Poor Predictive Performance Research Articles

Triple the Gamma—A Unifying Shrinkage Prior for Variance and Variable Selection in Sparse State Space and TVP Models

Time-varying parameter (TVP) models are very flexible in capturing gradual changes in the effect of explanatory variables on the outcome variable. However, in particular when the number of explanatory variables is large, there is a known risk of overfitting and poor predictive performance, since the effect of some explanatory variables is constant over time. We propose a new prior for variance shrinkage in TVP models, called triple gamma. The triple gamma prior encompasses a number of priors that have been suggested previously, such as the Bayesian Lasso, the double gamma prior and the Horseshoe prior. We present the desirable properties of such a prior and its relationship to Bayesian Model Averaging for variance selection. The features of the triple gamma prior are then illustrated in the context of time varying parameter vector autoregressive models, both for simulated dataset and for a series of macroeconomics variables in the Euro Area.

Machine learning highlights the deficiency of conventional dosimetric constraints for prevention of high-grade radiation esophagitis in non-small cell lung cancer treated with chemoradiation

Background and PurposeRadiation esophagitis is a clinically important toxicity seen with treatment for locally-advanced non-small cell lung cancer. There is considerable disagreement among prior studies in identifying predictors of radiation esophagitis. We apply machine learning algorithms to identify factors contributing to the development of radiation esophagitis to uncover previously unidentified criteria and more robust dosimetric factors. Materials and MethodsWe used machine learning approaches to identify predictors of grade ≥ 3 radiation esophagitis in a cohort of 202 consecutive locally-advanced non-small cell lung cancer patients treated with definitive chemoradiation from 2008 to 2016. We evaluated 35 clinical features per patient grouped into risk factors, comorbidities, imaging, stage, histology, radiotherapy, chemotherapy and dosimetry. Univariate and multivariate analyses were performed using a panel of 11 machine learning algorithms combined with predictive power assessments. ResultsAll patients were treated to a median dose of 66.6 Gy at 1.8 Gy per fraction using photon (89.6%) and proton (10.4%) beam therapy, most often with concurrent chemotherapy (86.6%). 11.4% of patients developed grade ≥ 3 radiation esophagitis. On univariate analysis, no individual feature was found to predict radiation esophagitis (AUC range 0.45–0.55, p ≥ 0.07). In multivariate analysis, all machine learning algorithms exhibited poor predictive performance (AUC range 0.46–0.56, p ≥ 0.07). ConclusionsContemporary machine learning algorithms applied to our modern, relatively large institutional cohort could not identify any reliable predictors of grade ≥ 3 radiation esophagitis. Additional patients are needed, and novel patient-specific and treatment characteristics should be investigated to develop clinically meaningful methods to mitigate this survival altering toxicity.

Unsupervised automatic seizure detection for focal-onset seizures recorded with behind-the-ear EEG using an anomaly-detecting generative adversarial network

Background and Objective: Epilepsy is a neurological disorder of the brain, which involves recurrent seizures. An encephalogram (EEG) is a gold standard method in the detection and analysis of epileptic seizures. However, the standard EEG recording system is too obstructive to be used in daily life. Behind-the-ear EEG is an alternative approach to record EEG conveniently. Previous researchers applied machine learning to automatically detect seizures with EEG, but the epileptic EEG waveform contains subtle changes that are difficult to be identified. Furthermore, the extremely small proportion of ictal events in the long-term monitoring may cause the imbalance problem and, consequently, poor prediction performance in supervised learning approaches. In this study, we present an automatic seizure detection algorithm with a generative adversarial network (GAN) trained by unsupervised learning and evaluated it with behind-the-ear EEG.Methods: We recorded behind-the-ear EEGs from 12 patients who have various types of epilepsy. Data were reviewed separately by two epileptologists, who determined the onsets and ends of seizures. First, we conducted unsupervised learning with the normal records for the GAN to learn the representation of normal states. Second, we performed automatic seizure detection with the trained GAN as an anomaly detector. Last, we combined the Gram matrix with other anomaly losses to improve detection performance.Results: The proposed approach achieved detection performance with an area under the receiver operating curve of 0.939 and sensitivity of 96.3% with a false alarm rate of 0.14 per hour in the test dataset. In addition, we confirmed distinguishability with the distribution of the anomaly scores in terms of EEG frequency bands.Conclusions: It is expected that the proposed anomaly detection via GAN with the behind-the-ear EEG can be effectively used for long-term seizure monitoring in daily life.

Abstract TP13: External Validation of Stroke Prognostication Scales - A Retrospective Cohort Study

Introduction: In recent years, many scoring models have been proposed to predict clinical outcomes after acute ischemic stroke. Aim of our study was to perform a comparative analysis of these scoring systems to assess predictive reliability. Method: This retrospective single center study included 166 community-based patients presenting with an acute ischemic stroke between 2015 and 2018 who had undergone mechanical thrombectomy with or without IV r-tPA administration prior to the procedure. Patients with unknown 90 day Modified Ranking Scale (mRS) were excluded from the study. We included SPAN-100, THRIVE, HIAT2, iScore , TPI, DRAGON, ASTRAL and HAT predictive models to our study. To predict MRS at 90 days, we first dichotomize mRS into two groups: scores of 0 and 1 and scores 2 and above. We then used logistic regression to find the association between a stroke score and the probability of having a 90-day mRS of 2 or above. Separate univariate logistic regressions were fit for each stroke score. We assessed the ability of each stroke score to predict 90-day mRS using the area-under-the-curve (AUC) of the receiver operating characteristic curve (ROC - plot of sensitivity against 1-specificity). AUC values range from 0.5 to 1 with values above 0.7 showing good discriminatory ability. Results: SPAN-100, HIAT2, iScore, and ASTRAL scores have similar predictive ability with AUC values over 0.7 (Table 1). The ASTRAL score had the highest predictive ability with a score above 31.5 indicating a high likelihood of a 90-day MRS ≥ 2 (Table 1). The TPI, DRAGON, and HAT scores all had AUCs below 0.65 indicating poor predictive performance in our data. Conclusion: The SPAN-100, HIAT2, iScore, and ASTRAL scores reliably predicts 90-day mRS of 2 or greater in patients with acute ischemic stroke.

Is there a place for polygenic risk scores and gwas in modern genetic testing?

Traditionally genetic testing has focused on rare high penetrance mutations such as BRCA1 mutations conferring risk of breast/ovarian cancer or MYOC mutations conferring risk of glaucoma. However, most Australians affected by common diseases such as these do not carry these mutations. Instead, most people are at high genetic risk due to the cumulative effect of many polygenes (that is, variants of individually small effect on risk). Recent genome-wide association studies have identified polygenes for virtually every complex human trait. There has been considerable recent excitement about polygenic risk scores (PRS) – these provide an index of genetic risk based on the cumulative effect of many polygenes. While there are now PRS for many diseases, their translational utility can be limited due to poor predictive performance or because of a lack of real world consequences. In my talk I will discuss applications of PRS across a range of traits including eye disease and cancer. In a general population setting I will show there are exciting opportunities for PRS to be used in improving risk stratification and screening. In the ‘high penetrance mutation’ setting, I will show that mutation penetrance can vary dramatically with PRS values, with important consequences for genetic testing.

37. Understanding the clinical utility and prognostic implications of the new AJCC8 staging system for cutaneous squamous cell carcinoma: A survey of head and neck cancer specialists

Background: In order to have clinical relevance and assist in treatment decisions, a cancer staging system must stratify patients into prognostically meaningful groups; outcomes should be homogenous within groups, but distinct between them (AJCC8, 2017). The staging system for head and neck cutaneous squamous cell carcinoma (HNcSCC) in the AJCC 8, has been criticised for its lack of homogeneity and distinctiveness, with no difference in survival between pT2 and pT3 tumours, nor between stage III and stage IV disease. The nodal staging system is likewise a poor predictor of patient outcomes. In addition, the AJCC8 staging system for cutaneous SCC is similar to that of mucosal SCC, despite the vastly different biological behaviour of these diseases. Aim: To understand the experience of head and neck cancer surgeons, pathologists, radiation and medical oncologists in using the AJCC8 in routine management of HNcSCC patients. Method: A literature review was undertaken to: (1) determine the prognostic utility of HNcSCC AJCC8 staging system, (2) identify tumour factors which may influence the staging system, and (3) identify avenues to improve the staging system. An eighteen-part online questionnaire was administered to surgeons, surgical pathologists, oncologists, and radiation oncologists, designed to elicit their views on the prognostic utility of this staging system. Results and discussion: Literature review using the terms ‘head and neck’, ‘cutaneous SCC’, ‘prognosis’ and ‘AJCC’ yielded over 200 results pertaining to the prognostic features of HNcSCC, with fewer than 20 peer-reviewed, indexed manuscripts directly relevant to the prognostic performance of the AJCC8 HNcSCC staging system. The evidence suggests that the omission of prognostically important characteristics (such as patient immunosuppression, depth of invasion, perineural invasion, extent of extranodal extension, and soft-tissue metastases) may contribute to the poor predictive performance of the staging system. The questionnaire is being currently circulated and the Results will be provided.

Kernel-Based Aggregating Learning System for Online Portfolio Optimization

Recently, various machine learning techniques have been applied to solve online portfolio optimization (OLPO) problems. These approaches typically explore aggressive strategies to gain excess returns due to the existence of irrational phenomena in financial markets. However, existing aggressive OLPO strategies rarely consider the downside risk and lack effective trend representation, which leads to poor prediction performance and large investment losses in certain market environments. Besides, prediction with a single model is often unstable and sensitive to the noises and outliers, and the subsequent selection of optimal parameters also become obstacles to accurate estimation. To overcome these drawbacks, this paper proposes a novel kernel-based aggregating learning (KAL) system for OLPO. It includes a two-step price prediction scheme to improve the accuracy and robustness of the estimation. Specifically, a component price estimator is built by exploiting additional indicator information and the nonstationary nature of financial time series, and then an aggregating learning method is presented to combine multiple component estimators following different principles. Next, this paper conducts an enhanced tracking system by introducing a kernel-based increasing factor to maximize the future wealth of next period. At last, an online learning algorithm is designed to solve the system objective, which is suitable for large-scale and time-limited situations. Experimental results on several benchmark datasets from diverse real markets show that KAL outperforms other state-of-the-art systems in cumulative wealth and some risk-adjusted metrics. Meanwhile, it can withstand certain transaction costs.

Takeovers and the Dividend Discount Model

The standard dividend discount model assumes an infinite stream of dividends, but many stocks disappear through merger at a premium at some point in their corporate life, with a current takeover probability of about 0.5% to 2% per year for publicly-traded firms in the U.S. Ignoring takeover probabilities (1) predicts stock prices that are too low; and (2) underestimates the implied cost of capital. This may help explain the poor predictive performance of implied cost of capital methods and the low returns found for most stocks over their lives, since firms that are not acquired are unlikely to earn their true cost of equity.

Properties of analysts’ consensus cash flow forecasts for Australian firms

PurposeThe purpose of this paper is to determine the properties of analysts’ cash flows from operations (CFO) forecast generated for Australian listed firms as a productive activity, within the wider processes of financial disclosure in Australia.Design/methodology/approachTwo categories of criteria are adopted: first, basic predictive statistical performance relative to a benchmark model and earnings forecasts; and second, relevance for equity pricing, as indicated by the market reaction to cash flow or forecast error reactions. The final sample comprised 2,138 observations between 2001 and 2016 and several regression models are estimated to determine the relative performance and market reaction.FindingsAnalysts’ consensus cash flow forecasts demonstrate poor predictive performance relative to earnings forecasts. Cash flow forecasts are typically naïve extensions of earnings forecasts. Furthermore, cash flow forecasts appear to be of minimal use for equity market participants in complementing earnings forecasts’ role in informing firms’ equity pricing.Practical implicationsWhile analysts’ earnings forecasts are useful for making predictions, the role of analysts’ cash flow forecasts in capital market functional efficiency appears quite limited.Originality/valueThis study is one of few that examines comparative usefulness of analysts’ earnings and cash flow forecasts and their predictive power using the Australian setting. Additionally, it enriches the sparse international literature on such forecasts.

Predictive performance of population pharmacokinetic software on vancomycin steady-state trough concentration

To estimate the predictive performance of the population pharmacokinetics software JPKD-vancomycin on predicting the vancomycin steady-state trough concentration, and to analyze the related factors affecting the predictive performance. The clinical data of patients who were treated with vancomycin and received therapeutic drug monitoring (TDM) admitted to Suzhou Hospital Affiliated to Nanjing Medical University from July 2013 to December 2018 were enrolled. All patients were designed an empirical vancomycin regimen (initial regimen) according to vancomycin medication guidelines. Steady-state trough concentrations of vancomycin were determined at 48 hours after the first dose and 0.5 hour before the next dose. Dosage regimen was adjusted when steady-state trough concentration was not in 10-20 mg/L (adjustment regimen), and then the steady-state trough concentration was determined again 48 hours after adjustment. First, the JPKD-vancomycin software was used to calculate the initial regimen and predict the steady-state trough concentration according to the results calculated by classic pharmacokinetic software Vancomycin Calculator. Second, the JPKD-vancomycin software was used to adjust the vancomycin dosage regime and predict the steady-state trough concentration of adjustment regimen. The weight residual (WRES) between the predicted steady-state trough concentration (Cpre) and the measured steady-state trough concentration (Creal) was used to evaluate the ability of the JPKD-vancomycin software for predicting the vancomycin steady-state trough concentration. The TDM results of initial regimen were divided into accurate prediction group (WRES < 30%) and the inaccurate prediction group (WRES ≥ 30%) according to the WRES value. Patient and disease characteristics including gender, age, weight, height, the length of hospital stay, comorbidities, vasoactive agent, mechanical ventilation, smoking history, postoperative, obstetric patients, trauma, laboratory indicators, vancomycin therapy and TDM results were collected from electronic medical records. Univariate and multivariate Logistic regression analysis was used to screen the related factors that influence the predictive performance of JPKD-vancomycin software, and the receiver operating characteristic (ROC) curve was drawn to evaluate its predictive value. A total of 310 patients were enrolled, and 467 steady-state trough concentrations of vancomycin were collected, including 310 concentrations of initial regimen and 157 concentrations of adjustment regimen. Compared with the initial regimen, the WRES of adjusted regimen was significantly reduced [14.84 (6.05,22.89)% vs. 20.41 (11.06,45.76)%, P < 0.01], and the proportion of WRES < 30% increased significantly [82.80% (130/157) vs. 63.87% (198/310), P < 0.01]. These results indicated that JPKD-vancomycin software had a better accuracy prediction for steady-state trough concentration of the adjusted regimen than the initial regimen. There were 198 concentrations in the accurate prediction group and 112 in the inaccurate prediction group. Univariate Logistic regression analysis showed that women [odds ratio (OR) = 0.466, 95% confidence interval (95%CI) was 0.290-0.746, P = 0.002], low body weight (OR = 0.974, 95%CI was 0.953-0.996, P = 0.022), short height (OR = 0.963, 95%CI was 0.935-0.992, P = 0.014), low vancomycin clearance (CLVan; OR < 0.001, 95%CI was 0.000-0.231, P = 0.023) and postoperative patients (OR = 1.695, 95%CI was 1.063-2.702, P = 0.027) were related factors affecting the predictive performance of JPKD-vancomycin software. Multivariate Logistic regression analysis indicated that women (OR = 0.449, 95%CI was 0.205-0.986, P = 0.046), low CLVan (OR < 0.001, 95%CI was 0.000-0.081, P = 0.015) and postoperative patients (OR = 2.493, 95%CI was 1.455-4.272, P = 0.001) were independent risk factors for inaccurate prediction of JPKD-vancomycin software. The ROC analysis indicated that the area under ROC curve (AUC) of the CLVan for evaluating the accuracy of JPKD-vancomycin software in predicting vancomycin steady-state trough concentration was 0.571, the sensitivity was 56.3%, and the specificity was 57.1%. The predictive performance of JPKD-vancomycin software was decreased when CLVan was lower than 0.065 L×h-1×kg-1. JPKD-vancomycin software had a better predictive performance for the vancomycin steady-state trough concentrations of adjustment regimen than initial regimen. JPKD-vancomycin software had a poor predictive performance when the patient was female, having low CLVan, and was postoperative. The predictive performance of JPKD-vancomycin software was decreased when CLVan was lower than 0.065 L×h-1×kg-1.

Diagnostics-Driven Nonstationary Emulators Using Kernel Mixtures

Weakly stationary Gaussian processes (GPs) are the principal tools in the statistical approaches to the design and analysis of computer experiments (or uncertainty quantification). Such processes are fitted to computer model output using a set of training runs to learn the parameters of the process covariance kernel. The stationarity assumption is often adequate, yet can lead to poor predictive performance when the model response exhibits nonstationarity, for example, if its smoothness varies across the input space. In this paper, we introduce a diagnostic-led approach to fitting nonstationary GP emulators by specifying finite mixtures of region-specific covariance kernels. Our method first fits a stationary GP and, if traditional diagnostics exhibit nonstationarity, those diagnostics are used to fit appropriate mixing functions for a covariance kernel mixture designed to capture the nonstationarity, ensuring an emulator that is continuous in parameter space and readily interpretable. We compare our approach to the principal nonstationary GP models in the literature and illustrate its performance on a number of idealized test cases and in an application to modeling the cloud parameterization of the French climate model.

Streaming parallel variational Bayesian supervised factor analysis for adaptive soft sensor modeling with big process data

Time-varying and state shifting are two of the main process factors that cause poor prediction performance of soft sensors. Adaptive soft sensor is commonly an alternative practice to ensure high predictive accuracy. However, the large scale of process data often leads to inefficiency of model updating. In this paper, a streaming variational Bayesian supervised factor analysis (S-VBSFA) model is first proposed to capture the process time-varying and state shifting features through online updating of the posterior of model parameters. During the updating process, the symmetric Kullback–Leibler (SKL) divergence is utilized to determine priors of the next variation Bayesian inference. To improve the modeling efficiency for large-scale process data, the parallel computing strategy is further applied to the streaming model. As a result, the proposed streaming parallel VBSFA (SP-VBSFA) algorithm not only relieves the computing pressure of modeling big process data, but also improves the prediction accuracy and further reduces the tracking time delay for process variations. Two case studies demonstrate the superiority of the proposed method, compared to conventional methods.

Time series prediction based on deep confidence network

The timing deep belief network model (T-DBN) is proposed to solve the problems of gradient dispersion, local minimum, poor long-term prediction performance of nonlinear time series prediction and high complexity of high-dimensional sequence data in traditional computer neural networks. In the pre-training stage, the improved greedy pre-training algorithm is adopted. In the pre-training process, the gradient fixing parallel tempering (GFPT) algorithm is used to determine the depth of network by reconstruction error. In the stage of reverse adjustment, the quasi-newton method (BFGS) is adopted to obtain more accurate prediction performance. Combined with the theory of phase space reconstruction and BP (back propagation) neural network, the total power of agricultural machinery in Jiangxi province of China from 2016 to 2020 is predicted. For the highly nonlinear stock data, the characteristics of the Shanghai stock exchange index from 1990-12-20 to 2018-03-30 are extracted from the stock data of Tonghuashun software. T-DBN, DBN and long short-term memory (LSTM) models areused for stock forecasting with the prediction accuracy of 79.3%, 77.9% and 74.6%, respectively. The experimental results demonstrate the better prediction performance of T-DBN model in comparison with DBN and LSTM models.

An efficient wavelength selection method based on the maximal information coefficient for multivariate spectral calibration

Spectral data on the modern spectroscopic instrument are commonly of high co-linearity and contain a large number of spectral variables, which may cause the poor predictive performance of the developed model. To address this problem, a novel method for wavelength selection, named maximal information coefficient screening combined with PLS regression (MICPLS), is proposed. MIC can capture a wide range of relationships between feature variables and target variable, including both functional and non-functional relationships. By employing the simple but effective MIC screening, MICPLS can rapidly and efficiently select strongly correlated spectral variables related to the response. The performance of MICPLS is investigated with two real spectroscopic datasets. Compared with commonly used methods for variable selection based on informative criterions including variable importance in projection (VIP) and selectivity ratio (SR), the results show the proposed method can improve the prediction performance, which is a good and promising strategy for variable selection in spectral analysis.

Soft-Sensor Development for Processes With Multiple Operating Modes Based on Semisupervised Gaussian Mixture Regression

Gaussian mixture regression (GMR) is an effective tool in developing soft sensors for online estimating difficult-to-measure variables in industrial processes with multiple operating modes. However, the GMR usually requires a sufficient amount of labeled samples to guarantee accurate probability density function (PDF) estimations because of its supervised learning process. Unfortunately, in soft-sensor applications, labeled samples could be very infrequent due to technical or economic limitations, which may lead the GMR-based soft sensors to unreliable parameter estimation and model selection, resulting in poor prediction performance. To tackle this problem, a semisupervised GMR (S2GMR) was proposed, where both labeled and unlabeled samples were effective. In the S2GMR, the PDFs of Gaussian components in input space and the functional dependence between input and output variables were learned simultaneously based on the expectation–maximization algorithm. Moreover, the Bayesian information criterion was employed to automatically determine the number of Gaussians for the S2GMR. The S2GMR was first investigated by a numerical example, and then applied to a real-life ammonia synthesis process for estimating the oxygen concentration at the top of the primary reformer. The two case studies verified the effectiveness of the proposed method.

Learning to Identify Severe Maternal Morbidity from Electronic Health Records.

Severe maternal morbidity (SMM) is broadly defined as significant complications in pregnancy that have an adverse effect on women’s health. Identifying women who experience SMM and reviewing their obstetric care can assist healthcare organizations in recognizing risk factors and best practices for management. Various definitions of SMM have been posited, but there is no consensus. Existing definitions are further limited in that they 1) are often rooted in existing clinical knowledge (which is problematic as many risk factors remain unknown), leading to poor positive predictive performance (PPV), and 2) have limited scalability as they often require substantial chart review. Thus, in this paper, a machine learning framework was introduced to automatically identify SMM and relevant risk factors from electronic health records (EHRs). We evaluated this framework with EHR data from 45,858 deliveries at a large academic medical center. The framework outperformed a state-of-the-art model from the U.S. Centers for Disease Control and Prevention (AUC of 0.94 vs. 0.80). Specially, it improved upon PPV by 59% (CDC: 0.22 vs. our model: 0.35). In the process, we revealed several novel SMM indicators, including disorders of fluid or electrolytes, systemic inflammatory response syndrome, and acidosis.

High-temperature oxidation of methyl isopropyl ketone: A shock tube experiment and a kinetic model

Pentanone has three types of isomers: 3-methyl-2-butanone (methyl isopropyl ketone, MIPK), 3-pentanone (diethyl ketone, DEK) and 2-pentanone (methyl propyl ketone, MPK). Since MIPK is the smallest branched ketone and a promising biofuel, it is important to fully understand its combustion kinetics. In this work, the ignition delay times were measured in the temperature range from 1150 to 1600 K at pressures from 1 to 5 bar and for the equivalence ratios of 0.5, 1.0 and 1.5 (Mixes 1–9). A high-temperature detailed combustion kinetic model for MIPK was developed via a hierarchical method. Previous theoretical computations of the H-atom abstraction by HO˙2 and O˙H from the tertiary carbon of MIPK were adopted, and the reaction rate coefficients of H-abstraction by H˙ from the MIPK tertiary carbon were calculated via quantum chemical techniques. The rate coefficients of the reactions that are related to the acetyl group and the isopropyl group of MIPK were estimated by analogy with MPK and DIPK, respectively. Eleven elementary reactions in the submodel that are related to MIPK were selected for the optimization due to the poor prediction performance of the raw model. The self-adaptive differential evolution algorithm was applied to optimize the raw kinetic model based on the ignition delay times of Mixes 1–6. The optimized MIPK model was validated against the ignition delay times of Mixes 7–9. A brute-force sensitivity analysis, in combination with a flux analysis, demonstrates that the amount of unsaturated fuel-related stable intermediate products is the key factor that determines the reactivity of MIPK and DEK.

Volatility forecasts using stochastic volatility models with nonlinear leverage effects

AbstractThe leverage effect—the correlation between an asset's return and its volatility—has played a key role in forecasting and understanding volatility and risk. While it is a long standing consensus that leverage effects exist and improve forecasts, empirical evidence puzzlingly does not show that this effect exists for many individual stocks, mischaracterizing risk, and therefore leading to poor predictive performance. We examine this puzzle, with the goal to improve density forecasts, by relaxing the assumption of linearity of the leverage effect. Nonlinear generalizations of the leverage effect are proposed within the Bayesian stochastic volatility framework in order to capture flexible leverage structures. Efficient Bayesian sequential computation is developed and implemented to estimate this effect in a practical, on‐line manner. Examining 615 stocks that comprise the S&amp;P500 and Nikkei 225, we find that our proposed nonlinear leverage effect model improves predictive performances for 89% of all stocks compared to the conventional stochastic volatility model.

Generating a Spatiotemporal Dynamic Map for Traffic Analysis Using Macroscopic Fundamental Diagram

Transportation simulation and analysis projects that utilize maps with inappropriate fidelity levels carry a significant risk of having poor runtime or poor prediction performance. To address this, researchers use map abstraction method to abstract out a simplified map with fewer links and nodes based on the original full detailed map. Traditional static abstraction methods produce analysis maps with a single fidelity across the entire planning horizon, which cannot reflect the dynamic changes of daily traffic. This paper proposes a spatiotemporal dynamic map abstraction approach that adopts a time series clustering method to segment the analysis time horizon adaptively based on a Macroscopic Fundamental Diagram (MFD) curve, which describes network-wide dynamic traffic states. Time periods with similar macro-performance are grouped into one subinterval. A map with a dedicated fidelity is produced for each subinterval. Furthermore, a simulation is run on multiple abstracted maps with different fidelities in a sequence according to their temporal order. A numerical experiment ascertains that the proposed approach has promising results in both analysis accuracy and efficiency for resource-constrained modeling agents.

Neonatal hyperamonemia as a consequence of a defect of the urea cycle

Propionic acidemia (PA) is a severe metabolic disorder characterized by multiorgan pathology, including renal disease. The prevalence of chronic kidney disease (CKD) in PA patients and factors associated with CKD in PA are not known.Thirty-one subjects diagnosed with PA underwent laboratory and clinical evaluations through a dedicated natural history study at the National Institutes of Health (ClinicalTrials.gov identifier: NCT02890342).Cross-sectional analysis of the creatinine-based estimated glomerular filtration rate (eGFR) in subjects with native kidneys revealed an age-dependent decline in renal function (P < 0.002). Among adults with PA, 4/8 (50%) had eGFR <60 mL/min/1.73 m2. There was a significant discrepancy between eGFRs calculated using estimating equations based on serum creatinine compared with serum cystatin C (P < 0.0001). The tubular injury marker, plasma lipocalin-2, and plasma uric acid were strongly associated with CKD (P < 0.0001). The measured 24-hour creatinine excretion was below normal, even after adjusting for age, height, and sex.CKD is common in adults with PA and is associated with age. The poor predictive performance of standard eGFR estimating equations, likely due to reduced creatine synthesis in kidney and liver, could delay the recognition of CKD and management of ensuing complications in this population.