Log Likelihood Research Articles

What is the optimal range of fasting stress hyperglycemia ratio for all-cause mortality in American adults: An observational study.

To date, no studies have been conducted to assess the impact of fasting stress hyperglycemia ratio (SHR) on all-cause mortality. Therefore, the objective of our study is to investigate the association between SHR and all-cause mortality in a population of American adults. The study population was derived from NHANES data spanning from 2005 to 2018. The exposure variable SHR was derived from fasting blood glucose (FBG) and glycosylated hemoglobin (HbA1c%), and the specific calculation formula was as follows: (FBG (mmol/L))/(1.59 × HbA1c (%) - 2.59). The outcome variable was all-cause mortality. A total of 18,457 participants were enrolled in this prospective cohort study. Following a median follow-up period of 90 months, all-cause mortality was observed in 10.32% of the subjects. Cox proportional hazards regression model indicates that there is no significant difference between SHR and all-cause mortality in the fully adjusted model, whether analyzed as a continuous or categorical variable (P for trend > 0.05). Through the 2-piecewise Cox proportional hazards regression model, we have determined that the inflection point of SHR in relation to all-cause mortality is 0.88. It has also been observed that when the value of SHR is on the left side of the inflection point (SHR ≤ 0.88), there is a significant 77% (HR: 0.23; 95% CI: 0.10-0.50) reduction in all-cause mortality for each additional unit increase in SHR. Conversely, when the value of SHR exceeds 0.88, there is a substantial 2.40-fold (HR: 2.40; 95% CI: 1.61-3.58) increase in the risk of all-cause mortality (P for log likelihood ratio test < .001). The subgroup analysis results demonstrate that sex has the potential to modify the association between SHR and all-cause mortality within the population exhibiting SHR ≤ 0.88. The relationship between SHR and all-cause mortality follows a U-shaped pattern, where in the lowest risk of death for the average American adult is observed at an SHR value of 0.88. Furthermore, in men with SHR ≤ 0.88, there is a significant inverse relationship between the increase in SHR and the risk of all-cause mortality.

Efficient algorithm for thermal nondestructive testing and evaluation by considering the heteroscedastic nature of noise sources in infrared thermography

Abstract Thermal Imaging is a promising Non Destructive Testing &amp; Evaluation (NDT &amp; E) approach to monitor the health&amp;#xD;of composite materials. Among various post processing approaches adopted in thermal imaging for NDT &amp; E, statistical&amp;#xD;analysis schemes gained importance due to their reliability and data reduction capabilities. This paper provides an insight&amp;#xD;to a factor analysis-based statistical approach to detect the hidden defects in the Glass Fiber Reinforced Polymer (GFRP)&amp;#xD;sample. The proposed approach models the observed data covariance into combination of temporal signal covariance&amp;#xD;and noise covariance matrices. The modeling of the diagonal covariance matrix (with different elements) is motivated by&amp;#xD;the presence of heterogeneity in the experimental data obtained from GFRP sample. This novel method is based on the&amp;#xD;coordinate descent technique, which estimates the covariance matrix of the noise variances iteratively by minimizing the&amp;#xD;negative log likelihood function. The obtained results from the chosen GFRP samples compared with the widely used&amp;#xD;statistical Principal Component Thermography (PCT) technique illustrate the improved performance in terms of defect&amp;#xD;detection with the proposed technique.&amp;#xD;

Generative adversarial networks for whispered to voiced speech conversion: a comparative study

AbstractGenerative Adversarial Networks (GANs) have demonstrated promising results as end-to-end models for whispered to voiced speech conversion. Leveraging non-autoregressive systems like GANs capable of performing conditional waveform generation eliminates the need for separate models to estimate voiced speech features, and leads to faster inference compared to autoregressive methods. This study aims to identify the optimal GAN architecture for the whispered to voiced speech conversion task by comparing six state-of-the-art models. Furthermore, we present a method for evaluating the preservation of speaker identity and local accent, using embeddings obtained from speaker- and language identification systems. Our experimental results show that building the speech conversion system based on the HiFi-GAN architecture yields the best objective evaluation scores, outperforming the baseline by $$\sim$$ ∼ 9% relative using frequency-weighted Signal-to-Noise Ratio and Log Likelihood Ratio, as well as by $$\sim$$ ∼ 29% relative using Root Mean Squared Error. In subjective tests, HiFi-GAN yielded a mean opinion score of 2.9, significantly outperforming the baseline with a score of 1.4. Furthermore, HiFi-GAN enhanced ASR performance and preserved speaker identity and accent, with correct language detection rates of up to $$\sim$$ ∼ 98%.

The association between triglyceride-glucose index and gallstones: NHANES 2017-2020

ObjectsIt remains unclear whether the triglyceride-glucose (TyG) index has correlations with gallstones. This study aimed to investigate the association between TyG index and gallstones.MethodsData was obtained from the 2017-2020 National Health and Nutrition Examination Survey (NHANES). Participants who provided complete data about TyG index and gallstones were included in the analysis. Multivariable regression analysis and subgroup analysis were preformed to estimate the relationship between TyG index and gallstones. Restricted cubic splines (RSC) was employed to calculate the cut off value.ResultsThe TyG index was independently associated with gallstones and demonstrates a clear positive correlation (OR = 1.10; 95% CI: 1.01–1.21; p = 0.033). The threshold value is 8.98, showing a positive correlation between TyG index and gallstones when the TyG index is less than 8.98 (Log likelihood ratio P &amp;lt; 0.001). Subgroup analysis indicates that the correlation between TyG and gallstones is mainly observed in individuals with obesity, females, younger individuals, and those with normal blood sugar levels, with these subgroups all acting as mediators between TyG and gallstones.ConclusionsHigher TyG index was linked to a higher chance of developing gallstones. Managing insulin resistance (IR) could help reduce the risk of gallstones since the TyG index is an indicator of IR.

A novel performance prediction method for harmonic reducers based on the trend-enhanced Fisher’s discriminant ratio-Wiener process

Abstract Harmonic reducers, as core components of industrial robots, play a critical role in maintaining robot health. Performance degradation assessment (PDA) and remaining useful life (RUL) prediction are essential for ensuring the operational reliability of such robots. To address the multistage characteristics and stochastic uncertainties in the performance degradation of harmonic reducers, a performance prediction method based on Fisher’s discriminant ratio and Wiener process is proposed. Firstly, the health indicator (HI) construction is defined as an optimization problem based on Fisher’s discriminant ratio and trend monotonicity constraints. By leveraging the sum of the multiplication of frequency amplitudes and optimized weights, precise segmentation of the multistage degradation process over the entire lifecycle is achieved. Notably, the optimized weights can automatically identify the resonance frequency bands caused by damage. Subsequently, a nonlinear Wiener process model with a drift coefficient in the form of a power function is established. The probability density function (PDF) expression for RUL is derived based on the concept of first hit time (FHT). Additionally, a logarithmic likelihood function (Log-LF) for the unknown parameters in the degradation model is constructed. The experimental results indicate that the proposed method surpasses the other two Wiener process models in terms of root mean square error (RMSE), mean absolute percentage error (MAPE), and cumulative relative accuracy (CRA). This provides robust support for preventive maintenance decision-making for harmonic reducers.

A novel data-driven state evaluation approach for photovoltaic arrays in uncertain shading scenarios

Accurately determining the operating state of photovoltaic (PV) power-generation equipment and providing accurate support for maintenance management requires evaluation of the states of PV arrays under partial shading conditions. Results of such evaluation enable the optimization of maintenance measures to reduce maintenance costs and extend the lifespan of PV modules, thereby increasing the overall revenue of the PV power station. Therefore, this study proposes a novel state-evaluation approach based on current-voltage (I-V) curves under partial shading conditions. Based on the size of the pixel values in this curve, the proposed novel state-evaluation was performed using the proportion of the values of the normal and shaded curves under the same environmental conditions. Specifically, the feature variables (open-circuit voltage, short-circuit current) and pixel values within the I-V curve were analyzed and calculated using a sensitivity algorithm and computer vision algorithms (Canny edge-detection algorithm and Green's theorem). Based on the mapping relationship between feature variables and pixels, an improved Gaussian-process regression method with a combined kernel function (dot product and Matérn function) was proposed to predict the pixel values under the corresponding normal curve. This combined kernel function effectively captured causal relationships and handled outliers in the test dataset to improve the robustness of the model. Finally, the ratios of the pixel values were calculated using the predicted pixels within the normal curve and the calculated pixels within the shading curve. Experiments demonstrated that the root mean square error (RMSE), mean absolute error (MAE), R squared (R2) and log likelihood of the improved Gaussian-process regression method reached 0.011, 0.0086, 0.9996, and 24.7096, respectively.

An Empirical Model for Skillful Rogue Wave Forecasts

Abstract One of the leading goals of rogue wave research is to develop a robust rogue wave warning system to mitigate the danger they pose. One such system has been developed by the European Centre for Medium-Range Weather Forecasts (ECMWF), called the freak wave warning system (FWWS), based on nonlinear wave effects. The FWWS predicts maximum expected wave envelope height as a risk parameter for forecasts. Recently, a data-driven alternative has been proposed by Häfner et al., which was distilled from a neural network using wave buoy observations. However, it has yet to be evaluated by a spectral wave model for application to operational wave forecasting. The data-driven, learned model emphasizes bandwidth-controlled linear superposition as the predominant mechanism in crest-to-trough rogue wave generation, while nonlinear effects are a secondary term. The present work evaluates the performance of the empirical model using output from an ECMWF global wave hindcast. We find that the prediction models based on bandwidth effects have the highest log likelihood scores, with the empirical model outperforming all other tested models. In contrast, the expected maximum envelope wave height from the FWWS does not predict the occurrence of rogue waves. These results indicate that the empirical model with wave model input is a skillful predictor and should be considered for operational implementation to improve rogue wave forecasting. Significance Statement Rogue waves are unexpectedly large and unpredictable waves. Encounters with rogue waves can result in damage to marine vessels and offshore infrastructures. Research is devoted to developing systems to predict the risk of rogue wave events. A novel predictive model has been shown to perform well in predicting rogue wave occurrences based on wave buoy observations. This model is a symbolic expression distilled from an artificial neural network that incorporates known rogue wave dynamics and that can be evaluated alongside traditional risk estimates with minor adjustments to operational forecasting systems. This study evaluates the newly proposed empirical model in a forecasting setting. Our work demonstrates the efficacy of the empirical model for operational forecasting purposes.

Profile of childhood headaches in a tertiary center

BackgroundThe aims of this study were to analyze the profile of childhood headache complaints in a tertiary headache center, verify the differences between children and adolescents, and provide the clinical characteristics that are related to the headaches.MethodsA review of 384 medical records of children aged between 4-12 years old (n=206) and adolescents (n=178) aged &gt;12 to 18 years old. The variables evaluated were reported as percentages, compared between children and adolescents and selected the variables that were related to headaches diagnosis.ResultsThe majority of the sample were females (60.7%), diagnosis of migraine (70.3%), pulsatile pain (60.2%), episodic attacks (60.2%) and no need for imaging tests associated with the diagnosis (69.3%), which were associated with different clinical characteristics (LR X2(52); p&lt;0.001; Log likelihood = -322.434; Pseud R2 = 0.154). The group of adolescents had a higher proportion of use of antidepressants (29.8%) as prophylactic medication than children, who on the other hand, had a higher prevalence of use of antivertiginous drugs (32%). The abortive medication used was common analgesics (37.8%) and the main outcome after treatment was discharge from the service (32%).ConclusionThe evaluation and the treatment for these patients should take into account the main complaints, considering the particularities of each type of headache and also the age group, in order to identify, treat the disease properly and avoid chronification and continuity in the tertiary service, referring them to less complex services.

Approaches to Modelling Marketing Strategies in E-Commerce

Introduction. Increasing the efficiency of business activities requires developing a marketing strategy using e-commerce to attract a wide range of customers and maximise the company's income. The key marketing metric for most companies is the customer lifetime value indicator, which is why studying approaches to forecasting this indicator in e-commerce requires special attention, considering the changing business operation environment. Aim and tasks. The study aims to develop approaches to modelling the marketing strategy of e-commerce enterprises. The tasks are a comparative characterisation of modern marketing planning tools and an empirical analysis of the possibilities of using the BG/NBD model for marketing planning in e-commerce companies, considering the need to implement scenario marketing planning in unstable conditions. Results. A comparison of the characteristics of the main marketing planning tools was conducted according to features such as content and focus, functional purpose, and planning period, which allows more purposeful use of these tools to solve specific tasks. In order to automatically obtain the required data format for modelling, a Visual Basic program was developed, a forecast of customer behaviour was built, and an indicator of customer transactions was determined for further calculation of customer lifetime value (CLV). During the simulation, four parameters of the BG/NBD model were evaluated, and the maximum value of the logarithmic likelihood function equal to 14460.54 was obtained. The expected number of transactions (forecast time = 26 weeks) was determined for each client from the database. After calculations the information was received, the selected client will make 2.35 transactions during the forecast period of 55-80 weeks. The recommended algorithm for using the BG/NBD model as part of a scenario approach for forecasting the cost of the customer's life cycle allows for more flexible forecasting of digital business revenues and building an optimal marketing strategy. Conclusions. An empirical study of the possibilities of using the BG/NBD model for marketing planning in e-commerce companies showed that, for the simulation of indicators that affect a company's marketing strategy, it is crucial at the initial stage to convert the database of customer purchases into the input format required for modelling. Using a scenario approach allows e-commerce enterprises to carry out variable marketing planning and reduce risks under growing uncertainty. A well-designed marketing strategy that can be adapted to different business development scenarios enables an e-commerce company to increase its sales.

Mapping Risk–Return Linkages and Volatility Spillover in BRICS Stock Markets through the Lens of Linear and Non-Linear GARCH Models

This paper explores the influence of the risk–return relationship and volatility spillover on stock market returns of emerging economies, with a particular focus on the BRICS countries. This research is undertaken in a context where discussions on de-dollarization and the expansion of BRICS membership are gaining momentum, making it a novel and distinct exercise compared to prior studies. Utilizing econometric techniques to investigate daily market returns from 1 April 2008 to 31 March 2023, a period that witnessed major events like the global financial crisis, the COVID-19 pandemic, and the Russia–Ukraine conflict, linear and non-linear models like ARCH, GARCH, GARCH-M, EGARCH, and TGARCH, are employed to assess stock return volatility behaviour, assuming a Gaussian distribution of error terms. The diagnostic test confirms that the distribution is non-normal, stationary, and heteroscedastic. The key findings indicate a lack of the risk–return relationship across all BRICS stock markets, except for South Africa; a more pronounced effect of unpleasant news over pleasant news; a slow mean-reverting process in volatility; the EGARCH model is the best fit model as evidenced by a higher log likelihood and lower Akaike information criterion and Schwardz information criterion parameters; and finally, the presence of significant bidirectional and unidirectional spillover effects in the majority of instances. These findings are valuable for investors, regulators, and policymakers in enhancing returns and mitigating risk through portfolio diversification and informed decision making.

Dictionary learning-based denoising algorithm with expected patch log likelihood in diffusion-weighted magnetic resonance image

Dictionary learning-based denoising algorithm with expected patch log likelihood in diffusion-weighted magnetic resonance image

Epidemiological Characteristics and Spatiotemporal Clustering of Pulmonary Tuberculosis Among Students in Southwest China From 2016 to 2022: Analysis of Population-Based Surveillance Data.

Pulmonary tuberculosis (PTB), as a respiratory infectious disease, poses significant risks of covert transmission and dissemination. The high aggregation and close contact among students in Chinese schools exacerbate the transmission risk of PTB outbreaks. This study investigated the epidemiological characteristics, geographic distribution, and spatiotemporal evolution of student PTB in Chongqing, Southwest China, aiming to delineate the incidence risks and clustering patterns of PTB among students. PTB case data from students monitored and reported in the Tuberculosis Information Management System within the China Information System for Disease Control and Prevention were used for this study. Descriptive analyses were conducted to characterize the epidemiological features of student PTB. Spatial trend surface analysis, global and local spatial autocorrelation analyses, and disease rate mapping were performed using ArcGIS 10.3. SaTScan 9.6 software was used to identify spatiotemporal clusters of PTB cases. From 2016 to 2022, a total of 9920 student TB cases were reported in Chongqing, Southwest China, with an average incidence rate of 24.89/100,000. The incidence of student TB showed an initial increase followed by a decline, yet it remained relatively high. High school students (age: 13-18 years; 6649/9920, 67.03%) and college students (age: ≥19 years; 2921/9920, 29.45%) accounted for the majority of student PTB cases. Patient identification primarily relied on passive detection, with a high proportion of delayed diagnosis and positive etiological results. COVID-19 prevention measures have had some impact on reducing incidence levels, but the primary factor appears to be the implementation of screening measures, which facilitated earlier case detection. Global spatial autocorrelation analysis indicated Moran I values of >0 for all years except 2018, ranging from 0.1908 to 0.4645 (all P values were <.05), suggesting strong positive spatial clustering of student PTB cases across Chongqing. Local spatial autocorrelation identified 7 high-high clusters, 13 low-low clusters, 5 high-low clusters, and 4 low-high clusters. High-high clusters were predominantly located in the southeast and northeast parts of Chongqing, consistent with spatial trend surface analysis and spatiotemporal clustering results. Spatiotemporal scan analysis revealed 4 statistically significant spatiotemporal clusters, with the most likely cluster in the southeast (relative risk [RR]=2.87, log likelihood ratio [LLR]=574.29, P<.001) and a secondary cluster in the northeast (RR=1.99, LLR=234.67, P<.001), indicating higher reported student TB cases and elevated risks of epidemic spread within these regions. Future efforts should comprehensively enhance prevention and control measures in high-risk areas of PTB in Chongqing to mitigate the incidence risk among students. Additionally, implementing proactive screening strategies and enhancing screening measures are crucial for early identification of student patients to prevent PTB outbreaks in schools.

Anomaly detection via Gumbel Noise Score Matching.

We propose Gumbel Noise Score Matching (GNSM), a novel unsupervised method to detect anomalies in categorical data. GNSM accomplishes this by estimating the scores, i.e., the gradients of log likelihoods w.r.t. inputs, of continuously relaxed categorical distributions. We test our method on a suite of anomaly detection tabular datasets. GNSM achieves a consistently high performance across all experiments. We further demonstrate the flexibility of GNSM by applying it to image data where the model is tasked to detect poor segmentation predictions. Images ranked anomalous by GNSM show clear segmentation failures, with the anomaly scores strongly correlating with segmentation metrics computed on ground-truth. We outline the score matching training objective utilized by GNSM and provide an open-source implementation of our work.

Exploring the influencing factors of scrub typhus in Gannan region, China, based on spatial regression modelling and geographical detector

Scrub typhus is a significant public health issue with a wide distribution and is influenced by various determinants. However, in order to effectively eradicate scrub typhus, it is crucial to identify the specific factors that contribute to its incidence at a detailed level. Therefore, the objective of our study is to identify these influencing factors, examine the spatial variations in incidence, and analyze the interplay of two factors on scrub typhus incidence, so as to provide valuable experience for the prevention and treatment of scrub typhus in Gannan and to alleviate the economic burden of the local population.This study employed spatial autocorrelation analyses to examine the dependent variable and ordinary least squares model residuals. Additionally, spatial regression modelling and geographical detector were used to analyze the factors influencing the annual mean 14-year incidence of scrub typhus in the streets/townships of Gannan region from 2008 to 2021. The results of spatial1 autocorrelation analyses indicated the presence of spatial correlation. Among the global spatial regression models, the spatial lag model was found to be the best fitting model (log likelihood ratio = −319.3029, AIC = 666.6059). The results from the SLM analysis indicated that DEM, mean temperature, and mean wind speed were the primary factors influencing the occurrence of scrub typhus. For the local spatial regression models, the multiscale geographically weighted regression was determined to be the best fitting model (adjusted R2 = 0.443, AICc = 726.489). Further analysis using the MGWR model revealed that DEM had a greater impact in Xinfeng and Longnan, while the southern region was found to be more susceptible to scrub typhus due to mean wind speed. The geographical detector results revealed that the incidence of scrub typhus was primarily influenced by annual average normalized difference vegetation index. Additionally, the interaction between GDP and the percentage of grassland area had a significant impact on the incidence of scrub typhus (q = 0.357). This study illustrated the individual and interactive effects of natural environmental factors and socio-economic factors on the incidence of scrub typhus; and elucidated the specific factors affecting the incidence of scrub typhus in various streets/townships. The findings of this study can be used to develop effective interventions for the prevention and control of scrub typhus.

Gender differences in the relationship between the triglyceride-glucose index and serum Klotho concentrations among the middle-aged and elderly: a cross-sectional analysis

BackgroundThe triglyceride-glucose (TyG) index is recognized as a robust indicator for evaluating insulin resistance (IR). Despite the well-documented anti-aging biological functions of Klotho protein, its correlation with the TyG index remains unexplored.MethodsA cross-sectional analysis was conducted involving participants from the National Health and Nutrition Examination Surveys (NHANES) 2007–2016. The TyG index was computed using laboratory data, while serum Klotho concentrations was determined using ELISA kit. After adjusting potential confounding variables, multivariate regression models were employed to evaluate the association between the TyG index and Klotho protein levels among middle-aged and elderly females and males separately. Additionally, smooth curve fitting and segmented regression model were applied to investigate potential threshold effects and identify the inflection point.ResultsA total of 6,573 adults qualified for inclusion, comprising 3,147 (47.88%) males and 3,426 (52.12%) females. Multivariate regression analysis revealed that females with a higher TyG index exhibited significantly lower serum Klotho concentrations (β=-83.41, 95% CI: -124.23 to -42.60, P < 0.0001). This association was not statistically significant in males (β = 15.40, 95% CI: -19.16 to 49.95, P = 0.3827). Subgroup analyses revealed a significant interaction effect by diabetes status in females (P-interaction = 0.0121), where non-diabetic females showed a stronger negative association between TyG index and serum Klotho levels compared to diabetic females. In the female group, when TyG index was divided into quartiles, individuals in the highest quartile of TyG index exhibited reduced levels of Klotho protein (Q4: -88.77 pg/ml) compared to those in the lowest quartile (Q1) after full adjustment (P = 0.0041). Segmented regression analysis indicated a turning point value of 9.4 in females. Notably, a 1-unit increase in TyG index was significantly associated with a decrease in Klotho levels by -111.43 pg/ml (95% CI: -157.34 to -65.52, P < 0.0001) when TyG index was below 9.4, while above this threshold, the association was not significant (Log likelihood ratio test: 0.009).ConclusionsThe findings highlight a non-linear correlation between the TyG index and serum Klotho concentrations among females, indicative of a saturation effect. This relationship was particularly pronounced in non-diabetic women. In contrast, no statistically significant association was observed in male participants.

Ophthalmological Examination and Imaging Features that Better Predict Risk of Angle Closure in Siblings

Characteristics of individuals with angle closure may be useful in targeted screening of family members. Here, we assess if findings gathered during examination and imaging of patients with a known angle closure diagnosis (probands) could better determine the risk of angle closure in the patients' siblings. Cross-sectional study of patients with known angle closure and their siblings. Participants, and Controls: South Indian patients (probands) 30 years and older with open angles, suspect primary angle closure (PACS), or primary angle closure/primary angle closure glaucoma (PAC/PACG), and a biological sibling age 30 years or older (n=292 proband/sibling pairs). Demographic data, relevant ocular history, and a comprehensive ophthalmic examination with Anterior Segment Optical Coherence Tomography (ASOCT) were obtained. Three clinically relevant models were created to analyze the contribution of specific proband factors in predicting sibling angle closure diagnosis, using demographic (age, gender), ocular exam (gonioscopy, optic nerve exam, visual acuity, intraocular pressure [IOP]), and ASOCT features to improve prediction beyond proband diagnosis alone evaluated by log likelihood ratio testing and statistical comparison of receiver operating characteristics (ROC). Sibling angle closure diagnostic accuracy. Demographic and ocular exam metrics did not improve the prediction of sibling angle closure for all three outcomes (sibling diagnosis: (1) PACS/PAC/PACG vs OA, (2) PAC/PACG vs PACS/OA, and (3) PAC/PACG vs PACS), adding no model improvement when compared to diagnosis alone. Models adding ASOCT metrics to the prior model including proband diagnosis, demographics and ocular exam measures led to significantly improved prediction of 2 of the 3 angle closure outcomes. Specifically, improvement was noted via likelihood ratio testing for prediction of PAC/PACG vs PACS/OA (p=0.01), or PAC/PACG vs PACS (p=0.001). For all 3 angle closure outcomes, ROC comparisons demonstrated significant improvement in AUC between the three models predicting sibling outcomes, demonstrating an increase in AUC with each successive nested model across all 3 sibling angle closure outcomes. Structural features of eyes with angle closure may assist in stratifying the risk of angle closure in patients' siblings. Further studies should consider evaluating this approach to achieve more targeted screenings.

Machine Learning Model for Anesthetic Risk Stratification for Gynecologic and Obstetric Patients: Cross-Sectional Study Outlining a Novel Approach for Early Detection.

Preoperative evaluation is important, and this study explored the application of machine learning methods for anesthetic risk classification and the evaluation of the contributions of various factors. To minimize the effects of confounding variables during model training, we used a homogenous group with similar physiological states and ages undergoing similar pelvic organ-related procedures not involving malignancies. Data on women of reproductive age (age 20-50 years) who underwent gestational or gynecological surgery between January 1, 2017, and December 31, 2021, were obtained from the National Taiwan University Hospital Integrated Medical Database. We first performed an exploratory analysis and selected key features. We then performed data preprocessing to acquire relevant features related to preoperative examination. To further enhance predictive performance, we used the log-likelihood ratio algorithm to generate comorbidity patterns. Finally, we input the processed features into the light gradient boosting machine (LightGBM) model for training and subsequent prediction. A total of 10,892 patients were included. Within this data set, 9893 patients were classified as having low anesthetic risk (American Society of Anesthesiologists physical status score of 1-2), and 999 patients were classified as having high anesthetic risk (American Society of Anesthesiologists physical status score of >2). The area under the receiver operating characteristic curve of the proposed model was 0.6831. By combining comorbidity information and clinical laboratory data, our methodology based on the LightGBM model provides more accurate predictions for anesthetic risk classification. Research Ethics Committee of the National Taiwan University Hospital 202204010RINB; https://www.ntuh.gov.tw/RECO/Index.action.

A Cross-sectional Comparative Analysis of Eleven Population Pharmacokinetic Models for Docetaxel in Chinese Breast Cancer Patients.

Various population pharmacokinetic (PPK) models have been established to help determine the appropriate dosage of docetaxel, however, no clear consensus on optimal dosing has been achieved. The purpose of this study is to perform an external evaluation of published models in order to test their predictive performance, and to find an appropriate PPK model for Chinese breast cancer patients. A systematic literature search of docetaxel PPK models was performed using PubMed, Web of Science, China National Knowledge Infrastructure, and WanFang databases. The predictive performance of eleven identified models was evaluated using prediction-based and simulation-based diagnostics on an independent dataset (112 docetaxel concentrations from 56 breast cancer patients). The -2×log (likelihood) and Akaike information criterion were also calculated to evaluate model fit. The median prediction error of eight of the eleven models was less than 10%. The model fitting results showed that the three-compartment model of Bruno et al. had the best prediction performance and that the three compartment model of Wang et al. had the best simulation effect. Furthermore, although the covariates that significantly affect PK parameters were different between them, seven models demonstrated that docetaxel PK parameters were influenced by liver function. Three compartment PPK models may be predictive of optimal docetaxel dosage for Chinese breast cancer patients. However, for patients with impaired liver function, the choice of which model to use to predict the blood concentration of docetaxel still requires great care.

Association between the oxidative balance score and estimated pulse wave velocity from the National Health and Nutrition Examination Survey (2005–2018)

BackgroundNo research report has been conducted to investigate the impact of oxidation balance score (OBS) on the estimated pulse wave velocity(ePWV).We aimed to examine the association between OBS and ePWV.MethodWe evaluated data for 13,073 patients from the National Health and Nutrition Examination Survey (NHANES). The exposure variable was OBS. The outcome variables was combination of ePWV and arterial stiffness.ResultsWe observed a significant negative correlation between OBS (Per 1SD increase) and ePWV in the gradually adjusted models. Based on the aforementioned results, a two-piecewise logistic regression adjusted model was subsequently employed to establish the association between OBS and elevated ePWV, and the inflection point was determined as 5. The increased risk of elevated ePWV (OR:0.70; 95%CI:0.51–0.94) gradually decreases with the increase of OBS on the left side of the inflection point; however, when OBS exceeds 5, this decrease in risk of elevated ePWV(OR:1.00; 95%CI:0.96–1.04) is no longer observed (P for log likelihood ratio test = 0.028).ConclusionsThere exists a significant association between OBS and ePWV in the context of American adults. Specifically, OBS exhibits a negative correlation with ePWV; however, when considering an elevated ePWV, a saturation effect is observed in relation to OBS.

Power-law properties of nocturnal arrhythmia avalanches: A novel marker for incident cardiovascular events

BackgroundBursting nonsustained cardiac arrhythmia events are a common observation during sleep. ObjectivesThe purpose of this study was to investigate the hypothesis that nocturnal arrhythmia episode durations could follow a power law, whose exponent could predict long-term clinical outcomes. MethodsWe defined “nocturnal arrhythmia avalanche” (NAA) as any instance of a drop in electrocardiographic (ECG) template-matched R-R intervals ≥30% of R-R baseline, followed by a return to 90% of baseline. We studied NAA in ECG recordings obtained from the Sleep Heart Health Study (SHHS), Osteoporotic Fractures in Men Study (MrOS) Study, and Multi-Ethnic Study of Atherosclerosis (MESA). The association of nocturnal arrhythmia durations with a power-law distribution was evaluated and the association of derived power-law exponents (α) with major adverse cardiovascular (CV) events and mortality assessed with multivariable Cox regression. ResultsA total of 9176 participants were studied. NAA episodes distribution was consistent with power-law vs comparator distributions in all datasets studied (positive log likelihood ratio of power-law vs exponential in MESA: 83%; SHHS: 69%; MrOS: 81%; power-law vs log-normal in MESA: 95%; SHHS: 35%; MrOS: 64%). The NAA power-law exponent (α) showed a significant association of with adverse CV outcomes (association with CV mortality: SHHS hazard ratio 1.39 [1.07–1.79], P = .012; MrOS hazard ratio 1.42 [1.02–1.94], P = .039; association with CV events: MESA HR 3.46 [1.46–8.21], P = .005) in multivariable Cox regression, after adjusting for conventional CV risk factors and nocturnal ectopic rate. ConclusionThe NAA power-law exponent is a reproducible, predictive marker for incident CV events and mortality.