Search Results (26)

The significant impact of road traffic accidents on public health requires clear and effective policies to combat them. However, public action can only be truly effective when supported by robust monitoring tools. This project aims to evaluate the effectiveness of a set of machine learning algorithms in predicting road accidents in Switzerland, utilizing open-access Confederation drive crash databases combined with environmental and socio-economic factors. Three different algorithms are tested: Logistic Regression Model (LRM), Random Forest with Ranger (RF), and Artificial Neural Network (ANN) with Keras. Among the predictive factors, road types are shown to be of high importance in all models. Regarding model performance, all the applied algorithms show a high level of accuracy, with all models achieving over 90%. The Random Forest algorithm, optimised using the Ranger application, exhibited the best performance, particularly in terms of specificity (0.88 compared to 0.34 and 0.40 for LRM and Keras, respectively) and negative predictive value (0.96 compared to 0.65 for LRM and 0.68 for Keras). These results suggest that this approach could support public policy for traffic management, if data collection and sharing activities are constantly carried out. Full article

Accumulating evidence suggests that the predisposition to metabolic diseases is established in utero through epigenomic modifications. However, it remains unclear whether childhood obesity results from preexisting epigenomic alterations or whether obesity itself induces changes in the epigenome. This study aimed to identify DNA methylation marks in the placenta associated with obesity-related outcomes in children at age 6 and to assess these marks in blood samples at age 6 and whether they correlate with obesity-related outcomes at that time. Using an epigenome-wide DNA methylation microarray on 24 placental samples, we identified differentially methylated CpGs (DMCs) associated with offspring BMI-SDS at 6 years. Individual DMCs were validated in 147 additional placental and leukocyte samples from children at 6 years of age. The methylation and/or gene expression of IRS1 in both placenta and offspring leukocytes were significantly associated with various metabolic risk parameters at age 6 (all p ≤ 0.05). Logistic regression models (LRM) and machine learning (ML) models indicated that IRS1 methylation in the placenta could strongly predict offspring obesity. Our results suggest that IRS1 may serve as a potential biomarker for the prediction of obesity and metabolic risk in children. Full article

As adiposity increases in youth, so does the prevalence of cardiometabolic risk factors (CMRFs). The etiology of adiposity-based chronic disease and CMRFs includes ethnoracial disparities that are rarely considered in current treatment approaches. Precision interventions require further characterization of these disparities among high-risk youth. The objective of this study was to characterize differences in CMRF among African American (AA) and Hispanic (H) adolescents with varying levels of adiposity. A cross-sectional analysis of 2284 adolescents aged 12–17 was conducted using 3-year clinical data from Lifedoc Health. CMRF prevalence were compared using χ², with logistic regression models (LRM) applied to explore the relationships between exposures (age, sex, ethnoracial group, adiposity) and CMRF outcomes. Prevalence of CMRF rose with increasing adiposity, which was the strongest determinant of risk overall. However, individual risk profiles differed between the two groups, with H having higher prevalence of metabolic syndrome (MetS), higher triglycerides and liver enzymes, and low high-density lipoprotein cholesterol (HDL-c). Meanwhile, AA had higher prevalence of elevated blood pressure (BP) in the overweight category, prediabetes in overweight to severe obesity, and type 2 diabetes in obesity. LRM showed 3.0-fold greater chance of impaired glucose metabolism in AA than H, who were 1.7, 5.9, and 8.3 times more likely to have low HDL-c, high liver enzymes, and high triglycerides, respectively. Overweight/obesity prevalence was very high among AA and H adolescents. Excess adiposity was associated with an increased prevalence of CMRF, with individual risk factors differing between groups as adiposity increased. Research within routine clinical settings is required to better characterize these discrepancies and ameliorate their adverse impact on health in the transition to adulthood. Full article

Cultural property includes immovable assets that are part of a nation’s cultural heritage and reflect the cultural identity of a people. Hence, information about armed conflict’s impact on historical buildings’ structures and heritage sites is extremely important. The study aims to demonstrate the application of Earth observation (EO) synthetic aperture radar (SAR) technology, and in particular Sentinel-1 SAR coherence time-series analysis, to monitor spatial and temporal changes related to the recent Russian–Ukrainian war in the urban areas of Mariupol and Kharkiv, Ukraine. The study considers key events during the siege of Mariupol and the battle of Kharkiv from February to May 2022. Built-up areas and cultural property were identified using freely available OpenStreetMap (OSM) data. Semi-automated coherent change-detection technique (CCD) that utilize difference analysis of pre- and co-conflict coherences were capable of highlighting areas of major impact on the urban structures. The study applied a logistic regression model (LRM) for the discrimination of damaged and undamaged buildings based on an estimated likelihood of damage occurrence. A good agreement was observed with the reference data provided by the United Nations Satellite Centre (UNOSAT) in terms of the overall extent of damage. Damage maps enable the localization of buildings and cultural assets in areas with a high probability of damage and can serve as the basis for a high-resolution follow-up investigation. The study reveals the benefits of Sentinel-1 SAR CCD in the sense of unsupervised delineation of areas affected by armed conflict. However, limitations arise in the detection of local and single-building damage compared to regions with large-scale destruction. The proposed semi-automated multi-temporal Sentinel-1 data analysis using CCD methodology shows its applicability for the timely investigation of damage to buildings and cultural heritage, which can support the response to crises. Full article

Protecting biodiversity and keeping the Earth’s temperature stable are both very important jobs performed by tropical forests. In the last few decades, remote sensing has given us new tools and ways to track changes in land cover. To understand what causes changes in forest cover, it is important to look at the things that affect those changes. However, there is not enough research that uses a logistic regression model (LRM) and compares the results with machine learning (ML) techniques to investigate the specific factors that cause forest cover change in remote mountainous areas like Thailand’s Mae Hong Son and Chiang Mai Provinces. Following a comparison of an LRM, a random forest, and an SVM, this study of the causes of changes in forest cover in Mae Hong Son found six important factors: soil series, rock types, slope, the NDVI, the NDWI, and the distances to city areas. Compared to the LRM, both the RF and SVM machine learning algorithms had higher values for the kappa coefficient, sensitivity, specificity, accuracy, positive and negative predictions, and sensitivity, especially the RF. Following what was found in Mae Hong Son, when the important factors were examined in Chiang Mai, the RF came out on top. It is believed that these results can be used in more situations to help make plans for restoring ecosystems and to promote long-lasting methods of managing land use. Full article

As a cigarette butt falls onto the forest surface fuel, it first smolders the fuel, then ignites into flames, and spreads as forest fire under certain conditions. In this study, the needles under a typical stand of P. massoniana were used as the research object. Needle beds with different moisture content and packing ratios were constructed indoors. Cigarette butt-ignition experiments were conducted under different wind velocities, and 30 experiment cycles were conducted under different conditions. There was a total of 5 (packing ratio) × 4 (moisture content) × 6 (wind velocity) = 120 sets of conditions, and a total of 3600 ignition experiments were conducted. The results showed that (1) the total ignition probability of the cigarette butts was 2.36%, which only occurred when the fuelbed moisture content was <10% and the wind velocity was >1 m/s. The ignition time of cigarette butts ranged from 2.73 to 7.25 min. (2) The fuelbed moisture content and wind velocity significantly influenced the ignition probability and time. With an increase in moisture content, the ignition probability of cigarette butts decreased, while the time required for ignition showed an increasing trend. Wind velocity had a dual effect on ignition. The ignition effect was optimal at a wind velocity of 4 m/s. With an increase in wind velocity, the ignition probability first increased and then decreased, and the ignition time first decreased and then increased. (3) The packing ratio had no significant effect on the ignition probability; however, the ignition time significantly decreased as the packing ratio increased. (4) The logistic regression method (LRM), general linear method (GLM), and nonlinear regression method (NLM) were used to establish a prediction model of ignition probability. The prediction effect of GLM was the worst, followed by LRM, and the NLM had the best prediction effect. The GLM was selected to establish the ignition time model, and the error was also within the allowance range. This study elucidated the underlying mechanism of factors affecting cigarette butt-based fuel ignition. In addition, the established prediction model provides a reference for human-caused forest fires and is highly significant for forest fire prevention. Full article

Purpose: We aimed to assess the efficacy of machine learning and radiomics analysis using magnetic resonance imaging (MRI) with a hepatospecific contrast agent, in a pre-surgical setting, to predict tumor budding in liver metastases. Methods: Patients with MRI in a pre-surgical setting were retrospectively enrolled. Manual segmentation was made by means 3D Slicer image computing, and 851 radiomics features were extracted as median values using the PyRadiomics Python package. Balancing was performed and inter- and intraclass correlation coefficients were calculated to assess the between observer and within observer reproducibility of all radiomics extracted features. A Wilcoxon–Mann–Whitney nonparametric test and receiver operating characteristics (ROC) analysis were carried out. Balancing and feature selection procedures were performed. Linear and non-logistic regression models (LRM and NLRM) and different machine learning-based classifiers including decision tree (DT), k-nearest neighbor (KNN) and support vector machine (SVM) were considered. Results: The internal training set included 49 patients and 119 liver metastases. The validation cohort consisted of a total of 28 single lesion patients. The best single predictor to classify tumor budding was original_glcm_Idn obtained in the T1-W VIBE sequence arterial phase with an accuracy of 84%; wavelet_LLH_firstorder_10Percentile was obtained in the T1-W VIBE sequence portal phase with an accuracy of 92%; wavelet_HHL_glcm_MaximumProbability was obtained in the T1-W VIBE sequence hepatobiliary excretion phase with an accuracy of 88%; and wavelet_LLH_glcm_Imc1 was obtained in T2-W SPACE sequences with an accuracy of 88%. Considering the linear regression analysis, a statistically significant increase in accuracy to 96% was obtained using a linear weighted combination of 13 radiomic features extracted from the T1-W VIBE sequence arterial phase. Moreover, the best classifier was a KNN trained with the 13 radiomic features extracted from the arterial phase of the T1-W VIBE sequence, obtaining an accuracy of 95% and an AUC of 0.96. The validation set reached an accuracy of 94%, a sensitivity of 86% and a specificity of 95%. Conclusions: Machine learning and radiomics analysis are promising tools in predicting tumor budding. Considering the linear regression analysis, there was a statistically significant increase in accuracy to 96% using a weighted linear combination of 13 radiomics features extracted from the arterial phase compared to a single radiomics feature. Full article

Background: As has been shown previously, patients with atrial fibrillation (AF) who have left atrial thrombus (LAT) also have elevated plasma concentrations of fibrinogen. In this study, we tried to determine if this is the consequence of a genetic trait and whether elevated concentrations of fibrinogen could be used to predict LAT in patients with AF. Methods: We recruited 181 consecutive patients scheduled for pulmonary vein isolation (PVI) or direct current cardioversion. The primary endpoint was the presence of LAT on transesophageal echocardiography (TOE). We recorded routine clinical and biochemical data as well as the polymorphism type of the fibrinogen gene for the β chain. To control potentially interfering variables, we performed propensity score matching (PSM). Multivariable and univariable logistic regression models (LRM) were computed using the CHA2DS2-Vasc score, the fibrinogen concentration and creatinine clearance as estimated by the Cockcroft–Gault equation. Results: 60 of 181 patients had LAT as detected by TOE. As expected, patients with LAT had significantly higher concentrations of fibrinogen (3.9 vs. 3.6 g/L); p = 0.01 in the unadjusted analysis. After performing PSM, there were no statistically significant differences between the groups, except for creatinine clearance (79.9 vs. 96.8 mL/min); p = 0.01. There were also no differences regarding the −455 G/A βfibrinogen polymorphism distribution between the two groups. After constructing the LRM, we found no performance enhancement for the CHA2DS2-Vasc score by adding the fibrinogen concentration or creatinine clearance alone, but when all three variables were put together, there was a significant improvement in LAT prediction (AUC 0.64 vs. 0.72), p = 0.026. Conclusion: Our study found no evidence of elevated levels of circulating fibrinogen in patients with LAT or a connection between those levels and the A/A and A positive polymorphism. When used together with renal function markers such as creatinine clearance, plasma fibrinogen concentrations can provide additional power to the CHA2DS2-Vasc score for predicting LAT. Full article

Background: Dual-energy CT has been reported to be useful for differentiating thymic epithelial tumors. The purpose is to evaluate thymic epithelial tumors by using three-dimensional (3D) iodine density histogram texture analysis on dual-energy CT and to investigate the association of extracellular volume fraction (ECV) with the fibrosis of thymic carcinoma. Methods: 42 patients with low-risk thymoma (n = 20), high-risk thymoma (n = 16), and thymic carcinoma (n = 6) were scanned by dual-energy CT. 3D iodine density histogram texture analysis was performed for each nodule on iodine density mapping: Seven texture features (max, min, median, average, standard deviation [SD], skewness, and kurtosis) were obtained. The iodine effect (average on DECT180s—average on unenhanced DECT) and ECV on DECT180s were measured. Tissue fibrosis was subjectively rated by one pathologist on a three-point grade. These quantitative data obtained by examining associations with thymic carcinoma and high-risk thymoma were analyzed with univariate and multivariate logistic regression models (LRMs). The area under the curve (AUC) was calculated by the receiver operating characteristic curves. p values < 0.05 were significant. Results: The multivariate LRM showed that ECV > 21.47% in DECT180s could predict thymic carcinoma (odds ratio [OR], 11.4; 95% confidence interval [CI], 1.18–109; p = 0.035). Diagnostic performance was as follows: Sensitivity, 83.3%; specificity, 69.4%; AUC, 0.76. In high-risk thymoma vs. low-risk thymoma, the multivariate LRM showed that the iodine effect ≤1.31 mg/cc could predict high-risk thymoma (OR, 7; 95% CI, 1.02–39.1; p = 0.027). Diagnostic performance was as follows: Sensitivity, 87.5%; specificity, 50%; AUC, 0.69. Tissue fibrosis significantly correlated with thymic carcinoma (p = 0.026). Conclusions: ECV on DECT180s related to fibrosis may predict thymic carcinoma from thymic epithelial tumors, and the iodine effect on DECT180s may predict high-risk thymoma from thymoma. Full article

In the management of sepsis, providing adequate empiric antimicrobial therapy is one of the most important pillars of sepsis management. Therefore, it is important to evaluate the adequacy of empiric antimicrobial therapy (EAMT) in sepsis patients admitted to intensive care units (ICU) and to identify the determinants of inadequate EAMT. The aim of this study was to evaluate the adequacy of empiric antimicrobial therapy in patients admitted to the ICU with sepsis or septic shock, and the determinants of inadequate EAMT. The data of patients admitted to the ICU units due to sepsis or septic shock in two tertiary healthcare facilities in Al-Madinah Al-Munawwarah were retrospectively reviewed. The current study used logistic regression analysis and artificial neural network (ANN) analysis to identify determinants of inadequate empiric antimicrobial therapy, and evaluated the performance of these two approaches in predicting the inadequacy of EAMT. The findings of this study showed that fifty-three per cent of patients received inadequate EAMT. Determinants for inadequate EAMT were APACHE II score, multidrug-resistance organism (MDRO) infections, surgical history (lower limb amputation), and comorbidity (coronary artery disease). ANN performed as well as or better than logistic regression in predicating inadequate EAMT, as the receiver operating characteristic area under the curve (ROC-AUC) of the ANN model was higher when compared with the logistic regression model (LRM): 0.895 vs. 0.854. In addition, the ANN model performed better than LRM in predicting inadequate EAMT in terms of classification accuracy. In addition, ANN analysis revealed that the most important determinants of EAMT adequacy were the APACHE II score and MDRO. In conclusion, more than half of the patients received inadequate EAMT. Determinants of inadequate EAMT were APACHE II score, MDRO infections, comorbidity, and surgical history. This provides valuable inputs to improve the prescription of empiric antimicrobials in Saudi Arabia going forward. In addition, our study demonstrated the potential utility of applying artificial neural network analysis in the prediction of outcomes in healthcare research. Full article

This study utilised multi-year data from 5354 incidents to predict pedestrian–road traffic accidents (PTAs) based on twelve socioeconomic and built-environment factors. The research employed the logistic regression model (LRM) and the fuzzy-analytical hierarchy process (FAHP) techniques to evaluate and assign weights to each factor. The susceptibility map for PTAs is generated using the “Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS)”. Subsequently, the probability of accidents in 2020 was predicted using real multi-year accident data and the Markov chain (MC) and cellular automata Markov chain (CA-MC) models, with the prediction accuracy assessed using the Kappa index. Building upon promising results, the model was extrapolated to forecast the probability of accidents in 2023. The findings of the LRM demonstrated the significance of the selected variables as predictors of accident likelihood. The prediction approaches identified areas prone to high-risk accidents. Additionally, the Kappa for no information (KNO) statistical value was calculated for both the MC and CA-MC models, which yielded values of 0.94 and 0.88, respectively, signifying a high level of accuracy. The proposed methodology is generalizable, and the identification of high-risk locations can aid urban planners in devising appropriate preventive measures. Full article

Background and Objectives: To describe the predictors of cribriform variant status and perineural invasion (PNI) in robot-assisted radical prostatectomy (RARP) histology. To define the rates of upgrading between biopsy specimens and final histology and their possible predictive factors in prostate cancer (PCa) patients undergoing RARP. Material and Methods: Within our institutional database, 265 PCa patients who underwent prostate biopsies and consecutive RARP at our center were enrolled (2018–2022). In the overall population, two independent multivariable logistic regression models (LRMs) predicting the presence of PNI or cribriform variant status at RARP were performed. In low- and intermediate-risk PCa patients according to D’Amico risk classification, three independent multivariable LRMs were fitted to predict upgrading. Results: Of all, 30.9% were low-risk, 18.9% were intermediate-risk and 50.2% were high-risk PCa patients. In the overall population, the rates of the cribriform variant and PNI at RARP were 55.8% and 71.1%, respectively. After multivariable LRMs predicting PNI, total tumor length in biopsy cores (>24 mm [OR: 2.37, p-value = 0.03], relative to <24 mm) was an independent predictor. After multivariable LRMs predicting cribriform variant status, PIRADS (3 [OR:15.37], 4 [OR: 13.57] or 5 [OR: 16.51] relative to PIRADS 2, all p = 0.01) and total tumor length in biopsy cores (>24 mm [OR: 2.47, p = 0.01], relative to <24 mm) were independent predicting factors. In low- and intermediate-risk PCa patients, the rate of upgrading was 74.4% and 78.0%, respectively. After multivariable LRMs predicting upgrading, PIRADS (PIRADS 3 [OR: 7.01], 4 [OR: 16.98] or 5 [OR: 20.96] relative to PIRADS 2, all p = 0.01) was an independent predicting factor. Conclusions: RARP represents a tailored and risk-adapted treatment strategy for PCa patients. The indication of RP progressively migrates to high-risk PCa after a pre-operative assessment. Specifically, the PIRADS score at mpMRI should guide the decision-making process of urologists for PCa patients. Full article

The binary logistic regression model (LRM) is practical in situations when the response variable (RV) is dichotomous. The maximum likelihood estimator (MLE) is generally considered to estimate the LRM parameters. However, in the presence of multicollinearity (MC), the MLE is not the correct choice due to its inflated standard deviation (SD) and standard errors (SE) of the estimates. To combat MC, commonly used biased estimators, i.e., the Ridge estimators (RE) and Liu estimators (LEs), are preferred. However, most of the time, the traditional LE attains a negative value for its Liu parameter (LP), which is considered to be a major drawback. Therefore, to overcome this issue, we proposed a new adjusted LE for the binary LRM. Owing to numerical evaluation purposes, Monte Carlo simulation (MCS) study is performed under different conditions where bias and mean squared error are the performance criteria. Findings showed the superiority of our proposed estimator in comparison with the other estimation methods due to the existence of high but imperfect multicollinearity, which clearly means that it is consistent when the regressors are multicollinear. Furthermore, the findings demonstrated that whenever there is MC, the MLE is not the best choice. Finally, a real application is being considered to be evidence for the advantage of the intended estimator. The MCS and the application findings pointed out that the considered adjusted LE for the binary logistic regression model is a more efficient estimation method whenever the regressors are highly multicollinear. Full article

The Indian Himalayan region is experiencing frequent hazards and disasters related to permafrost. However, research on permafrost in this region has received very little or no attention. Therefore, it is important to have knowledge about the spatial distribution and state of permafrost in the Indian Himalayas. Modern remote sensing techniques, with the help of a geographic information system (GIS), can assess permafrost at high altitudes, largely over inaccessible mountainous terrains in the Himalayas. To assess the spatial distribution of permafrost in the Alaknanda Valley of the Chamoli district of Uttarakhand state, 198 rock glaciers were mapped (183 active and 15 relict) using high-resolution satellite data available in the Google Earth database. A logistic regression model (LRM) was used to identify a relationship between the presence of permafrost at the rock glacier sites and the predictor variables, i.e., the mean annual air temperature (MAAT), the potential incoming solar radiation (PISR) during the snow-free months, and the aspect near the margins of rock glaciers. Two other LRMs were also developed using moderate-resolution imaging spectroradiometer (MODIS)-derived land surface temperature (LST) and snow cover products. The MAAT-based model produced the best results, with a classification accuracy of 92.4%, followed by the snow-cover-based model (91.9%), with the LST-based model being the least accurate (82.4%). All three models were developed to compare their accuracy in predicting permafrost distribution. The results from the MAAT-based model were validated with the global permafrost zonation index (PZI) map, which showed no significant differences. However, the predicted model exhibited an underestimation of the area underlain by permafrost in the region compared to the PZI. Identifying the spatial distribution of permafrost will help us to better understand the impact of climate change on permafrost and its related hazards and provide necessary information to decision makers to mitigate permafrost-related disasters in the high mountain regions. Full article

If sufficient historical failure life data exist, the failure distribution of the system can be estimated to identify the system initial hazard function. The conventional proportional covariate model (PCM) can reveal the dynamic relationship between the response covariates and the system hazard rate. The system hazard rate function can be constantly updated by the response covariates through the basic covariate function (BCF). Under the circumstances of sparse or zero failure data, the key point of the PCM reliability assessment method is to determine the proportional factor between covariates and the hazard rate for getting BCF. Being devoid of experiments or abundant experience of the experts, it is very hard to determine the proportional factor accurately. In this paper, an improved PCM (IPCM) is put forward based on the logistic regression model (LRM). The salient features reflecting the equipment degradation process are extracted from the existing monitoring signals, which are considered as the input of the LRM. The equipment state data defined by the failure threshold are considered as the output of the LRM. The initial reliability can be first estimated by LRM. Combined with the responding covariates, the initial hazard function can be calculated. Then, it can be incorporated into conventional PCM to implement the reliability estimation process on other equipment. The conventional PCM and the IPCM methods are respectively applied to aero-engine rotor bearing reliability assessment. The comparative results show that the assessing accuracy of IPCM is superior to the conventional PCM for small failure sample. It provides a new method for reliability estimation under sparse or zero failure data conditions. Full article