of Breast Imaging Reporting and Data System category IV between observers 1 and 2 (kappa 0.85, 95% confidence interval: 0.70-0.99), between observers 1 and 3 (kappa 0.77, 95% confidence interval: 0.61-0.94), and between observers 2 and 3 (kappa 0.63, 95% confidence interval: 0.43-0.83). Breast Imaging Reporting and Data System BI-RADS IV was found to have a high positive predictive value in the diagnosis of breast cancer. Besides the interobserver agreement in terms of identifying BI-RADS IV subcategories was good.

Comprehending Cardiac Dysfunction by Oxidative Stress: Untargeted Metabolomics of In Vitro Samples.

Cardiovascular diseases (CVDs) are noncommunicable diseases known for their complex etiology and high mortality rate. Oxidative stress (OS), a condition in which the release of free radical exceeds endogenous antioxidant capacity, is pivotal in CVC, such as myocardial infarction, ischemia/reperfusion, and heart failure. Due to the lack of information about the implications of OS on cardiovascular conditions, several methodologies have been applied to investigate the causes and consequences, and to find new ways of diagnosis and treatment as well. In the present study, cardiac dysfunction was evaluated by analyzing cells’ alterations with untargeted metabolomics, after simulation of an oxidative stress condition using hydrogen peroxide (H2O2) in H9c2 myocytes. Optimizations of H2O2 concentration, cell exposure, and cell recovery times were performed through MTT assays. Intracellular metabolites were analyzed right after the oxidative stress (oxidative stress group) and after 48 h of cell recovery (recovery group) by ultra-high-performance liquid chromatography coupled to mass spectrometry (UHPLC-MS) in positive and negative ESI ionization mode. Significant alterations were found in pathways such as “alanine, aspartate and glutamate metabolism”, “glycolysis”, and “glutathione metabolism”, mostly with increased metabolites (upregulated). Furthermore, our results indicated that the LC-MS method is effective for studying metabolism in cardiomyocytes and generated excellent fit (R2Y > 0.987) and predictability (Q2 > 0.84) values.

Organization of Research Activities as a Factor in Increasing the Efficiency of Training International Students in Host Russian Universities

This article summarizes the practices of organizing research activities that can improve the training of international students in Russian universities. The authors analyzed and identified the correlation between the internal costs of research and development (R&amp;D), the equipment in university research laboratories, and the number of foreigners studying at Russian universities. The authors used the methods of analysis and synthesis and a systematic approach to explore the experience of Russian universities in the training of international students. To identify the impact of internal R&amp;D costs and the equipment of university research laboratories on the number of foreigners studying in Russian universities, the authors applied correlation and regression analysis, which included building a regression equation, calculating the correlation coefficient, the t-test, the coefficient of elasticity, and the coefficient of determination. This research paper revealed a strong correlation between the number of international students in the Russian Federation on the internal R&amp;D costs and the cost of fixed assets and the equipment of Russian universities, which was proven by the calculated correlation coefficients, elasticity coefficients, and determination coefficients. The authors concluded that universities could influence the number of international students. This article proposes some methods for organizing the research activities of international students that increase their academic mobility and form the most relevant scientific and professional competencies. Higher educational institutions of any major can implement these recommendations for managing the research activities of international students. The novelty of this study lies in the fact that the authors performed the correlation and regression analysis and revealed the dependence of the number of international students in the Russian Federation on internal R&amp;D costs, the cost of fixed assets, and the equipment of Russian universities. The authors illustrated the analysis results with the trend predictive values of factorial features and the value of the effective feature estimated according to the regression equations built. Using the calculated MAPE and Forecast Accuracy indicators for these predicted values, the authors concluded that the level of factorial features and the effective feature were predicted with high accuracy. Thus, host Russian universities can increase the number of international students through effective organization of research activities.

A Novel Non-Invasive Device For Screening And Optimized Management To Improve Heart Failure Outcomes In Patients With Diabetes Mellitus

<h3>Background</h3> Evidence shows diabetes mellitus (DM) is major independent risk factor for several cardiovascular disorders including heart failure (HF). Large, randomized clinical trials for screening and revascularization of stable macrovascular disease in diabetics, have failed to demonstrate a significant reduction in cardiac events and HF episodes. Studies have demonstrated that early detection of left ventricular dysfunction and prevention of microvascular complications through glycemic control in diabetes patients is a critical mechanism for reducing the incidence and severity of left ventricular dysfunction and HF. There is a need for a non-invasive, affordable, accurate, absolute and actionable method to facilitate optimized management of diabetes patients across the continuity of care. HEMOTAG, is a small portable device that uses micro-sensors to capture cardiac vibrations and electrocardiogram, transduced via thoracic electrodes. It is a viable option to measure cardiac time intervals (CTIs), surrogate markers for measurements of left ventricular dysfunction. <h3>Methods</h3> In 98 diabetes patients, the predictability value of HEMOTAG-derived CTIs- systolic time ratio index (STRi) = aortic valve opening/ (aortic valve opening to aortic valve closing) normalized by heart rate, aortic valve closing (AVC) and mitral valve opening (MVO) were assessed as valid markers for left ventricular dysfunction measured simultaneously at the time of transthoracic echocardiogram; to determine reduced ejection fraction (EF) <50%, <35%; elevated E/e'>14; and elevated left atrial volume index (LAVI) > 34 ml/m². Linear correlation analysis was performed. Sensitivity, specificity, positive, and negative predictive values were calculated. <h3>Results</h3> Mean age was 70.3 ± 11.6 years, mean EF was 59.0 ± 12.7%, mean E/e' was 17.9 ± 8.5, mean LAVI was 35.4 ± 16.1 ml/m², 30.2% were women. The following was noted with a paired dichotomous interpretation; for STRi and EF <35%: sensitivity 100%, specificity 88%, AUC 0.94; for STRi and EF <50%: sensitivity 94%, specificity 87%, AUC 0.93; for MVO and E/e' >14: sensitivity 92%, specificity 45%, AUC 0.63; for AVC and LAVI>34 ml/m²: sensitivity 92%, specificity 53%, AUC 0.65. <h3>Conclusions</h3> HEMOTAG detected systolic and diastolic dysfunction with high sensitivity and high specificity. This preliminary study has demonstrated the feasibility of estimating left ventricular dysfunction using a portable, non-invasive device that can facilitate screening and optimized management to improve HF outcomes in patients with DM. Further studies are required to validate HEMOTAG as a point-of-care non-invasive assessment for patients with DM. HEMOTAG® development and the clinical study is supported by NHLBI SBIR grant # R44HL149561.

Differences in the Spirometry Parameters Between Indigenous and Non-Indigenous Patients with COPD: A Matched Control Study.

BackgroundComparison of spirometry parameters between Indigenous and non-Indigenous patients with underlying chronic obstructive pulmonary disease (COPD) has been sparsely reported in the past. In this study, differences in the lung function parameters (LFPs), in particular spirometry values for forced vital capacity (FVC), forced expiratory volume in one second (FEV1) and FEV1/FVC ratio between Indigenous and non-Indigenous patients with COPD were assessed.MethodsIn this retrospective study, Indigenous and non-Indigenous patients with a diagnosis of COPD between 2012–2020 according to spirometry criteria (ie; post-bronchodilator (BD) FEV1/FVC < 0.7) were included. A further analysis was undertaken to compare the differences in the spirometry parameters, including lower limit of normal (LLN) values matching for age, sex, height and smoking status between these two diverse ethnic populations.ResultsA total of 240/742 (32%) Indigenous and 873/4579 (19%) non-Indigenous patients were identified to fit the criteria for COPD. Indigenous patients were significantly younger (mean difference 9.9 years), with a greater proportion of females (50% vs 33%), underweight (20% vs 8%) and current smokers (47% vs 32%). Prior to matching, Indigenous patients’ post-BD percent predicted values for FVC, FEV1, and FEV1/FVC ratio were 17, 17%, and −2 points lower (Hedges G measure of effect size large (0.91), large (0.87), and small (0.25), respectively). Among the matched cohort (111 Indigenous and non-Indigenous), Indigenous patients LFPs remained significantly lower, with a mean difference of 16%, 16%, and −4, respectively (Hedges G large (0.94), large (0.92) and small (0.41), respectively). The differences persisted despite no significant differences in LLN values for these parameters.ConclusionIndigenous Australian patients with COPD display a significantly different demographic and clinical profile than non-Indigenous patients. LFPs were significantly lower, which may or may not equate to greater severity of disease in the absence of normative predictive lung function reference values specific to this population.

Non-invasive Assessment Of Elevated Left Ventricular End Diastolic Pressure Using A Small Portable Device: Design For Tele-Monitoring

Background Left ventricular end-diastolic pressure (LVEDP) provides crucial information on the LV operating compliance and is a key marker for congestive heart failure and pulmonary hypertension. The diagnosis and monitoring of the left-sided filling pressure often requiring invasive procedures. HEMOTAG, is a small portable device that uses micro-sensors to capture cardiac vibrations and electrocardiogram (ECG), transduced via thoracic electrodes. It is a viable option to measure cardiac time intervals (CTIs), surrogate markers of hemodynamics and intracardiac pressure. In this study, diastolic-to-systolic time ratio (DSR) and systolic time ratio (STR) along with the patient's age, height, and weight were assessed as valid markers for elevated LVEDP, compared against left heart catheterization LVEDP measurement. Methods In 44 female and 49 male patients, mean LVEDP was measured during cardiac catheterization. Simultaneously, Hemotag generated CTIs: systolic time (ST): mitral valve closure (MVC) to aortic valve closure (AVC), diastolic time (DT): AVC to MVC, STR: Aortic valve opening/(Aortic valve opening to Aortic valve closing), and DSR: DT/ST. The predictability value of non-invasively HEMOTAG-derived CTI ratios - diastolic time normalize by heart rate (DT/HR), DSR, and STR - plus biological information of age, height, and weight were assessed to determine elevated LVEDP (> 15 mmHg). Logistic regression analysis between the markers and LVEDP was performed. Sensitivity, specificity, correlation, and p-values were calculated. Results Mean age for the female (male) patients was 69.41 ± 11.84 (63.12 ± 10.80) and mean LVEDP was 17.20 ± 6.17 (16.57 ± 5.42). Using logistic regression with DSR, STR, age, height, and weight as features, sensitivity of 88.00%, specificity of 94.74%, AUC of 0.94, r = 0.74, p-value = 4.31e-06, were obtained for female patients; using DT/HR, STR, age, height, and weight, a sensitivity of 89.29%, specificity of 71.43%, AUC of 0.79, r = 0.40, and p-value = 0.0419 were obtained in the male patients. Conclusions HEMOTAG detected elevated LVEDP with high sensitivity and high specificity for both female and male patients. This preliminary study has demonstrated the feasibility of estimating LVEDP using a portable, non-invasive device that empowers patients to transmit such data from home instantaneously to providers. Further studies are required to validate HEMOTAG as a point-of-care non-invasive assessment for heart failure and pulmonary hypertension patients. HEMOTAG® development and the clinical study is supported by NIMHD SBIR grant # R44MD009556. Left ventricular end-diastolic pressure (LVEDP) provides crucial information on the LV operating compliance and is a key marker for congestive heart failure and pulmonary hypertension. The diagnosis and monitoring of the left-sided filling pressure often requiring invasive procedures. HEMOTAG, is a small portable device that uses micro-sensors to capture cardiac vibrations and electrocardiogram (ECG), transduced via thoracic electrodes. It is a viable option to measure cardiac time intervals (CTIs), surrogate markers of hemodynamics and intracardiac pressure. In this study, diastolic-to-systolic time ratio (DSR) and systolic time ratio (STR) along with the patient's age, height, and weight were assessed as valid markers for elevated LVEDP, compared against left heart catheterization LVEDP measurement. In 44 female and 49 male patients, mean LVEDP was measured during cardiac catheterization. Simultaneously, Hemotag generated CTIs: systolic time (ST): mitral valve closure (MVC) to aortic valve closure (AVC), diastolic time (DT): AVC to MVC, STR: Aortic valve opening/(Aortic valve opening to Aortic valve closing), and DSR: DT/ST. The predictability value of non-invasively HEMOTAG-derived CTI ratios - diastolic time normalize by heart rate (DT/HR), DSR, and STR - plus biological information of age, height, and weight were assessed to determine elevated LVEDP (> 15 mmHg). Logistic regression analysis between the markers and LVEDP was performed. Sensitivity, specificity, correlation, and p-values were calculated. Mean age for the female (male) patients was 69.41 ± 11.84 (63.12 ± 10.80) and mean LVEDP was 17.20 ± 6.17 (16.57 ± 5.42). Using logistic regression with DSR, STR, age, height, and weight as features, sensitivity of 88.00%, specificity of 94.74%, AUC of 0.94, r = 0.74, p-value = 4.31e-06, were obtained for female patients; using DT/HR, STR, age, height, and weight, a sensitivity of 89.29%, specificity of 71.43%, AUC of 0.79, r = 0.40, and p-value = 0.0419 were obtained in the male patients. HEMOTAG detected elevated LVEDP with high sensitivity and high specificity for both female and male patients. This preliminary study has demonstrated the feasibility of estimating LVEDP using a portable, non-invasive device that empowers patients to transmit such data from home instantaneously to providers. Further studies are required to validate HEMOTAG as a point-of-care non-invasive assessment for heart failure and pulmonary hypertension patients. HEMOTAG® development and the clinical study is supported by NIMHD SBIR grant # R44MD009556.

Credit default swap prediction based on generative adversarial networks

PurposeFinancial price forecast issues are always a concern of investors. However, the financial applications based on machine learning methods mainly focus on stock market predictions. Few studies have explored credit risk predictions. Understanding credit risk trends can help investors avoid market risks. The purpose of this study is to investigate the prediction model that can effectively predict credit default swaps (CDS).Design/methodology/approachA novel generative adversarial network (GAN) for CDS prediction is proposed. The authors take three features into account that are highly relevant to the future trends of CDS: historical CDS price, news and financial leverage. The main goal of this model is to improve the existing GAN-based regression model by adding finance and news feature extraction approaches. The proposed model adopts an attentional long short-term memory network and convolution network to process historical CDS data and news information, respectively. In addition to enhancing the effectiveness of the GAN model, the authors also design a data sampling strategy to alleviate the overfitting issue.FindingsThe authors conduct an experiment with a real dataset and evaluate the performance of the proposed model. The components and selected features of the model are evaluated for their ability to improve the prediction performance. The experimental results show that the proposed model performs better than other machine learning algorithms and traditional regression GAN.Originality/valueThere are very few studies on prediction models for CDS. With the proposed novel approach, the authors can improve the performance of CDS predictions. The proposed work can thereby increase the commercial value of CDS predictions to support trading decisions.

A Real‐Time Heart Rate Detection Algorithm Based on Peripheral Electrical Bioimpedance

Impedance cardiography (ICG) and peripheral bioimpedance may be useful for monitoring several cardiovascular parameters such as cardiac output, stroke volume, blood pressure and heart rate. It is a noninvasive and low‐cost method and could be useful for diagnosis and treatment in different clinical situations. This study involved 100 volunteers and their age ranging from 18 to 26 years. The acquired bioimpedance signal was pre‐processed and used to estimate the heart rate. Thus, heart rates of each subject were calculated and then compared with actual measurement from FUZZY model MP3 digital blood pressure monitor. The results showed high precision with a 1.38 mean error a high sensitivity of 99.81% and a positive predictivity value of 99.63%, respectively. Compared with six other methods selected from the literature review and which are initially dedicated to ECG signals, the proposed method has been found to be more efficient and more precise with an error in determining the percentile HR = 1.78%. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Selection of optimal models for predicting growth stress in Artemisia desertorum by comparison of linear regression and multiple neural networks: Take the construction of a green mine in the Bayan Obo mine as an example

In recent years, more and more countries are focusing on the control of mining sites and the surrounding ecological environment, and the new environmental concept of green mines has been proposed. By investigating the ecological background of a mine site, pollution and ecological imbalances in the mine can be predicted, managed or transformed. This study investigated the effects of rare earth elements on plant growth in the Baotou Bayan Obo Rare Earth Mine and evaluated soil contamination and subsequent remediation through the measured plant height. Using linear regression, BP(Back Propagation) neural networks, GA-BP(Genetic Algorithm- Back Propagation) neural networks, ELM(Extreme Learning Machine) and GA-ELM(Genetic Algorithm- Extreme Learning Machine) model prediction instruments, the different rare earth solution concentrations were set as input values and the heights of Artemisia desertorum, which as the model plant, were set as output values in the prediction. The results showed that the linear regression predicted the standard error of single La(III), Ce(III) solution and compound La(III) + Ce(III) solution for Artemisia desertorum growth stress was on the high side, 7.02%− 8.92%; the efficiency range of each group of models under BP neural network, GA-BP neural network and ELM neural network were 1.15%− 2.53%, 0.85%− 1.28%, 1.76%− 3.53%; while the efficiency range under GA-ELM neural network was 0.59%− 0.68%, with average error values and predicted values close to the true values. Among them, the MAPE of GA-ELM neural network are significantly lower than other models, and the error decreases with increasing concentration of the compound solution. So GA-ELM neural network can be used as an efficient, fast and reasonable optimal model for predicting the growth stress of Artemisia desertorum in Bayan Obo mining area. The experimental results can provide a theoretical basis for assessing the risk of soil rare earth contamination in the area, evaluating the expectation of later remediation, and provide a degree of new ideas for the construction of green mines.

STROKE RISK PREDICTION MODEL FOR AFRICAN POPULATIONS: AN APPROACH FOR EARLY IDENTIFICATION OF POPULATIONS AT RISK

Background & Aim Early identification and treatment of populations at risk of stroke in Africa have been delimited by the data to design a risk prediction model tailor-made for indigenous Africans. This study developed and tested an Afrocentric risk scoring model to predict stroke occurrence among West Africans. Method 7066 case-control pairs (identified in the SIREN study) were randomly partitioned into testing and validation datasets (85:15), and conditional logistic regression models were developed from seventeen traditional factors of stroke in the training dataset. Significant factors were assigned constant, statistically weighted (based on regression coefficient), and receiver operating characteristics curves were constructed (from the testing dataset) to estimate cut-off points for discriminate stroke cases from healthy controls (to build an additive risk scoring model for stroke). Sensitivity, specificity, positive and negative predicted values were estimated to assess the strength of the stroke risk prediction model at P < 0.05. Results Cohen's kappa for validity was maximal at a total risk score of 56% using statistical weighting approaches for risk quantification in both datasets after identifying 15 traditional vascular factors for stroke occurrence. The stroke risk prediction model had a strong predictive accuracy (76%, 95%CI: 74%–79%), sensitivity (80.3%), specificity (63.0%), positive predictive value (68.5%) and negative predictive value (76.2%) in the validation dataset. Conclusion The Afrocentric stroke risk models had good accuracy and might be helpful for stroke prediction as a viable approach in de-escalating the rising burden of stroke across multi-ethnic indigenous African populations. Early identification and treatment of populations at risk of stroke in Africa have been delimited by the data to design a risk prediction model tailor-made for indigenous Africans. This study developed and tested an Afrocentric risk scoring model to predict stroke occurrence among West Africans. 7066 case-control pairs (identified in the SIREN study) were randomly partitioned into testing and validation datasets (85:15), and conditional logistic regression models were developed from seventeen traditional factors of stroke in the training dataset. Significant factors were assigned constant, statistically weighted (based on regression coefficient), and receiver operating characteristics curves were constructed (from the testing dataset) to estimate cut-off points for discriminate stroke cases from healthy controls (to build an additive risk scoring model for stroke). Sensitivity, specificity, positive and negative predicted values were estimated to assess the strength of the stroke risk prediction model at P < 0.05. Cohen's kappa for validity was maximal at a total risk score of 56% using statistical weighting approaches for risk quantification in both datasets after identifying 15 traditional vascular factors for stroke occurrence. The stroke risk prediction model had a strong predictive accuracy (76%, 95%CI: 74%–79%), sensitivity (80.3%), specificity (63.0%), positive predictive value (68.5%) and negative predictive value (76.2%) in the validation dataset. The Afrocentric stroke risk models had good accuracy and might be helpful for stroke prediction as a viable approach in de-escalating the rising burden of stroke across multi-ethnic indigenous African populations.

Diabetic Ketoacidosis With Acute Pancreatitis in Patients With Type 2 Diabetes in the Emergency Department: A Retrospective Study.

ObjectiveThis study aims to explore the incidence and clinical features of acute pancreatitis (AP) in patients with type 2 diabetes diabetic ketoacidosis (DKA) in the emergency department and discuss the predictive value of some pathological indicators for AP in DKA.MethodsInpatient medical data of DKA patients hospitalized to our hospital's emergency department between January 2017 and January 2021 were evaluated retrospectively. These DKA patients were split into two groups based on whether they had AP or not. We examined the two groups' epidemiologic features, baseline laboratory results, and clinical outcomes. The Bedside Index for Sequential Organ Failure Assessment (SOFA), Acute Physiology and Chronic Health Evaluation II (APACHE II), and Logistic Organ Failure System (LODS) scores were computed and compared across groups.ResultsThe prevalence of AP in DKA patients was 15.53%. The difference in Abdominal pain between the two groups of patients was statistically significant (p < 0.001), and there was no statistical difference in age, gender, and BMI. The DKA and AP group LOS (P < 0.001), ICU admission rate (P = 0.046), anion gap (P < 0.001), red blood cell (P = 0.002), hemoglobin (P < 0.001), hematocrit (P = 0.002), serum triglyceride (P < 0.001), serum cholesterol (P < 0.001), serum amylase (P = 0.004), random glucose (P = 0.028), plasma fibrinogen (P < 0.001), glycosylated hemoglobin [HbA1c (%); P = 0.008] higher than the DKA group, pH (P < 0.001), carbon dioxide combining power (CO2CP; P < 0.001), ionized calcium (Ca2+; P = 0.022), ionized sodium (Na+; P = 0.001), and correction Na (P = 0.034) lower than the DKA group. Multivariate analysis showed that low pH (P < 0.05), hypertriglyceridemia (P = 0.001), and hypercholesterolemia (P = 0.01) were risk factors for DKA combined with AP. ROC curve analysis showed that the three cut-off value: serum triglycerides of 10.52 mmol/L, serum cholesterol of 9.03 mmol/L, and pH of 7.214. Serum triglyceride has the largest area under the curve (0.93). Under this cut-off value, the sensitivity (80%) and specificity of serum triglyceride, the degree (93.7%) is the highest, while the positive predictive value (62.0%) and negative predictive value (94.7%) of serum cholesterol are the highest.ConclusionsA severe episode of DKA with significant acidosis and hyperlipidemia is more likely to be linked with AP. The frequently used critical illness score is ineffective in determining the severity of the condition. When the serum triglyceride cut-off value is 10.52mmol/L, it has a higher predicted value for AP in DKA.

The Relationship of Partial Pressure of Carbon Dioxide (PaCO2) with Disease Severity Indicators Such as BODE and GOLD in Hospitalized COPD Patients.

Purpose This study aimed to investigate the relationship of partial pressure of carbon dioxide (PaCO2) with BODE and GOLD in stable COPD subjects and to explore the predictive value of PaCO2 for severe COPD (BODE index score ≥5 or GOLD index score ≥3). Patients and Methods. In total, 80 participants with COPD and free from other conditions affecting PaCO2 were recruited. Arterial blood gases, BODE, GOLD, SGRQ, lung function, and other data were collected. The BODE index was calculated, and patients were divided into two groups according to the BODE index and PaCO2 median, respectively. We used Pearson's correlation test and the receiver operating characteristic curves to evaluate the utility of PaCO2. Besides, the univariate and multivariate logistic regression analyses were conducted to verify whether PaCO2 was an independent factor associated with BODE grades. Results COPD subjects with BODE ≥5 and GOLD ≥3 had significantly higher levels of PaCO2 (p = 0.004, p = 0.001, respectively). In the high PaCO2 group, patients underwent poorer outcomes than the low PaCO2 group. PaCO2 was negatively correlated with forced expiratory volume in 1 second in percent of the predicted value (FEV1%) (r = −0.612, p < 0.001). The performance of PaCO2 levels in predicting BODE ≥5 and GOLD ≥3 was 0.748 and 0.755, respectively. The logistic regression analyses proved that PaCO2 was associated with BODE ≥5 in COPD patients (odds ratio = 1.160, 95% CI: 1.025–1.313, p = 0.019). Conclusions A higher level of PaCO2 was associated with a higher index for BODE or GOLD in COPD and had the predictive value for severe COPD.

Decline in Respiratory Functions in Hospitalized SARS-CoV-2 Infected Cancer Patients Following Cytotoxic Chemotherapy-An Additional Risk for Post-chemotherapy Complications.

BackgroundPatients recovering from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection demonstrate impaired lung function and those requiring chemotherapy after recovering from SARS-CoV-2 infection have yet to be explored. In this study, we sought to investigate the possible pulmonary functional changes during and after administering chemotherapy in patients with prior SARS-CoV-2 infection.MethodsIn this study, a total of 37 SARS-CoV-2 infected patients with cancer who were discharged from hospital and received subsequent cytotoxic chemotherapy were enrolled and prospectively followed-up. The following parameters were prospectively measured before (P1), after first chemotherapy cycle (P2), and 10 weeks after the end of chemotherapy (P3), to assess their impact on respiratory complications in terms of diffusion capacity of the lungs for carbon monoxide (DLCO), forced expiratory volume in 1-s (FEV1), forced vital capacity (FVC), 6-min walking distance (6MWD) test and levels of key inflammatory markers.ResultsAll patients completed at least 2 cycles of chemotherapy without showing overt respiratory complications. Six patients (16%) complained about dyspnea during chemotherapy or at follow-up period. DLCO was significantly impaired during follow-up period [from P1 78 to P3 60% of predicted values; interquartile range (IQR) 55–89] and in 32 of 37 (86% of patients) from P1 to P2 (65% of predictive value; IQR 58–70; p < 0.001). Several patients experienced post-chemotherapy respiratory complications. As expected, all patients from control groups showed persistent improved pulmonary functions.ConclusionThe risk of pulmonary impairments due to cytotoxic chemotherapy in prior SARS-CoV-2 infected patients is linked to the loss of DLCO. Accordingly, we recommend that for patients with cancer requiring chemotherapy after recovering from prior SARS-CoV-2 infection, pulmonary tests to be performed routinely before and during chemotherapy treatment to monitor the pulmonary performance.